Skip to main content

Full text of "Rejoinder to Professor Edgerton"

See other formats


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 

Read more about Early Journal Content at 
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 

Rejoinder to Professor Edgerton 93 


Edwin W. Fay 
University of Texas 

1. To make an Irish reply to Professor Edgerton 's hypothet- 
ical question (p. 87), what I wish to know is whether the author 
or editor of the Vikramaearita and the late users of the verb 
lati employed Sanskrit as a vernacular and mother-tongue, 
whether they thought in Sanskrit (I do not mean to the exclu- 
sion of a Prakritic or Hindi dialect). If these authors had 
received Sanskrit viva voce vivisque auribus it is entirely possi- 
ble that they introduced into Sanskrit literature words not 
written into our record but, in point of origin, of hoary antiq- 
uity. Grammatical citations apart, parut ( : -n-ipvai) is not of 
record. This shows the possibility of a most ancient word never 
being included in the literary record (supposing us to have it 
all!), and leaves us to infer that Panini took the example from 
the speech of his own time. The IE. character of parut would 
have guaranteed its authenticity even if, without Panini 's cita- 
tion, it had emerged as late as lati. Again, the history of the 
root stigh, long known only through the questionable medium of 
Dhatupatha, shows us how a word of most certain IE. origin 
was restricted, not (so far as I know) to a definitely ascertain- 
able locality, but to the canticles of a restricted Vedic sect. The 
relation of literary Sanskrit to the genuine vernaculars is a 
thorny problem. From the time of the great Epics on, Sanskrit 
was not, in the narrow sense, a vernacular. But the language 
was imparted viva voce and received vivis auribus, so that it 
actually functioned as a standardized class or caste dialect, and 
its speakers were bilingual. In a genuine, if restricted, sense, 
this dialect must have begun as speech, so that the question 
arises at what time, in which century (sorites- wise) from 200 
B. c. (shall I say?) down to 1500 a. d., the colloquial founts 
dried up. For lati and adesa there is also the other question of 
a possible bookish source (see § 9, note) . If a word of good IE. 
stamp appeared first in the learned Epic of Apollonius or in 
Callimachos I should not question its genuineness as Greek even 

* Revised by the author after reading Edgerton 's following ' Counter- 
Rejoinder. ' 

94 Edwin W. Fay 

though the vernacular of these authors was Hellenistic. I can- 
not think the lateness of lati substantially different from the 
lateness of sthagayati (covers) : Lat. tegit; or of hadati which, 
exception made of Epic -hdda, is classical only, but surely of IE. 
provenance. Also note itar, primary derivative of i, but not 
found till Yasavadatta, see Gray's edition, pp. 202, 214. 

The vocalism of lati. 

2. I could not think, because of the conflict of vowels in Sk. 
lati and Hindi le-na, that the lexicon of Monier "Williams meant 
to assert the express derivation of the one from the other; nor 
did I feel sure — though I am compelled to speak without due 
lexical aids — that the contracted Hindi form lana<. le-ana was 
earlier than the emergence of lati. [And now exactly so for the 
Bengali root la.] On the other hand, the morphological rela- 
tion between lati and labhati has so many analogues to confirm 
it in IE. grammar that a theory of late emergence, but early 
origin, for lati is not to be put out of court till something like 
philological proof of origin from an Indie vernacular is assured. 
In brief, a colloquial option between lati and labhati may always 
have existed in that Primary Prakrit from which Sanskrit came, 
without one of the terms having emerged till a late period. 
Even what one takes for the commonest words may emerge rela- 
tively late into the written record, for instance Eng. leg die bull 
(see Eoyster in Studies in Philology, 14. 235 ) . 

[2a. In my original critique I failed to mention — because I 
did not know it then — that Wackernagel (Ai. Oram. § 80) had 
tentatively proposed the correlation of lati (root lau) with Lat. 
lucrum (gain, takings). The very dialect forms cited by Pro- 
fessor Edgerton, however, make for the root lei — perhaps from 
(t)Ui, cf. my explanation of Lat. cle-mens: raXai-^ptav as toyed 
with by "Walde on p. 868 and then on p. xx. There is an 
undoubted Prakrit root le and, whatever Pischel may have 
thought when he was translating Hemacandra, he categorically 
correlates the absolutives lev i leppinu levinu with Sk. la in his 
Prakrit Grammar § 588. Then Pk. le is from hi (: lei : : Av. 
pai : pai, see Bartholomae's Grammar, § 122. 10). We actually 
have Pk. lenti in the Karpura-manjarl 1. 13, as follows : 

lenti na taha angammi (loe. sg.) kuppasaam 
and do not put on a bodice (Lanman). 

Rejoinder to Professor Edgerton 95 

After Plautus Amphitruo 999, capiam coronam mi in caput, I 
feel free to render our sentence by 
capiunt non turn (for neque, postponed) <sibi> in membra uesticulam. 

How a proper sense for lenti here — and I have gone over the 
usage of li carefully in the Petersburg lexica — can be arrived at 
from Sk. ll (cling) I cannot divine. — In Sanskrit the flexion of 
the root ~kHi (to lie) — so Brugmann correctly writes it in IF 6. 
98; cf. Bartholomae, Lex. 1571 — generalizes the midgrade k l n 
(sete, accent abnormal) . In Greek /cen-ai Wei is generalized. In 
Sk. Id [i] ti : Pk. lenti we have the alternation e [i] /vi. That 1% 
would be a legitimate form of ld[i] in Sanskrit is true enough, 
and we might in fact derive Pk. levi from Hitvl, cf. Sk. pitvl : 
pdti (root poi). An Indie root Idi | Mi is recognized by Franke, 
BB 23. 177, in Pali layati (harvests). Now this is the root of 
lati. For the sense of reaps (i. e. harvests, gathers) from takes 
(seizes) cf. Cicero, Sen. 70, tempora demetendis fructibus et 
percipiendis, with Cato's more generalized usage {Agr. 4. 1) in 
the turn fructi plus capies. Further note Skt. V grabh ( : Eng. 
grabs), cognate with Germ. Oarbe (sheaf of' the reapers).] 

i. Whether 1 ddesa (indicium) came to mean salutation. 
3. If a sage could utter a benediction to a Hindu king in 
response to a merely mental salutation (an assumed glum 
silence, one suspects, to intensify the test of the sage's presci- 
ence) our sage might well have acknowledged the same king's 
intimation (cf. Lat. indicat) or signal (to proceed, of attention ; 
look of recognition), 1 and that quite duly. When a king of 
England 'commands' a singer or other artist, what remains 
formally a command is in fact a great courtesy, with all the 
effect of a salutation. Note that in Latin, by way of ellipsis, 
but ellipsis is one of the standing elements in semantic develop- 
ment, iubeo (sc. saluere) means saluto. — I still think that one 
who said distyd (salue; lit. with homage) might have turned for 
its cases to ddesa, a flexional word in being. In Iranian the cor- 
respondent of ddesa is Av. ddisti, whence the semantic propor- 
tion Indo-Iran. d-disti (indicium) : Sk. d-desa : : distyd, (with 

1 The closest synonym of Me&a is ajna, which means not only command 
but also, as I here assume for adeSa, permission. 

96 Edwin W. Fay 

homage): (2) adesa (if = salutation) . In Latin, salus (greet- 
ing) was adopted as the flexional form of the word of greeting, 
impv. salue (be whole). What I have in mind is a semantic 
correlation such as we employ when we use appurtenance as the 
noun corresponding to the technical adjective phrase pertaining 
to, in the formulae of derivation and definition. The correla- 
tion appurtenance x pertaining to is desk English, not the ver- 
nacular. Cognate words do interchange their meanings as when, 
to employ a standard example, to execute a man is developed out 
of the execution of a sentence. It is perfectly legitimate to sup- 
pose that from dirty a (salue) dis + a, or derivatives thereof, 
might have gathered up the force of salutem dico (saluto) ; it 
is quite legitimate, as a question of genesis, to say that a-desa 
does not derive from a -+- 1 dis, but rather from a + das (do 
homage, acclaim), in alternation with a -\-dis. For another 
example of the gradation a : i in interior position — at root ends 
nothing is commoner — cf. Jehad : khid, with intermediate e in 
khedd (not secondary, pace Wackernagel Ai. Gram. § 15), Av. 
sas : sis, see Bartholomae's Grammar §"122. 8. 

ii. The etymology of 2 adesa ( 1 salutation) . 
4. If in a formula of politeness such as adesam dattva — for- 
mulae may be very old — adesa meant salutation, it may well have 
come by its meaning through honest descent. The equation 
of SeUwrai (greets) with dasnoti (does homage) has not been 
responsibly questioned for 40 years (see literature in Brug- 
mann-Thumb, Gr. Gram. § 342), nor do I understand Professor 
Edgerton now to question it ; and we are now devising, to satisfy 
our craving for system, a fit gradation diagram with a place for 
the root de(i)ft\ a place for its derivative a-desa (of IE. type) ; 
with a place for Lat. dignus* a place for dicat (consecrates), 
and a place for decus. — On the late development of 2 adesa from 
adis see § 9 fn. 

2 Be it said in passing that dignus has certainly for its nearest of kin 
(morphologically and semantically, I mean) ONorse tiginn (eminent 
•Cdigito monstratus, see the lexicon of Falk-Torp, p. 1251). I call par- 
ticular attention to the tlmbrian perfect stem pur-dins' (see AJP 32. 414), 
with the sacral sense of offered. Here we have a nasal variety of the root 
of dicat; cf. Sk. puro-daiam (ace), offering. 

Rejoinder to Professor Edgerton 97 

iii. Hindi ddes: ddesa (ddesam dattva). 

5. I assume that ddes came from ddesa (indicium) and that, 
excluding the temporary expedient of 2 ddesa, its alleged sense 
of salutation, so far as we may list a contextual shading for a 
definition, was at some time and place developed by way of con- 
notation (a polite signal to proceed is a salutation) or by way of 
ellipsis. A situation apt for the development of the connotation 
lies in fact before us, where tasya ddesam dattvd etc. = ei inti- 
matione < ? sui> facta (rex ipse a sapient e salutatus est). Or, 
if we inform ourselves that Lat. indicium means not only testi- 
mony but also leave to testify, we may grant that, by a like shift 
of usage, ddesa might mean, not only announcement, but leave 
to announce' (? himself, the sage): ei indicatione <.ipsius~> 

iv. The meaning of a -\- dis. 

6. In support of my substantially correct version of RV 6. 
56. 1 (p. 83) I go on to demonstrate that this verb means pretty 
nearly what Lat. inclamare means, both in its good sense of 
invoke and in the bad sense of jeer at, abuse. Why should one 
who recalls Lat. f acinus or valetudo or inclamare or acclamatio 
object to the exhibition by a word of both bad and good senses ? 
As a vox media Eng. challenge is a good rendering of a + dis; 
or Lat. provocare (but with all the range between salutare and 
lacessere, or even imprecari) . In 9. 70. 5, ddedisdnah saryaheva 
surudhah = inclamans ut Sagittarius 3 iaculatores (suru- : Sabine 
Lat. curis, spear), and in 10. 61. 3, dsrinita ddisam = paravit 
(lit. coxit, cf. coquere iras, verba) inclamationem (impreca- 
tionem). One thinks of the 'brag' of Homeric combatants 
before beginning to fight. The reader may easily go through 
the ensuing examples from Professor Edgerton 's list and sub- 
stitute due forms of inclamo or of challenge. 

7. In the three next passages also ddis has the nominal sense 
of inclamatio, but varying, like acclamatio, between cheers (laus, 
honor) and jeers (inrisio, minae). The passages are as follows: 
(1) 8. 60. 12 b , tdranto aryd ddisah = superantes hostis inclama- 
tiones (minas) . For the situation cf. again the brag and threats 
of any pair of Homeric warriors, e. g. Tlepolemos and Sarpe- 

"The archer and spearmen, typically taken, may have belonged either 
to hostile armies or, as rival arms of the service, to the same army. 
7 JAOS 40 

98 Edwin W. Fay 

don in E 633 sq. (2) In 6. 4 Agni is besought to fetch the other 
gods to the sacrifice (st. 1), and in st. 5 (text of Aufrecht) we 
read, turydma yds ta adisdm drdtir = super emus <.eum> qui 
tibiinvocationum (laudum) invidus <est> (cf. 9. 21. 5, below). 
(3) I render 8. 93. 11 as follows: 

ydsya te nu cid adi&am nd mlndnte svarajyam \ nd devo nddhrigur jdnah 
cuius illi quidem laudem non impediunt eius <ve> imperium 

neque deus <alius, see § 9> neque semperfestinans(?) gens. 

8. In 9. 21. 5 (and likewise for the next stanza), dsmin . . 
dddhdta vendm adise etc. = apud nos facite voluntatem incla- 
mare (eum qui nobis invidus est), i. e. confirm in (or unto) us 
our desire, viz. to rebuke him who is stingy toward us. 

9. — 6. 56. 1. To give a hostile sense to adise here involves 
taking karambhdd (Pultiphagus), the title of Pusan, as defama- 
tory. This seems to me a grave literary error in the interpre- 
tation maintained by Both and Grassmann. Inasmuch as 
karambhd was the special food of Pusan it would be strange to 
summon his worshippers in the first stanza of a hymn by 
recounting a jeer of the 'pagans' "(in this case 'cits') that 
honored him not. Professor Edgerton will have it that the first 
stanza of a Pusan hymn says 'whosoever shall aim at Pusan 
(our god) with the taunt of "Porridge-eater," the god is not his 
to aim at.' To me the stanza can only mean what Sayana 
thought it meant — and he rendered ddidesati by abhitfauti 
(praises) — 'Whosoever shall invoke (praise) Pusan (our god) 
by his favorite title need invoke no other god.' 4 As for kar- 
ambhd, it was mixed-with-the-food (karambhin) of Indra, but 
besides (shade of Dr. Samuel Johnson!) it was also shared [and 
not only in 'porridge-punch'] by Indra — unless we mean to dis- 
qualify the evidence of Ait. Br. 2. 24 — and Indra was no weak- 

4 Among the Vedic clerks and priors, the scholars and men of letters, 
before and after his time (say 1350 a. d.), Sayana would not have been 
alone in holding and teaching the equation ddideSati = abhistauti (laudat, 
eelebrat). I confess I am casual enough to believe, even in the face of 
Professor Edgerton 's ordered genealogical and chronological criteria, that 
among these scholars many, one or another, even the redactor of the Vikra- 
macarita, seeking to vary the monotony of namas (solus, laudatio, honor), 
might have hit upon adeSam dattva (laudationem dans) as a fit substitute 
for namasTcrtya, so giving to adesa, a word in being, the sense of ddidesati. 

Rejoinder to Professor Edgerton 99 

ling, nor yet a hind. 5 The real vocative karambhdd (here 
turned to a nominative before iti) is a virtual invitation to 
Pusan to come and eat karambha; and the Vedic poet said in 
effect, to make a slight change in my previous version, 

qui hune inclamat (invocat) Pultiphagum nomine Pusanam, 
non ab eo deus invocando <est>. 

This version leaves the ambiguity of the original. If, to begin 
with the less probable, deus = Pusan, the apodosis means that 
Pusan will not wait for a second invitation, but accept instanter 
the call to his favorite food. If deus is not Pusan the apodosis 
means : not a god is to be invoked by the worshipper, for Pusan 
alone is sufficient. In my first version I supplied, after Ludwig, 
alius; but neither Ludwig (I will suppose) nor I actually sup- 
plied any as to the original (see also for na <.anyo> devo 8. 93. 
11 in § 7). "We have here a partitive relation, and Pusan is 
tacitly excluded from the other gods. [In passing I will state 
that I think Ludwig was entirely right in interpreting priydd 
. . preyo in 1. 140. 11 by dearer than <any other, or the typ- 
ical dear.] One thinks of Corinthians 15. 27 : But when he 
saith, All things are put in subjection, it is evident that he is 
excepted who did subject all things unto him. Cf. on veiWos 
aWw Class. Rev. 8. 456, and the colloquialism, He runs faster 
than anybody (for anybody else) ; or, none such = no other like. 
On the other hand, there have been grammatical sticklers who, 
in respect to Milton's famous line, 'the fairest of her daughters, 
Eve,' objected to the inclusion of Eve; cf. Odyssey 5. 262, 
where Calypso includes herself with Ulysses (those two, and no 
others) in the words rots apa ^vOiav ripx f - — The omission of 'other' 
is common enough, though lists of examples lack. Note, with 
consideration of the context (Sw/mra in 1. 299 = 8o>o S in 1. 302), 
Odys. 6. 301, oi ph . . . Sti/wtra &air)K<av = no < other > residence 
of the Phaeacians. 

10. — 6. 48. 14. Omitting the unessential and accepting (with- 
out reserve as to the metre) Ludwig 's disposition of the adjec- 
tive complement of Visnu, I would thus render : 

•I am not unaware that Pusan was a Pan among the gods. To Pro- 
fessor W. Schulze he is Pan, and the sectarian character of Pusan, of 
which note is made below ($ 12), reminds us again of the difficulty of 
getting recognition for Pan throughout Greece. 

100 Edwin W. Fay 

tdm (se. Pusdnam, again!) . . . | srprdbhojasaih visnurh nd stusa adise 
eum ut Vishnum adipicibum <habentem> laudo invocando. 

But for adise (invocando) we must-supply a subject like us or 
you (the worshippers), which yields the meaning ut invocemus 
(invocetis) ; cf. 1. 52. 8, ddharayo divy a suryam drse = posuisti 
in caelo solem videndo i. e. ut videremus (ut homines viderent) . 
Also see excellent examples for subjectless infinitives in Monro's 
Homeric Grammar, § 231. It were possible, but harsher, to 
render adise by the imperative, invocate. Or stusa adise = I 
(re) commend to (be) invoke(d). 

11. The evidence for a -\- dis = inclamare has been submitted. 
The definition recognizes derivation from the root deik 1 . I 
doubt not that Professor Edgerton admits the propriety of try- 
ing, so far as may be, to utilize IE. derivation and etymology 
in the effort to fix the definition of Vedic words. To know the 
approximately original meaning of a word certainly helps in 
fixing the sense of its further ramifications, as in the case of 
distya (with homage) § 3. 

12. In conclusion I suggest that the two Pusan stanzas I have 
interpreted seem to constitute a sectarian recommendation of 
Pusan as the equal or superior of other gods. It is because of 
this sectarian quality that karambhdd cannot be a jeer (adis), 
but must be a word of praise (adis) , see § 9. 


Franklin Edgerton 
University of Pennsylvania 

Professor Fay (§3) seems to miss the point of the story of 
the 'mental salutation,' which appears to me to prove absolutely 
that, to the feeling of its author, no sage would bless a king with- 
out first receiving a salutation. There was no 'assumed glum 
silence' — except perhaps to an ignorant bystander who lackt the 
sage's omniscience; certainly the sage, if he had assumed a glum 
silence (that is, lack of salutation), would not hav blest the king. 
That is the whole point of the story. The silence was only tech-