On the Name Devanāgarī

Author(s): Walter H. Maurer

Source: Journal of the American Oriental Society , Jan. - Mar., 1976, Vol. 96, No. 1
(Jan. - Mar., 1976), pp. 101-104
Published by: American Oriental Society

Stable URL: http://www.jstor.com/stable/599893

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
https://about.jstor.org/terms

American Oriental Society is collaborating with JSTOR to digitize, preserve and extend access
to Journal of the American Oriental Society

This content downloaded from

149.233.192.93 on Tue, 21 Jul 2020 15:23:26 UTC
All use subject to https://about.jstor.org/terms
 MAURER: On the Name Devandgari 101

The
The phraseology
phraseologyof ofverse
verse3 3isiscurious:
curious:the
the
rsirsi
became
became
an an
by way of contrast permits no such such obscurity
obscurity toto shroud
shroud
anthill (valmika) and was not simply built into one. the background of its legendary author.
author.
Can Cyavana be the Valmika from whom Valmiki is So astonishing is this omission, which
which leaves
leaves the
the com-
com-
derived? It is an intriguing thought but it seems impos- mentators fumbling in clumsy efforts
efforts to
to provide
provide consistent
consistent
sible that both epics could then be ignorant of this etymologies for the sage's three epithets, that one is
relationship. forced to the conclusion that the information is almost

The similarity between the two episodes incorporating wholly unknown to the epic redactors. In the highly
this strange motif is so striking that Citrav gSstri wasBhrguisedled Mlahdbhdrata, it appears impossible to prove,
to conclude that it gave rise to a mixing up of the two and only in one instance even to suggest that Valmiki is
sages and that it is for this reason that Valmiki was given known as a Bhargava. In the Ramayana itself this
the designation Bhargava.34 This is a brave attempt to identification is made only twice, both times in the late
cut through the maze of the Valmiki-Bhargava issue. It Uttarakdanda. The Mahdbhdrata, the central core of the
probably contains a grain of truth in that it is far from Rdmdyana and even the late Bdlakdnda, otherwise a
unlikely that the late story of Valmiki's anthill was source of much information concerning Valmiki, appear
borrowed from the Cyavana episode in an effort to provide not to know of it. The matter of Pracetas and Pracetasa

a convincing etymology for at least one of the ddi kavi's will require further investigation, but again, in the epic
three names. But, as the epics do not themselves apply litterature only the Ultarakcdnda seems to know both that
the anthill motif to Valmiki, the argument as a whole is sage and that epithet in connection with Valmiki.
not convincing. On the basis of this evidence one may hazard several
interesting conclusions. For one it would appear that the
CONCLUSIONS Uttarakdnda is, at least in part, later than the Bdlakdn-
da but, unless we can find another source for Agvaghosa's
It appears little short of extraordinaryline, that earlier than the Buddhacarita. For another it may be
the Sans-
krit epics which revel in the accounts of the asserted that
deeds, if at all the process of Bhrguisation can be
origins
and antecedents of even the most trivial characters, said to have affected the Rdmayana, it has touched only
the two latest kdndas, the Bala and the Uttara. In the
should offer such scanty and muddled evidence concern-
ing a figure of the stature of Valmiki. The Mahabhdrata,former it has operated, in the same way as in the Ma-
hdbhdrata, although to a much less pronounced degree, b
the inclusion or insertion of a number of Bhargava
used again
again in
in the
the case
case of
of Cyavana
Cyavanawho
whoititisissaid
saidremained
remained
references and episodes. In the latter it would seem that
immobile
immobile underwater
underwater at
at the
the confluence
confluenceof
ofthe
theGafig
Gafig
the Bhrguisation has affected not so much the poem as
and Yamuna, became intimate with the fish and was the poet.
covered with water weeds, shells, etc., Mbh. XIII. 50.9-19. ROBERT P. GOLDMAN
34 Prdcina Caritra Kosa, s. v. Valmiki. UNIVERSITY OF CALIFORNIA, BERKELEY

On the
the Name
Name Devanmgari
Devanmgari

The
The paper
paper discusses
discussesthe
thename
name"Devanagari"
"Devanagari"as as
applied
applied
to to
thethe
Indian
Indian
script
script
and and
the the
pos-pos-
sible derivation of the word.

The
The name
name "Devanagari"
"Devanagari"has
haslong
longbeen
been something
somethingof of
a a
devanagara
devanagara "city
"city of
ofthe
thegods"
gods"and
andhence
henceinterpret
interpretit it
as as
puzzle
puzzle to
to scholars.1
scholars.1Some
Someregard
regard it it
as as
a derivative
a derivative
of of
" [script]
[script] of
of the
the city
cityof
ofthe
thegods"
gods"ororby
bysome
someequivalent
equivalent
phrase.2
phrase.2 Yet
Yet others,
others,and
andititisishard
hardtotosay
saywhich
whichform
form
the
the
1 That
That the
the name
nameisisobscure
obscureininmeaning
meaningand
and
origin
origin
hashas
been
been asserted
assertedin
invarious
variousSanskrit
Sanskritgrammars,
grammars,e.g.,
e.g.,
those of2 of
those E.g.,
E.g., A.
A. L.
L. Basham,
Basham,The
TheWonder
WonderThatThatWas
WasIndia
India
Whitney, Macdonell, Thumb-Hauschild; cf. also Hans (London, 1954), p. 397; M. R. Kale, A Higher Sanskrit
Jensen, Sign, Symbol, and Script: an Account of Man's Grammar, 7th ed. (Bombay, 1931), p. 1: ". . . the alphabet
Efforts to Write, 3rd ed., transl. from German by George
in which it is written is called Devanagari (sic 1), or that
Unwin (New York, 1969), p. 375, n. 1: "The name cannot employed in the cities of the gods"; A. Ballini, article in
be accounted for very clearly (' [script] of the divinethe Enciclopedia italiana, vol. XIX, p. 52f: "della citt/
city' ?)." degli dei"; A. M. Pizzagalli, Elementi di grammatica

This content downloaded from

149.233.192.93 on Tue, 21 Jul 2020 15:23:26 UTC
All use subject to https://about.jstor.org/terms
102 Journal of the American Oriental Society 96.1 (1976)

majority, view it as a compound, consisting of deva and by a European is apparently to be found in the letters of
ndgari, meaning either "city [-script] of the gods" or the Italian traveller Pietro della Valle, who in 1622, while
"city [-script] of the Brahmans" or "city [-script] of the in Persia on his way to India, became acquainted with an
kings," the word deva being capable of any of these Indian Brahman (Damodel Sami) who at Della Valle's
renderings.3 request wrote down his own name in characters which he
But "Devanagari" is also known more simply as "Na- learned from him were called "Nagher" (i.e., Nagari).6
gari," a name that is probably even more prevalent in
India than the longer name, which appears to be more Kupferplatten des Rastrakiutak6nigs Dantidurga von
popular among scholars outside of India. "Nagari," 754 p. Chr., aus Samangadh im siidlichen Marathenlande
as the name of a particular script, is fairly old, being(Kolapur)."
mentioned along with several other scripts by Al-Beruni,6 For a general discussion of the various passages
c. 1030 A.D., who refers to an alphabet called Nagarawhere Della Valle mentions the Nagari script, v. Theodor
Zachariae "Pietro della Valle fiber das Nagari-Alphabet"
(sic 1), which he says was current in the Malava country.4
in his Kleine Schrifften (Bonn, 1920), pp. 8-12, also in
This statement is particularly interesting, as the earliest
example of a text written entirely in Nagari characters, Wiener Zeitschrift fur die Kunde des Morgenlandes, XVI
an inscription of the Rastrakfita King Dantidurga,stems (1902), pp. 205-210.
from 754 A.D.5 The earliest mention of the word "Nagari" Another European, the French traveller Jean Chardin,
who went to Persia and India not long after Della Valle,
sanscrita (Milano, 1931), p. 4: ". . . scrittura della citthbut whose account was not published in full until 1711,
degli dei = scrittura divina, dei brdhmani"; Thumb- actually brought back with him a fairly complete chart
Hauschild, Handbuch des Sanskrit, 3rd completely rev. ed. of the Nagari alphabet, which, however, he does not call
(Heidelberg, 1958), vol. I, p. 188, ? 43: "Vielleicht heisstby name and erroneously imputes to the ancient Zoro-
devanagarl (sc. lipih) einfach "(die Schrift) von Devana-astrians of Persia ("J'ai inser6 dans cet ouvrage, pour la
gara," d. h. der "Gotterstadt" oder "g6ttlichen Stadt" = satisfaction des Curieux, un Alphabet de ces anciens
"Himmel," ein Ausdruck, durch den wohl die gottlichePerses, ou Guebres, en grandes et petites Lettres");
Herkunft dieser Schrift demonstriert werden soll...." v. Zachariae, "Das Nagari-Alphabet bei Jean Chardin,"
As a secondary derivative from devandgara, the form in his Kleine Schriften, pp. 12-14, also in WZKM, XIX
*Daivanagari (i.e., with vowel strengthening of the (1905), p. 243ff. A reproduction of Chardin's alphabet-
initial syllable instead of the second member of the table may be found on p. 519 of the article by Hauschild
compound) ought normally to be expected, but Whitney, referred to farther on in this note.
Sanskrit Grammar, 2nd ed. (Cambridge, Mass., 1955), Curiously, neither the names "Nagari" nor "Devana-
?1204e, gives many examples of this interchange gari"of
occur in the tables of the Nagari script prepared in
vowel modification. For a discussion of this point, 1664 by Pater Heinrich Roth, the Bavarian missionary
v. Thumb-Hauschild, p. 188, n. 11. On the curious to India, for inclusion in Athanasius Kircher's China
variant Daivanagari (!) with double vrddhi, used by Illustrata (Amstelodami, i.e., Amsterdam, 1667). On
the early German scholar Othmar Frank, v. note 9 below. the one table, which gives a transcription of the Lord's
3 E.g., A. A. Macdonell, Sanskrit Grammar for Students,
Prayer into Nagari letters, they are referred to as "Indian
3rd ed. (London, 1927), p. 2: "city writing of the gods";
letters" (Pro Exercitio huius Linguae ponam hic Pater
Max Miiller, Sanskrit Grammar (London, 1866), p.noster 1: Literis Indicis scriptum). Nor does Kircher him-
"Nagari of the gods, or possibly, of the Brahmans";
self anywhere in this work nor in any of his other volu-
St. Petersburg lexicon, under the entry Devanagari:
minous writings use the word "Nagari"; on this and many
"gottliche d.i. heilige Stadtschrift"; L. Renou and J. Fillio-
other particulars, including reproductions of Roth's
zat, L'Inde classique (Paris, 1953), p. 678: "citadine des tables, v. Richard Hauschild, "Die erste Publikation der
dieux" (but questioningly). indischen Nagari-Schriftzeichen in Europa durch Atha-
4 Alberuni's India, An Account of the Religion, Philo- nasius Kircher und Heinrich Roth" in Wissenschaftliche
sophy, Literature, Geography, Chronology, Astronomy, Zeitschrif der Friedrich-Schiller-Universitdt Jena, Jahrg.
Customs, Laws and Astrology of India about A.D. 1030, 5, 1955/56, gesellschaffs- und sprachwissenschaftliche
transl. into English by Eduard Sachau (London, 1910), Reihe, Heft 4/5, pp. 499-520); cf. also Zachariae "Das
vol. I, p. 173. Devanagari-Alphabet bei Athanasius Kircher" in his
5 G. Biihler, Indische Palaeographie, in Grundriss der Kleine Schriften, pp. 1-7, also in WZKM, XV (1901),
indoarischen Philologie und Altertumskunde (Strassburg,
pp. 313-320.
1896), p. 51: "Die alteste Inschrift aber, welche ganz in
That the name "Nagari" does not occur in Roth's
diesem Alphabete geschrieben ist, findet sich auf den tables does not, of course, exclude the possibility of its

This content downloaded from

149.233.192.93 on Tue, 21 Jul 2020 15:23:26 UTC
All use subject to https://about.jstor.org/terms
 MAURER: On the Name Devandgari 103

On the other hand, Devanagari does not appear to be "city of the gods," this being doubtless a periphrasis for
an old name,7 the earliest attestation possibly being that "heaven. "9

found in the introduction to Halhed's Code of Gentoo If, then, as seems probable from the scanty evidence
Laws, 1776: "The Shanscrit Character, used in Upper available, the original name of the script was simply
Hindostan, is said to be the same original Letter that was "Nagari," and "Devanagarl" but a later extension of it,
first delivered to the People by Brihma, and is now called the question next to be considered is the explanation of
Diewnagur, or the Language of Angels ...."8 In fact, the name "Nagar.L"
the use of the adverb "now" in this statement, if it is not According to the etymology that most readily presents
just a turn of phrase (which seems unlikely), even implies itself-indeed that usually followed by those who treat
that in Halhed's view the longer name had by the time "Devanagari" as a tatpurusa or karmadharaya compound
of writing come to replace an older name, which we can-"Nagari," as the feminine of the adjective ndgara
reasonably presume to have been "Nagari." (modifying lipz "script" understood), is a secondary de-
It is not difficult to account for the expansion of Nagari rivative of nagara "city, town." But in view of the wide
into Devanagari: from the beginning writing in India has range of meanings that can be implied in an adjective
been looked upon as divine in origin, as shown by the of appurtenance, the specific meaning involved here is
name Brahmi " [script] of Brahma," applied to the oldest obscure. It seems unlikely that by "Nagari" was meant
of India's scripts, and accordingly "Deva" was prefixed the mode of writing that was current in the city as oppos-
to the name "Nagari" to invest the script with a divine ed to the country, as this implies the existence of a
provenance. Incidentally, this same tendency toward "country-script" which can scarcely be imagined in any
the sanctification of the written character is also to be circumstances that prevailed in ancient or medieval
seen in the alternative explanation of Devanagari, alluded India.

to earlier, which makes of it a derivative from devanagara Possible, though not provable, is the explanation that
the name "Nagari" meant the script that originated in the
occurrence in his Sanskrit grammar, the first to be written
city, i.e., as a product of the refinement and culture of
by a European, which has unfortunately never been the city. Perhaps a particular city, from which its use
published, the MS last having been seen in 1800 by the spread, was meant, a view suggested in the St. Petersburg
Spanish scholar Lorenzo HervAs. On the vicissitudes lexicon. It is remarkable that Della Valle hazards the
of this MS, v. Bruno Zimmel's interesting article "Die
guess that "Nagari" is derived from the name of a small
erste Sanskrit-Grammatik: die merkwiirdigen Schicksale
village called "Naghra," a few miles from Cambay,
eines verschollenen Manuskripts" in Biblos, Jahrg. 5, he visited a temple dedicated to Brahma. But he
where
Heft 2 (1956), pp. 48-63; an enlarged version of this hesitates in this conjecture because he feels that in
article, published separately, Vienna, 1957, referred to in
Thumb-Hauschild, p. 172, n. 126, is unavailable to me. 9 But Othmar Frank in Grammatica Sanscrita nunc
7 So Macdonell, p. 2: "a term of late but obscure origin"; primum in Germania (Wirceburgi, i.e., Wiirzburg, 1823),
Muiller, p. 1: "No authority has yet been adduced from pp. 1-2, a work not available to me, but cited from Thumb-
any ancient author for the employment of the word Hauschild, p. 188, n. 11, calling the script "Daevanagarica"
Devanagari"; Renou and Filliozat, p. 678: ". . cette (i.e., Daivanagari 1), makes the curious assertion that it is
designation ne se rencontre pas dans les textes et paraisse so named because "it was employed in Daevanagara (i.e.,
n'avoir et6 r6pandue que par des pandits et surtout des Daivanigara), a city of the gods which is said to have
Europeens depuis le xvnIe siecle." There is no reason to once flourished near Kabul" (Praecipua vero ea est,
suppose any connection between the name "Devanagari"quae Daevanagarica nominatur, i.e., Daevanigarae, in
and "Devalipi" which is given in the Lalitavistara in a urbe deorum usitata, quae olim prope Cabulem floruisse
list of 64 scripts, where it occurs along with such names
fertur). This statement is remarkable not only because
as "Nagalipi, Yaksalipi, Gandharvalipi, Kinnaralipi,
nowhere else, so far as I know, is a city of this name found,
Mahoragalipi, Asuralipi and Garudalipi," all of which, but especially because he calls the script "Daivanagari"
if they are not entirely fanciful names, are but examples
instead of "Devao." Perhaps this peculiar form of the
of the old tendency in India to enshroud the written name, which seems to be an artificial overcorrection of the
symbol with a divine aura; v. S. Lefmann's ed. (Halle, 1.usual name (v. pendant to note 2 above) as well as the
Teil, 1902), pp. 125-126. explanation derive ultimately from the English scholar
8 Nathaniel Brassey Halhed, A Code of Gentoo Laws, Alexander Hamilton, under whom Frank studied (v. Ro-
or, Ordinations of the Pundits, from a Persian Translation,
sane Rocher's monograph Alexarndler Hamilton (1762-
Made from the Original, Written in the Shanscrit Language
1824), a Chapter in the Early History of Sanskrit Philology
(London, 1776), p. xxiv. (New Haven, 1968), pp. 109-110.

This content downloaded from

149.233.192.93 on Tue, 21 Jul 2020 15:23:26 UTC
All use subject to https://about.jstor.org/terms
104 Journal of the American Oriental Society 96.1 (1976)

explaining names, especially place names, resemblances tradition, as it is wholly possible that it arose from the
to ordinary words cannot be trusted.10 The history of the fortuitous similarity between the name of this caste and
Nagari script set forth by Biihler does not seem to favor that of the script. But it is certainly conceivable that the
the view of a particular city as responsible for its inven- Nagara Brahmans, while not the inventors of the NSgari
tion. The earlier inscriptions, prior to that of Dantidurga, script, which, as we have seen, was the result of the
are in fact often characterized by a mixture of Nagari gradual development of the tendency to spread out the
letters with letters of the Siddhamatrka type whose serifs atop the perpendicular strokes of the letters, were
vertical strokes are capped by wedges instead of horizontal somehow connected with its dissemination and populariza-
lines. This implies a gradual development of Nagari, tion. One small point which may perhaps lend some
in the course of which the wedges were extended on either support to the derivation of the name from the N5gara
side of the vertical strokes so as to become lines. Since Brahmans is that Al-Berfini calls the script "Nagara,"
these inscriptions are found in scattered localitiesnot in "Nagari,"
Gu- though in the same passage he speaks of
jarat and Maharastra, it may be that the typesanother of lettersscript which he calls "Ardhanagari," "Half-
with flared serifs or headlines were styled "nagari" Nagari,"inas being an amalgamation of "N5gara" and the
the sense that they were a product of a development Siddhamatrka character. The fact that he uses the
common to various cities throughout the whole area. feminine form in "Ardhanagari" makes it unlikely that
Agrawala is of the opinion that the name "Devanagari" failed to understand the simpler term "Nagari" correctl
derives from the eastern city of Pataliputra which was Can it be that the script was in those days called "Ni-
famed simply as Nagara, "the city, the metropolis," so gara [lipi]" as well as "Nagari"? This possibility is
that anything connected with it, whether a style of perhaps strengthened by the recurrence of the double
architecture or writing, etc., was termed "Nagara" forms "Nagara" and "Nagari" in the italianized transcrip-
(feminine "Nagari") "[the such-and-such] of The City tions "Nagher" and "Nagheri" in the writings of Pietro
(i.e., Pataliputra)." To the name "Nagari" thus formed della Valle centuries later.l4
was prefixed "Deva," which he states is found in Gupta Though "Devanagari" is but an extension of the much
inscriptions as the personal name of King Candragupta II older name "Nagari" and is therefore synonymous with
Vikramaditya.ll But this view seems entirely unsup- it, the two are not interchangeable. Thus, what is De-
portable, since, among other reasons, the writing on the vanagari may be called "Nagari," but not all that is Na-
Gupta coins, which would accordingly have to be De- gari may be termed "Devanagari." The reason for this is
vanagari par excellence, is not Devanagari, lacking as it partly implicit in what was said earlier: Nagari developed
does the horizontal headline, its chief characteristic. gradually over a large area, and in the course of this
Moreover, the Nagari script was, as we have noted, not development there arose varieties, some of which at least
developed in the eastern part of India, but in the west, have come to bear names of their own, e.g., "Balbodh"
where all the earliest inscriptions in it have been found. and "Modl," which are variants used in writing Marathi,
The adjective ndgara can also be derived from the and "Nandinagari" ("Nagari of Nandi," i.e., "giva or
proper name "Nagara," a particular caste of Gujarati Visnu," thus virtually equivalent in meaning to "De-
Brahmans,12 who according to Diringer claim to have vanagari !"), formerly widely employed in southern India
given the Nagari script its name.13 In the absence of any to transcribe Sanskrit Mss. None of these variants can be
evidence, however, it is very difficult to appraise thiscalled "Devanagari," but they may all be designated
"Nagari." The expanded name "Devanagari" may
10 The Travels of Pietro della Valle in India, from the indeed have arisen as a means of distinguishing the
Old English Translation of 1664 by G. Havers, ed. bybasic Nagari style of writing from some of the less im-
Edward Grey (Hakluyt Society, no. 84, London, 1892), portant varieties, or, what is perhaps more plausible, it
pp. 112-113. may have come into use first under the impulse, already
11 V. S. Agrawala, "The Devanagari Script" in Indian
explained, of sanctifying the terrestrial name "Nagari"
Systems of Writing (Delhi, 1966), p. 15. and was in the course of time followed by "Nandinagari"
12 According to J. H. Hutton, Caste in India: its and the others.
Nature, Function and Origins, 4th ed. (Oxford University WALTER H. MAURER
Press, 1963), under the entry "Nagar Brahman" in the UNIVERSITY OF HAWAII
glossary: "A caste of Gujarati Brahman associated in
origin with half a dozen towns in the north-east of Guja- 14 v. Note 6 above; the 'h' in these spellings is simp
rat (< nagar = town)," p. 290. a device regularly employed in Italian to prevent a 'g'
13 David Diringer, The Alphabet: a Key to the History followed by 'e' or 'i' from being pronounced like the
of Mankind, 3rd ed. (New York, 1968), vol. I, p. 290. 'j' in English 'jump'.

This content downloaded from

149.233.192.93 on Tue, 21 Jul 2020 15:23:26 UTC
All use subject to https://about.jstor.org/terms