Scribd
Upload
121 views4 pages

Bulletin of The School of Oriental and African Studies

Uploaded by

Sajad Amiri
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
121 views4 pages

Bulletin of The School of Oriental and African Studies

Uploaded by

Sajad Amiri
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Bulletin of the School of Oriental

and African Studies


http://journals.cambridge.org/BSO

Additional services for Bulletin


of the School of
Oriental and African Studies:

Email alerts: Click here


Subscriptions: Click here
Commercial reprints: Click here
Terms of use : Click here

A bactrian god

Nicholas Sims-Williams

Bulletin of the School of Oriental and African Studies / Volume 60 / Issue 02 / June 1997, pp
336 - 338
DOI: 10.1017/S0041977X00036430, Published online: 05 February 2009

Link to this article: http://journals.cambridge.org/abstract_S0041977X00036430

How to cite this article:


Nicholas Sims-Williams (1997). A bactrian god. Bulletin of the School of Oriental
and African Studies, 60, pp 336-338 doi:10.1017/S0041977X00036430

Request Permissions : Click here

Downloaded from http://journals.cambridge.org/BSO, IP address: 141.13.254.246 on 22 Jul 2014


A Bactrian god
The Kushan god /xo£Sooavo is attested on a group of rare coins of
Kanishka I, where he is depicted as a bearded male bearing a trident in his
right hand and riding a two-headed horse.1 It has been universally assumed
that the name /xo^Sooavo is connected in some way with that of Ahura Mazda,
the supreme deity of the Zoroastrian religion, the second element of the
Bactrian form being generally interpreted as an epithet meaning ' victorious '.2
The interpretation of yuol,hooavo as 'Mazda the victorious' has been
opposed by Humbach, who objects that such an epithet is not applied to
Ahura Mazda anywhere in the Zoroastrian literature and draws attention to
the fact that the Bactrian word for ' victorious' is attested in the form oavwho
(rather than *oavo). His solution is to postulate a compound *mazda-wana-
' Winner of Wisdom'.3 According to Humbach, ixo£,8ooavo 'is portrayed as
a Kusana king ..., [which] gives the impression that he is a sort of tribal god
of the Kusanas, or a family god of their royal family'. Since a double-headed
horse would be regarded by Zoroastrians as a creature of the Evil Spirit,
/xo^Sooavo is 'likely to be a non-Zoroastrian antagonist of Ahura Mazda'.
Humbach's objections to the interpretation of jxo^Sooavo as ' Mazda the
victorious' are valid if not absolutely conclusive. His own reconstruction is
open to similar objections, however. No god named ' Winner of Wisdom' is
attested anywhere else in the Iranian world. Nor is it clear why the -a- of a
compound such as *mazda-wana- should have been lost (contrast the personal
names /xo^Sa-fiavSayo and /xo^Sa-ixaprjyo, both meaning 'servant of
Mazda', which are attested in unpublished documents of later date).
The divine name /xo^Booavo occurs once more in a Bactrian inscription of
Kanishka I recently discovered at Rabatak in the province of Baghlan,
Afghanistan.4 This inscription contains a list of divinities headed by two
goddesses: la a/xaa vava oho ta a/xaa ofx/xa, probably to be understood as
' the lady Nana and the lady Umma'. There follows a series of male deities:
aopofxqt,ho iLot,hooavo opopapSo vapaaao p*upo 'Ahura Mazda, /xo£8ooavo,
Srosh-ard, Narseh, Mihr'. The natural interpretation of this wording is that
/xol,8ooavo is a divinity distinct from Ahura Mazda.5 It cannot perhaps be
ruled out that /xo^Sooavo may here be used as an epithet (rather than a name)

1
J. M. Rosenfield, The dynastic arts of the Kushans (Berkeley, 1967), 82-3; R. Gobi, System
und Chronologie der Miinzprdgung des Kusanreiches (Wien, 1984), 42 and pi. 167.
2
So far as I know, this etymology was first proposed by J. Duchesne-Guillemin, Paideuma,
vn, 1960, 213-14.
3
H. Humbach, Mithraic Studies, I, (ed.) J. R. Hinnells (Manchester, 1975), 139, followed by
G. Dj. Davary, Baktrisch: ein Worterbuch auf Grund der Inschriften, Handschriften, Miinzen und
Siegelsteine (Heidelberg, 1982), 234.
4
N . Sims-Williams and J. Cribb, 'A new Bactrian inscription of Kanishka the Great', Silk
Road Art and Archaeology, iv, [1995-]1996, 75-142.
5
The name of Ahura Mazda appears here in a more archaic spelling than the cuopoju.o{8o
attested a generation later on a coin of Huvishka (the authenticity of which is doubtful according
to Humbach, art. cit., 139-40). Forms such as coupofto^So, copopo^So, aiomxo^So, attested in
the Kushano-Sasanian period (as a personal rather than a divme name) probably represent MP
o(h)rmuzd. In my edition of the text I contemplated an alternative reading a6po fjLO^So, which
could be understood as ' Fire (son) of Mazda', cf. Av. atars ahurahe mazda ' Fire (son) of Ahura
Mazda', but it is highly doubtful whether this pregnant use of the genitive would have survived
into Bactrian.

© School of Oriental and African Studies, University of London 1997


NOTES AND COMMUNICATIONS 337

of Ahura Mazda, but an epithet containing a repetition of the god's name is


scarcely plausible.
Since the iconography of /Lio£Sooavo provides no support for his identifica-
tion with Ahura Mazda, the connexion in fact rests solely on a phonetic
coincidence in one syllable. It is time to be resolute in abandoning it. The
alternative which I should like to propose—and which, as I am happy to
acknowledge, has occurred independently to Dr Ilya Gershevitch—is to derive
/u.o£Sooavo from a form cognate with Vedic midhvams- 'generous, gracious'.
For the existence of this term in Iranian and its use as a name of a divinity
one may refer to Gersheyitch's reconstruction of the feminine equivalent
*mizdusi- (=Vedic mldhusi-) as the name of a goddess mentioned in two
Elamite tablets from Persepolis.6 In addition, it seems likely that Late
Khotanese mdjsai- ' woman' derives from *mizdusakT-, literally ' gnadige
(Frau)'. 7
In form midhvams- appears to be an unreduplicated perfect participle,8 of
which the expected form in Iranian would be *mizdwah-. The nasal suffix of
jiio^Sooavo finds a parallel in Khot. made, md(s)ddn(a)- 'gracious', which
Ernst Leumann long ago compared with mldhvams-? The comparison deserves
to be retained, though Leumann's formulation of the relationship between the
two forms (-vdrhs- active, -ana- middle participle) is ruled out by the athematic
nom. sg. m. made. Emmerick rightly classifies the Khotanese form as an -n-
stem, which he explains as *mizddn- from *mizda-Han-, i.e. *mizda- ' reward'
(Av. mizda-, Vedic midhd-, Greek /xioOos) plus the possessive suffix *-Han-
identified by Karl Hoffmann.10 This derivation is phonologically unexception-
able. However, since Vedic midhvhms- is known to have had cognates in Iranian
and probably even in Khotanese, it requires the assumption that one unusual
and archaic formation was replaced by another. Such an assumption seems
implausible, especially in view of the fact that Khot. made would be the regular
outcome of Olran. *mizdwah, nom. sg. m. of the expected *mizdwah-.n The
replacement of *-h- by *-n- in the rest of the paradigm could in theory be a
comparatively recent development due to the ambiguity of the nom. sg. in
*-wdh: in Old Iranian stems in *-wan- had nom. sg. m. in *-wa, but this was
systematically extended to *-wdh (>-(v)e) in Khotanese, 12 with the result that
a nom. sg. made < *mizdwdh could have been reinterpreted as belonging to a
stem *mizdwan-. If Bactr. fxolfiooavo is to be compared with Khot. md(s)ddn-,
however, it is necessary to place the creation of the stem *mizdwan- at a much
earlier period. So far as I can see, such a development would in fact be perfectly
plausible even in the remotest antiquity, in view of the interchange of *-wah-,
*-wan-, and *-want- in cases such as the Avestan name of the mythical hero
vTuuarj hant- (Vedic vivdsvant-) beside *vluuag han- and *viuuar) hah- (in the

6
1 . Gershevitch, TPS, 1969, 174, and apud R. T. Hallock, Persepolis fortification tablets
(Chicago, 1969), 732.
7
Rather than from *mazisakl- as previously proposed (N. Sims-Williams, Proceedings of the
first European Conference of Iranian Studies, I, (ed.) Gh. Gnoli and A. Panaino (Rome, 1990
[1991]), 292-4). I have discussed the phonological details of the new etymology in a contribution
to R. E. Emmerick, Studies in the vocabulary of Khotanese, III, Vienna, 123-4 (forthcoming).
8
F. B. J. Kuiper, Ada Orientalia, xu, 1934, 234; J. Wackernagel and A. Debrunner, Altindische
Grammatik, n/2 (Gottingen, 1954), 910.
9
E. Leumann, Zur nordarischen Sprache und Literatur (Strassburg, 1912), 72.
10
R. E. Emmerick, Saka grammatical studies (London, 1968), 338, referring to K. Hoffmann,
'Ein grundsprachliches Possessivsuffix', MSS, vi, 1955, 35-40.
11
For the simplification of the cluster *dv in made < *md(s)dve cf. nade < *nrtdwah, nom. sg.
of nadaun- < *nrtdwan- ' man'.
12
cf. Emmerick, op. cit., 337. The secondary addition of *-h to the nom. sg. m. is also found
in Khot. -/--stems, e.g. pate ' father' < *pitd-h. Its function was perhaps to characterize these forms
more clearly as masculines.
338 NOTES AND COMMUNICATIONS

patronymic adjectives viuuaij hana- and vluuat) husa-), or the Avestan feminine
personal names vispa. tauruuain- and vispa. taurusT-, which are formed as if to
stems *°tauruuan- and *"tauruuah- respectively, though the original form of the
stem was probably *°tauruuant- (cf. Hittite tarhunt-, Vedic tUrvant-, Av. nom.
sg. m. tbaeso. tauruud ' overcoming enmity' and varsdra. tauruua ' overcoming
resistance', inflected on the analogy of possessives in -uuant-).13
If the inflexion of Khot. made, md(s)ddn(a)- as an -n-stem is comparatively
old, as the equation with /xo^Sooavo would imply, this may help to account
for the apparently identical inflexion of Khot. urmaysde ' sun', whose nom.
sg. evidently derives directly from Old Iranian *ahura-mazddh (cf. Av. ahuro
mazdd, OP auramazdd) but whose other forms indicate a stem urtnaysddn(a)-.14
In the Khotanese Buddhist literature mdfs)ddn(a)- is chiefly used as an epithet
of the Buddha. If the underlying *mizdwan- was used in pre-Buddhist Khotan,
as in Bactria, as the name or epithet of a divinity there would be a clear
associative reason for it to have provided a source for the analogical restructur-
ing of the awkward paradigm of *ahura-mazdd-.ls Cf. the development of
forms such as Vedic pdnthdnas, Avestan pantdno as nom. pi. to the stem pdnthd-
Ipantd- ' p a t h ' by analogy with the synonymous -«-stem ddhvan-/aduuan-.16
Several facts suggest that one might consider identifying ixot,8ooavo with
the god who is depicted on the coins of virtually all the Kushan rulers in the
form of the Indian Siva and who is named as or/po ( = Av. vaiius, MPers. way,
Sogdian wysprkr) on coins of Kanishka and his successors.17 Since this god
was evidently one of the chief deities in the Kushan pantheon, one naturally
expects him to be included in the list in the Rabatak inscription. If fiot^Sooavo
means ' the Gracious one', the name is virtually synonymous with that of Siva
( = Vedic sivd- 'kind, benevolent, auspicious'). The portrait of jiio^Sooavo
shows him holding a trident, the symbol of Siva, and riding a two-headed
horse, which could possibly be related to the dual nature of the god Vayu, the
way T weh ' Way the better' and way T wattar ' Way the worse' of the Pahlavi
texts, and more distantly to the three-faced wysprkr of the Sogdian Vessantara
Jataka and the three-headed orjpo/Siva portrayed on certain Kushan coins.18
In other respects, however, the iconography of /juo^Sooavo is entirely different
from that of orjpo, who is always shown standing and beardless. Until a closer
parallel to the figure of /xo^Sooavo is found, the identification of the god must
remain speculative.
NICHOLAS SIMS-WILLIAMS
School of Oriental and African Studies, London

13
See further Wackernagel-Debrunner, op. cit., 894, 903-5; J. Schindler, Investigationes
philologicae et comparativae: Gedenkschrift fiir Heinz Kronasser, (ed.) E. Neu (Wiesbaden,
1982), 200-201.
14
Emmerick, op. cit., 347-8.
15
cf. the OP gen. sg. Auramazdaha, as explained by F. B. J. Kuiper, On Zarathustra's language
(Mededelingen der Koninklijke Nederlandse Akademie van Wetenschappen, Afd. Letterkunde,
N. R., XLI/4), (Amsterdam, 1978), 7: gen. ending *-ah secondarily added to disambiguate the
nomVgen. in *-ah.
Conversely, the GAv. nom. sg. aduud may be due to influence of pantd. See C. Bartholomae,
Grundriss der iranischen Philologie, I, (ed.) W. Geiger and E. Kuhn (Strassburg, 1895), 118.
17
See H. Humbach, Monumentum H. S. Nyberg, I (Acta Iranica, iv), Tehran-Liege, 1975,
402-8. As Frantz Grenet has pointed out to me, one of the problems to be confronted is the fact
that both orjjx) and fio^Sooavo are depicted on coins belonging to the same (third) emission
of Kanishka.
18
This last point too I owe to Frantz Grenet. Humbach, art. cit., 405, suggests that the third
face of wysprkr may correspond to a third ' neutral' aspect of the god.

You might also like