Kalidasa and the Agnimitra problem

kak at max.ee.lsu.edu kak at max.ee.lsu.edu
Mon Jan 25 19:38:10 UTC 1993

.nr PS 11
.nr VS 14
.nr LL 5i
.nr PO 2i
.nr HM 2i
.nr FM 2i
delim $$
Ka\*_lida\*_sa and the Agnimitra Problem
Subhash C. Kak
Louisiana State University, Baton Rouge 70803, USA
The Agnimitra problem is as follows: If Ka\*_lida\*_sa lived around 400 A.D.,
as many scholars now believe, why did he write a play on a minor king
who had lived more than 500 years earlier?
This question is especially significant since Agnimitra is the only historical
character in his plays, and one might further ask why did Ka\*_lida\*_sa
not use more famous kings and heroes of contemporary times in his plays.
This problem has not been discussed because of 
the assumptions which lie at the basis of the accepted chronology of
Indian literature.
However, in recent years,
many of these assumptions have been shown to be incorrect.
In the absence of any archaeological evidence to support it,
the theory of invasion of the Aryans is now characterized as a myth.\**
J.G. Shaffer, "The Indo-Aryan invasions: cultural myth and archaeological
reality", in \fIThe People of South Asia\fR, edited by J.R. Lukacs.
New York: Plenum, 1984.
The Bra\*_hmi\*_ script has been shown\**
S.C. Kak," A frequency analysis of the Indus script," \fICryptologia\fR,
vol. 12, 1988, pp. 129-143.
S.C. Kak, "Indus writing," \fIMankind Quarterly\fR, vol. 30, 1989, pp. 113-118.
S.C. Kak, "Indus and Brahmi: further connections," \fICryptologia\fR,
vol. 14, 1990, pp. 169-183.
to be derived from the Indus
script of the third millennium B.C.
There is reason now\**
S.B. Roy, "Chronological framework of Indian protohistory -- the lower
limit," \fIJournal of the Baroda Oriental Institute\fR, vol. 32,
1983, pp. 254-274.
S.C. Kak, "On the chronology of ancient India," \fIIndian Journal of
History of Science\fR, vol. 22, 1987, pp. 222-234.
to judiciously
use the astronomical and literary evidence
which suggests greater antiquity for the earliest Indian literature
than the arbitrary chronology popularized
by Max Mu\*:ller and his successors.
Furthermore Seidenberg's research has shown\**
A. Seidenberg, "The ritual origin of geometry", \fIArchive for History of
Exact Sciences\fR, vol. 1, 1962, 488-527.
A. Seidenberg, "The origin of mathematics", \fIArchive for History of
Exact Sciences\fR, vol. 18, 1978, 301-342.
that there exist
reasons to reopen the
question of the dating of the \fIsu\*_tra\fR literature.
His work shows that the late dating of \fIs\*'ulvasu\*_tras\fR was
partly prompted by the attempt to see a Indian geometry as
following the rise of Greek geometry.
His demonstration that \fIS\*'atapatha Bra\*_hman\*.a\fR, which is
conservatively dated centuries before earliest Greek geometry,
itself contains Indian geometry compels a new look at the entire question.
It also calls for a review of the evidence used to fix the chronology of the
literature that followed the \fIsu\*_tra\fR period.
This provides our motivation to review the controversy related to
Ka\*_lida\*_sa's date.
The dating of Ka\*_lida\*_sa
poses several interesting problems.
The time window to seek Ka\*_lida\*_sa's
date is provided by Agnimitra (150 B.C.),
the subject of his play \fIMa\*_lavika\*_ and Agnimitra\fR and the Aihole
inscription praising him which is dated 634 A.D.
There also exists the tradition associating Ka\*_lida\*_sa with Vikrama\*_ditya
of Ujjayini\*_,
who is believed to have founded the Vikrama era in 58 or 57 B.C.
But this tradition is manifestly late:
the earliest account connecting Ka\*_lida\*_sa with Vikrama is by the eleventh
century king and author Bhoja in his \fIS\*'r\*.nga\*_ra Praka\*_s\*'a\fR.
Legend describes Ka\*_lida\*_sa as one of the nine jewels at the court of
King Vikrama\*_ditya together with Amarasim\*.ha, Veta\*_labhat\*.t\*.a,
Vararuci, and Ghat\*.akarpara.
The tradition that a Vikrama\*_ditya founded the Vikrama era has been traced
back only to Sam\*.vat 1050 (993 A.D.).
Initially this era was apparently
called the Kr\*.ta era.\**
Buddha Prakash, \fIStudies in Indian History and Civilization\fR. Agra: Shiva
Lal Agarwala, 1962.
There exist several inscriptions from 225 A.D. to 424 A.D. that use this name.
Later it is called the
Ma\*_lava era and the earliest reference
to this name\**
F. Edgerton, \fIVikrama's Adventures\fR. Cambridge: Harvard University
Press, 1926.
appears in Sam\*.vat 428 (371 A.D.).
But there is a Jain tradition according to which Vikrama began this era
470 years after Maha\*_vi\*_ra's nirva\*_n\*.a which agrees with the
dating of 58/57 B.C.
According to this tradition the era was founded to celebrate the
magnanimity of King Vikrama\*_ditya who freed his subjects from debt
to mark his conversion by Siddhasena Diva\*_kara.
But it is conceivable that these late Jain works interpolated the
account of Vikrama as the founder of this era at a time when the facts
were quite forgotten.
Al-Bi\*_runi\*_ (1030 A.D.)
believed there had been two Vikrama\*_ditya's, with the
latter being the one who expelled the S\*'aka, and the former being the
one after the era was named.
He states:\**
E.C. Sachau, \fIAlberuni's India\fR. 1910, Reprint 1989 (Delhi).
The epoch of the era of S\*'aka or S\*'akaka\*_la falls 135 years later
than that of Vikrama\*_ditya.
The here-mentioned S\*'aka tyrannised over their country between the
river Sindh and the ocean, after he had made A\*_ryavarta in the midst
of this realm his dwelling-place....
The Hindus had much to suffer from him, till at last they received help
from the east, when Vikrama\*_ditya marched against him, put him to flight
and killed him in the region of Karu\*_r, between Multa\*_n and the castle
of Loni\*_.
Now this date became famous, as people rejoiced in the news of the death
of the tyrant, and was used as the epoch of an era, especially by the
They honour the conqueror by adding S\*'ri\*_ to his name, so as to say
S\*'ri\*_ Vikrama\*_ditya.
Since there is a long interval between the era which is called the era of
Vikrama\*_ditya and the killing of S\*'aka, we think that the
Vikrama\*_ditya from whom the era has got its name is not identical with
the one who killed S\*'aka, but only a namesake of his.
The significance of this quotation is that Vikrama of the era fame had
nothing to do with the defeat of the S\*'akas.
This agrees with the account in the Jain tradition.
Many scholars believe that, notwithstanding the above, the association of 
with Vikrama really refers to Candragupta II (375-413 A.D.) of the Gupta
dynasty who bore the title Vikrama\*_ditya.
However, Candragupta II's capital was Pa\*_t\*.aliputra, whereas 
poetry describes Ujjayini\*_ in great detail in his \fIMeghadu\*_ta\fR.
And if the tradition of 
being a courtier of Vikrama\*_ditya has any
truth to it, one would have expected references to 
On the other hand, Candragupta II defeated the S\*'aka ks\*.atrapa
of Ma\*_lava\*_ some time after 390 A.D.
Annexing this region, he established a provincial capital at Ujjayini\*_.
Nevertheless, it is quite conceivable that the legends and references related
to Ka\*_lida\*_sa
were penned when the facts were quite mixed up and, therefore, 
association with Candragupta II was transferred to the era that bore the
name of Vikrama.
This means that the controversy can be resolved only by considering
evidence within the Ka\*_lida\*_sa corpus itself.
This brings us to Agnimitra, the one historical character in his plays.
Why did Ka\*_lida\*_sa write a play about him?
Plays are generally written about heroes from myth or history, or
characters, otherwise insignificant, who are from the contemporary world.
This is true of Shakespeare whose major characters are either mythical
heroes like Hamlet, or Roman generals and conquerors, or famous English kings
from not too distant past.
One would expect this to be true of Ka\*_lida\*_sa as well.
Agnimitra was a minor king who, according to Pura\*_n\*.as, ruled for a mere
8 years.
But before this he served as the viceroy of his father Pus\*.yamitra
S\*'unga in Vidis\*'a\*_.
In \fIMa\*_lavika\*_ and Agnimitra\fR Pus\*.pamitra (Pus\*.yamitra)
is described as Agnimitra's general.
Now it is conceivable that Pus\*.yamitra, who had been the general of
Br\*.hadratha, the last Mauryan king of Pa\*_t\*.ali\*_putra,
continued to call himself sena\*_pati during his reign.
But during the time of the imperial Guptas, when the Pura\*_n\*.as were
being expanded and revised, Pus\*.yamitra was remembered as the
king who overthrew the Mauryas and ruled for 36 years.
There existed another reason why Pus\*.yamitra was remembered long after
his death.\**
Buddha Prakash, \fIop. cit.\fR,
for this and other information in this paragraph.
This is because Patan\*~jali in his \fIMaha\*_bha\*_s\*.ya\fR
mentions the As\*'vamedha sacrifice by him.
A stone inscription from Ayodhya\*_ declares that he performed two
As\*'vamedha sacrifices.
According to Buddhist tradition, Pus\*.yamitra persecuted that religion.
Jain tradition claims that after his death he was succeeded by
Balamitra-Bha\*_numitra, and it assigns them a reign of 60 years.
It has been suggested, therefore, that after Pus\*.yamitra death
the empire was divided and Agnimitra ruled in Pa\*_t\*.liputra for 8
years, and Balamitra-Bha\*_numitra set up an independent state in the
This suggests that it would have been highly unlikely for Ka\*_lida\*_sa to
base his play on a minor king, 500 years after his death.
We return now to the traditional account which places Ka\*_lida\*_sa
about 50 B.C.
Considering the Jain accounts that suggest that Vikrama was a minor
regional king, it is easy to visualize the court poet writing a play
about a recent king who was famous because of the As\*'vamedha
And since Ka\*_lida\*_sa was apparently from Ujjayini\*_, it would have
been natural to write about the regional viceroy.
Furthermore, Ka\*_lida\*_sa would have known that Pus\*.yamitra continued
to style himself sena\*_pati after seizing the throne.
In this scenario, one could assume that Ka\*_lida\*_sa was born about
100 B.C. and, therefore, Agnimitra's stewardship of the As\*'vamedha
in Ujjayini\*_ would have made him well remembered hero.
On the other hand, in 400 A.D. the many intervening As\*'vamedhas
of numerous kings, including the Guptas, would have made the one
stewarded by Agnimitra a long-forgotten event.
The above arguments do not prove that the traditional account of 50 B.C.
being the date of Ka\*_lida\*_sa is true.
But it makes this date more probable than 400 A.D. accepted by many
Journal of the Oriental Institute, vol 40, pp 51-54, 1990.

More information about the INDOLOGY mailing list