Manadeva Samvat: an investigation into an historical fraud.

Don't take anybody's words for granted. Motto of the Royal Society, London

There is no Manadeva II, King or Feudatory


Between Ganadeva (Samvat 479-487) and sivadeva I (Samvat 512-535) there
is no king named Manadeva. Although the gap is of about 25 years, there is no trace of "Manadeva II" in Mangal Bazar, Patan
inscription of Bharavi, dated Samvat 492, nor in Desabhattarika's
sankhamula inscription, dated Samvat 495. Had there been an outstanding
king, able to found an epoch era with a name, Manadeva, this era should
have been used, by sivadeva I, in his Visnupaduka Phedi inscription of
Samvat 512. At least, Jayadeva II's Pasupati inscription dated Samvat
157 should have mentioned him if there were an illustrous king able to
found an epoch era which Jayadeva II himself had used.


"Manadeva II" in the Nepalavamsavali, Kirkpatrick, Levi etc., is
clearly a result of scribal error which has been handed down to all the
19th century Bhasavamsavalis. This is quite clear from the comparison
of the details mentioned about Ganadeva in the Gopalarajavamsavali and
of "Manadeva II" in Nepalavamsavali, i.e.,
Ganadeva>Manadeva>Manadeva. Based on Kirkpatrick's summary of the
Nepalavamsavali, Levi wrote,

Between Udayadeva and Ganadeva
(Gunakamadeva), the Vamsavalis place Manadeva II, under this reign
Nepal suffered for three years from a terrible drought; Manadeva
brought an end to it by offering all his treasures to Pasupati. The
Vamsavali of Kirkpatrick alone registers this tradition (Levi, 1905 Vol
II: 121).

This conclusion of Levi is based on the following summary by Kirkpatrick, in his "Historical Sketch of Nepal",


Maun Deo (the 2nd) 45 in whose reign Nepal was afflicted during three
years with a severe drought, which ceased on the Rajah's propitiating
the god Pusputty by an oblation of all his reaches (1811: 260).

This summary by Kirkpatrick, in turn, is based on the following entry in the Nepalavamsavali,


Raja srimanadeva varsa 45//tena pasupatibhattarakaya varsa trayam
navristih vrstiakarsanamna kosam maniyuktamca dattavan// (folio 5b
lines 3-6)

This entry in NV is based on the misreading of the following entry in the Gopalarajavamsavali,


Rajasriganadeva varsa 45// tasya rajyem nepalabhumi varsatrayam
anavrsti varisovrsti akamksanaya kamanenah sripasupatibhattarikaya
mahanaga nirjityah tasya maniyukte ganadeva nama kosa kritam
pradhokitam tatprabhavat mahavrsti kritam praja sukhi bhavati// (folio
21a line 4- 21b line 1)

At least from that point onward, the
contamination of the Vamsavali tradition goes on uninterrupted.
Beginning with Pandit Gunananda's (A.D. 1829) work (Wright, 1877), to
the Vamsavalis compiled in the 1890s, most of them invariably insert a
king named Manadeva as a son of Udayadeva, between Udayadeva and
Ganadeva (or the chronicler's Gunakamadeva). In Siddhiman Simha
Basnet's work (dated A.D. 1878), published under the title
Rajabhogamala (the National Archives Cat. No IV 332, another copy in
Leo E. Rose collection in the Berkeley Campus, University of
California), this insertion is visibly done with the help of a
correction mark at the bottom of folio 45a line 13, where the King's
name is corrupted to Raja Mamdeva with a rule of 53 years. In the
Kaiser Library, there is an interesting vernacular chronicle (No
9/1276) offered to General Kaiser by Pandit Ramnath Calise of
Mahottari. Here King Manadeva is recorded as the son of Udayadeva, with
25 as reginal years. Similarly, another text of Bhasa Vamsavali (last
record A.D. 1877) in our personal collection has the following entry,


srimanadeva, years of role 45. On hearing the former story of Vir
Vikramaditya Sen, the King re- introduced the Vikram Samvat all over
the eastern and western dominions and made it famous everywhere.
(folios 91- 92)

Nearly the same text appears in the Bhasa
Vamsavli Part II edited by Lamsal (1966:1) where the King appears as
Raja Nandadeva, with only 13 years' rule. Since every well- to- do
family in Nepal has a copy of such chronicles the contamination went on
multiplying with every new copyist, almost ad absurdum.

Manadeva Samvat: A Bogus Interpolation in the Text


"Mandevabda 304" is clearly a bogus interpolation in both the British
Museum copy and the National Archives copy of Sumatitantra. There the
interpolation, a faint line in a later hand inserted at the bottom of
the page, is all too evident. The correction mark, a hamsapada at the
top as well as at the bottom of the line 5 following the punctuation
mark which concludes the line ending in raja kramenatu, is clearly
visible. The post--colophon stray folio has "Manndevsya rajyabda 304'
by a recent scribe. In the British Museum copy, this is a prose line in
the midst of verses in Anustubha metre. Probably, it is an elaboration
of a verse similar to the modern Surya Siddhanta, Chapter I:23 which
enjoins that

In the present twenty- eighth, Age, this Golden
Age is past : from this point, reckoning up the time, one should
compute together the whole number. (Burgess, 1860/1997: 13.)


As Sumatitantra is a tantra, its six ways of calculating ahargana
(literally, total number of mean civil days elapsed since the beginning
of the Kali) begins from the start of the Kali Yuga, expressing the
expired years of the Kali as "bhavisyam sampravaksami kalikanca
yathakramam" i.e. "introducing the future Kali years in a sequence."


sankaranarayana (ca. A.D. 825-900) of Quilon, in Kerala, rose to
eminence through his commentary on Laghubhaskriya, and later appointed
chief court astronomer of Ravivarman of the Cera dynasty of Kerala. He
quotes a verse from acarya Sumati in his Vivarana on Laghubhaskariya
(dated A.D. 869), in connection with lunar and solar eclipses, ten
years before the so-called Manadeva Samvat 304. This line is clearly an
interpolation in the Sumatitantra texts. These texts give the mean
positions of the planets of mid- night of Saturday/Sunday, March 20/21,
505 A.D. in Avanti.

The Sumatitantra gives an R- Sine Table
with 90 divisions of a quadrant, one for each degree of the arc. It is
sometimes accurate up to the 12th decimal place (e.g., 889 for
15[degrees], 1790 for 30[degrees], 2431 for 45[degrees]" 2977 for
60[degrees]) whereas for other degrees the text only gives round
figures by ignoring decimal points lower than 0.50, (e.g., 3321 for
75[degrees]). Most classics on Indian astronomy dated before Vatesvara
(b.800 A.D.) give 24 Sines for a quadrant, with the value of Radius=
3438. This is so in Paitamaha Siddhanta, Aryabhata I, Brahmagupta,
Lalla, and modern Surya Siddhanta. The Sines for Other degrees can, of
course, be derived on the basis of the 24 Sine tables, and smaller
decimal points are inevitable if lesser arcs are taken as units.
Varahamihira, on the other hand, gives Sines for a radius of 120 with
225" seconds as interval. As the table differs from all others,
including the one given by Varahmihira, the text of Sumatitantra seems
to be infested with interpolations by later hands.

Of the two
available copies of the text, the copy in the National Archives is by
three different hands, copied for the second time, as mentioned clearly
in its colophon. In the British Museum copy, there are a number of
folios with lines erased, added, and the whole folios erased after
having copied them. A number of folios are also blank. A number of
folios are post-colophon, such as 121a-b, 122a-b, 123a-b. The folio
124a-b has no corresponding equivalent in the National Archives copy.
At the end of it comes another colophon which dates the text to 16 days
later. The first colophon dates the text on Samvat 476 Pausa sukla 9;
the second one dates on Samvat 476 Pausa Krsna 10. There are 124 folios
in this recension. All folios are numbered in letters consecutively on
the left hand side and in numerals on the right hand side.

In
the National Archives copy, on the other hand, there are 152 folios;
folio 13a 13b are missing, and after 15a there are 3 folios which are
unrelated to the context. Upto 143b folio there is consecutive
numbering--first in letters upto folio 62b, then in decimal numerals up
to the end. Decimal numbering is also used in the main text. This
corrupt and contaminated manuscript is copied at least by two different
hands on two different kinds of palm.

Unlike the popular
Caitradi expired saka (computed from Tuesday, March 3, 78 A.D.) the so-
called "Manadeva Samvat" is none other than Karttikadi current saka
which is to be computed from Thursday, October 18, 76 A.D. Karttikadi
saka was also prevalent in Saurastra. It is referred to by Bhaskara I
(A.D. 550-629) in his aryabhatiya-bhasya ( folio 127 of the transcript
from the two manuscripts in Malayalam script in the collection of the
Government Library of Oriental Manuscripts, Madras, No R- 14850 ).
First used by Amsuvarma and his successors, it is a lokakala, with 500
dropped from the total current year. saka Samvat, the astronomer's era
par excellence, too, was a lokakala, begun with the figure for century
left out in computation. Al-Biruni notes,

Common people in
India date by the years of a centennium which they call samvatsara. If
a centennium is finished, they drop it, and simply begin to date by a
new one. This era is called lokakala i.e., the era of the nation at
large. But of this era people give such totally different accounts,
that I have no means of making out the truth. (Sachau, 1910, II;
Chapter XLVIII.)

All current eras begin 1 year before the
expired one. In Karttikadi samvat, Karttika comes before Caitra whereas
in Caitradi samvat. Caitra comes before Karttika. Only five months are
common to both the systems i.e., Karttika, Margasirsa, Pausa, Magha,
Phalguna, whereas uncommon ones are Caitra, Vaisakha, Jyestha, asadha,
sravana, Bhadra, and asvina. So the conversion of one era into another
is not such a simplistic addition/subtraction arithmetical operation as
made of it by some "authorities" (See Bhandari, Purnima 103, VS 2058
Pausa, pp 2-13).

Amsuvarma, most likely under the influence
of Harsa Samvat which is also Karttikadi, current, and amanta, began to
use the then prevalent era with 500 left out since year 529. So far, in
the last 144 years, we have not found a single inscription dated
between Samvat 1 to 28. Nor is there any reasonable argument for
launching an epoch era with Samvat 29. Obviously, Amsuvarma just
dropped the figure for hundreds and used the current epoch era. Several
scholars had speculated such a possibility as early as Bhagwanlal
Indraji (1885:45). Amsuvarma makes it more than clear when he installs
a gold repousse kvaca image of Garudanarayana at Cangu where the date
is mentioned as

Om ekatrimsattme varse vartamane svasamsthaya maghasuklatrayodasyampusyena saviturdine


The late Dhanavajra Vajracarya as well as his guru, the late Pandit
Nayaraja Pant, was both unable to interpret the correct meaning of the
words vartamane and svasamsthaya because both believed in the so-called
"Manadeva Samvat." They thought that the words qualified the kavaca,
set up at present, rather than the epoch era. The current epoch years
are qualified with words such as pravartmane or vartamane whereas
expired ones are prefixed or followed by words such as gate, atite,
pratite, yute etc. (On the subject of current year and expired year and
their conversion, (See Sewell and Dikshit 1896:40; Pillal, 1922: 52-53;
Ketkar, 1923: 18-19)

When Franz Keilhorn calculated 200
Vikrama Samvat inscriptions with weekdays from India and 29 Nepala
Samvat inscriptions with weekdays from Nepal, he found that some are
verifiable as current years and others as expired ones, while a number
of inscriptions, both from India and Nepal, have irregular dates that
cannot be verified ( Kielhorn, 1890, January: 20-40; June:160-187, and
November: 354-374; September: 1888,. 246-253 )

Of the 29
Nepalese-inscriptions with weekday, examined and computed by Kielhorn,
4 have years mentioned as expired. The words used are gate, prayate,
yate, yute, 19 have expired years without any appellation; and 2 are
dated according to current year, without mentioning this. Others were
irregular ones. In the ease of the Vikrama Era, of the 200 inscriptions
Kielhorn examined, 50 have irregular dates: some are regular by one
siddhanta, irregular by another; some dates are doubtful readings; some
are with a wrong weekday; some will work only if the immediately
following year is taken into consideration. Some tally only if the
preceding year is considered. Some work with amantamana others with
purnimantmana; both also have Southern and Northern variety. All these
variations in dating the inscriptions are dated in the Vikrama Samvat.


It will, therefore, be a naive generalization if we assume that in all
the Licchavi inscriptions, particularly with intercalation, the dates
are all regular, uniform, and absolutely flawless renderings and
readings. Not all the inscriptions are uniformly preserved. There is a
serious problem of illegible and variant readings by different
epigraphists--both in the case of numerals and ligatures (e.g., where
Gnoli read Ganadeva, Dhanavajra reads Gangadeva, in Capaligaon
Inscription, dated Sara. vat 489 sravana sukla 12). In some
inscriptions or documents the same epoch year may be expired whereas in
others they can be verified only as current years. This task is made
less than surmountable by the fact that of the 200+ Licchavi
inscriptions, only 2 have weekdays.

The current year begins
with month 1 year 1, whereas the expired one begins with year, day 1 of
month 1 and becomes year 1 and month 13th in the 13th month only. As
most inscriptions in Nepal or India do not explicitly mention whether
the given year is current one or expired one, the only way to ascertain
a given date is to compute all the available elements of pancanga, i.e,
tithi, vara, naksatra, karana, and yoga, if they are explicitly
mentioned in an inscription. As Pillai puts it,

The weekday
is the crucial test in the vast majority of verifiable Indian dates and
in the absence of a weekday, an Indian date is usually pronounced
unverifiable; unless there is an eclipse on that date. Where we have a
date that merely gives a tithi, a naksatra, and a year without the
week- day, we say that the day cannot be verified, i.e., proved free
from the probability of error, because every year must contain such a
tithi, and such a naksatra, we cannot assert with any degree of
confidence that the year- data is free from error (Pillai, 1922: 4-5).

Or as Ketkar puts it,


Citations about Samvatsaras, months, tithis are not sufficient for the
determination of a date. Details about the era: whether the samvat is
current at the Mesadi or at date; whether the date is expired or
current; whether Caitradi or Karttikadi, whether the date is derived
from the Surya Siddhanta or Arya Siddhanta, these details should be
clearly and fully made out before commencing the calculation. These are
the uncertainties that often beset the work of an epigraphist (Ketkar,
1923:80-81).

Unfortunately for us, there are only two
documents from ancient Nepal, one the gold repousse inscription dated
Current Karttikadi Samvat 31 Magha sukla 13 Sunday when the Moon was in
Pusya naksatra (verified by Regmi, 1983: 268, for Sunday, February 4,
608. A.D.); the second one, a colophon dated Expired Karttikadi Samvat
301 Vaisakha sukla 7 Sunday, while the Moon was in Pusya Naksatra and
the Yoga was Siddha (verified by Petech, 1984:29 for Sunday 13, 878,
A.D.; wrongly verified for Sunday April 23, 878 A.D. by Regmi, 1983:
268; but rightly verified by himself for A.D. 878 April 13, Sunday on
p. 23) We believe that these two documents with weekdays were verified
by the two historians by using Pillai's tables (1911/1922). We have
also cross- checked both the dates by using Jacobi, 1892, Sewell and
Dikshit, 1896, Ketkar, 1923, and Sewell, 1926. Except for a small
variation of a few ghatikas and palas, their verifications are found as
correct ones.

Of all the newly found Licchavi inscriptions,
just a little over a dozen in the last 33 years, none has a weekday.
Even if all other elements of Hindu calendar are mentioned, such as
naksatra or muhurta as in Cafigu Inscription of Manadeva, dated Samvat
386 Jyestha sukla Pratipad, Abhijit muhurta, with the Moon in the
Rohini naksatra, its date cannot be verified as totally free from
error, because such astronomical conjunctions are not rare. So the late
Pundit Nayaraj Pant's claim that he successfully calculated the date of
this inscription on the basis of Jyautisa vedanga does not carry any
significance. Similar claims had been made in the past by other
authorities using different siddhantas, and epoch eras.

For
instance, Shankar Man Rajvamsi calculated these two Licchavi documents
with weekdays by using his own theory of a Licchavi Era founded 22
years before the founding of saka Era in 78 A.D. He came out with the
following startling results: A.D. 442 April 27 Monday for Manadeva's
Cangu inscription, A.D.584 February 19 Sunday for the Amsuvarma's
repousse inscription, and A.D. 854 April 6 Sunday for the colophon of
the Sausruti Samhita Sahottaratantra. He claimed that all the elements
given in the inscription and the colophon match perfectly well with his
own theory of the Licchavi Era founded in A.D. 56 and Amsuvarma Era
founded in A.D. 552.(Rajavamsi: 1970:43-47.). In his magnum opus. NR
Pant himself wrote,

In the past, all have claimed that they
had verified the date in the Cangu inscription of Manadeva. Such date
occurs 6-7 times within a century. So it is clear that conclusions
cannot be drawn on the basis of such computations of dates (NR Pant et
al., 1987:556).

The Tibetan Text does not Mention "Manadeva II"


Among the so- called four evidences in support of the "Manadeva
Samvat," the Tibetan text does not mention Manadeva at all. On the
contrary, bSo nams rtse mo (A.D.1142-1182) mentions that Amsuvarma
counted years from saka 438 (probably a misinterpretation for 498,
resulting out of misreading candra for randhra, because candra also
means 3), and 242 years after the appearance of Amsuvarma, King Khri
gtsugs lde btsan (Ral pa cen) came to power. The relevant quote from
the text is the following:

The Buddha entered womb in the
fire- rabbit year and was born in the iron-dragon year. He got the
perfect enlightenment in, the water- tiger year and entered Nirvana in
the process of beginning the earth- mouse year (2133 B.C.). 137 years
later, King Nanda appeared. This account occurred in Tarkajvala
(=Toh.3856). After 800 years from this King, Candragupta appeared. 132
years after him, King sudraka appeared. After him, counting years, when
274 years passed by, the Nepalese continued to count years after it,
and still after 438 (498 ?) years, King Amsuvarman appeared. After him,
by counting years, when 242 years passed by, it reaches King Khri
gtsugs lde btsan's reign. (bSod names rtse mo: 1968: folio 315b line 1-
folio 316a line 4.)

Instead of lending support to "Manadeva
Samvat" this chronology completely falsifies it. Among Tibetan
historians, there is a controversy as to when Ral pa cen actually came
to power, in A.D. 814 (The Blue Annals by gZon nu dpal, compiled in
A.D. 1476-78), or in 817 (Bod kyi rgyal tabs, compiled by Gras pa rgyal
mtshan, in A.D.1545), or in A.D.823 (Chos b'yung, compiled by Buston,
in 1322). The short chronology is based on the Buddha's Nirvana in the
year 2133 B.C. bSo nams rtse mo says that he noted down the chronology
from the colophon of the Tibetan translation of Vavaviveka's (ca.450
A.D.) Tarkajvala done by Atisa and Jayasila (Nag lo cha ba). It cannot,
however, be traced in the T'angur, dbu- ma No 3856= Toh (or mDo Edition
XIX 2 K 96/5256/19- 4- 7 on folio 40B7-329B2).

In Tibet,
prior to the adoption of the system based on Kalacakratantra, derived
from the 60- year cycle of Jupiter, there was a calendar named mekha-
gya- tsho (Skt. agni.- ambara- abdhi, i.e., 403). As Levi puts it,


the word is a compound of numeral symbols: me, the fire, expresses 3;
kha, the space, 0; gya- tsho, the lakes, 4; melcha gya tsho signifies
403. Read according to the Indo- Tibetan method, me kha gya tsho
signified 403, and 403 deducted from AD. 1025 would then be A.D. 622.
But I have already more than once pointed out to what extent these
expressions in numerical symbols lend themselves to inversion of
figures. If one rephrases the hypothesis as Kha me gya- tsho one will
read 430 instead of 403. It is the very date I was led to by
astronomical calculation to the year A.D. 595 as the epoch year of the
Thakuri Era (Levi, 1905: II: 154).

Levi refers to one of the
fundamental rules of Indian chronology, ankanam, vamto gatih, according
to which the figures rendered by indicative words are to be counted
backward, but the Tibetan historical tradition began to adopt this
tradition only since mid- seventeenth century. So we have to calculate
the numeral nominals according to the sequence of the given words. It
would have resulted in (A.D. 1025- 304) A.D. 622- the date of founding
of Hijri Era (July 15, 622 A.D.). That date would have upset Levi's own
theory.

Petech's Puppet Theory

The authenticity of
this theory is just as doubtful as Luciano Petech's own "puppet
theory". He claims that Amsuvarma launched this epoch era by placing on
the throne a puppet of his- - - a "Manadeva II." In his own words,


In 576 Amsuvarman, then the man behind the throne, installed a puppet
of his, Manadeva, followed later by Gunakamadeva. Both are mentioned in
Kirkpatrick's vamsavali as the immediate predecessors of sivadeva ...
Then sivadeva was placed on the throne. But in 606 he was deposed or
died, and Amsuvarman began to rule without a puppet king, employing (or
starting) the era of his first protegee Manadeva.... The Manadeva of
576 was the first Buddhist king (Petech, 1961:230).

Why would
an ambitious regent use his puppet's epoch era is none too clear to us.
In the whole Indian sub- continent there is hardly any evidence of a
feudatory under a paramount ruler launching an epoch era.

Nayaraj Pant's Vintage Theory : Mahasamanta Manadeva II


The above theory is as good as Nayaraj Pant's vintage theory that
"Manadeva II," the Imaginary founder of the epoch era, was a powerful
mahasamanta, not a king.

In A.D. 576, when the reigning Kings
were too weak, the chief feudatory (mahasamanta) Manadeva II, took
power in his own hands and also founded an epoch era. This epoch era is
known as Manadeva Samvat Its use continued for 300 years (Pandey and
Pant, 1947:13-14).

The Ranguru's Theory: Epoch Era of Manadeva I


Similarly, the late Hem Raj Pandey thought that the epoch era mentioned
in Sumatitantra was founded by Manadeva I. "The Rajaguru was of the
opinion that this Manadeva for whom he supplied the date from the
manuscript was the Manadeva of Cangu Narayna and that it proved a
Manadeva Era which was the era used by Amsuvarma'--a theory which Kashi
Prasad Jayaswal politely turned down (1936:36).

Distorting
these historical facts, Nayaraj Pant and his school enlist the name of
Hemraj Pandey as the first proposer of the so- called "Manadeva
Samvat". Clearly, he was not proposing a Manadeva II; he was proposing
Manadeva I as the founder of a new epoch era in saka 498/A.D. 576.

The Historian- Laureate's Theory : the Non- existent Rupavarma


Almost similar is the nature of the late Historian-Laureate Kharidar
Baburam Acarya's theory that Manadeva Samvat was founded by Amsuvarma
in the memory of his father, Rupavarma who probably belonged to the
royal line of Vrsadeva! (Acharya, 1949.8). He continued to believe in
this theory in his essay on "Nepal's Relations with China and Tibet,"
where he wrote,

In Nepal, upto A.D. 576, all political power
was controlled by the Abhiras. In this very year Rupavarma took the
support of a Licchavi prince born in that family and suppressed the
Abhiras. Around A.D. 587 Amsuvarma became a ruler succeeding the
Licchavi prince. In A.D. 606 the coronation of Amsuvarma took place,
and he founded a new epoch era beginning from A.D. 576. (Acarya,
1956:8- 9).

Acarya fails to explain why Amsuvarma did not
launch his epoch era from Year 1. We are kept in dark about what
happened during the interval of Year 1 and 28. In an interview he gave
to Professor T.R. Vaidya, the late Dhanavajra Vajracarya, Basudeva
Tripathi and Churamani Bandhu, on April 23, 1970 at his residence,
Acarya says,

In Samvat 528 Amsuvarma came to power. After
that for the convenience of computation, he dropped the century figure
5 and continued with number 28 and so on (Sharma, Vajracarya and
Thakur, 1973:24-25).

Acarya names the Samvat used by Manadeva
and his successors as, Kosanu Samvat and the one used Amsuvarma and his
successors by dropping the century figure "Later Kosanu Samvat". Thus,
in complete defiance of Acharya's views as well as Hemraj Pandey's
view, Nayaraj Pant and his school enlist both Baburam Acharya and
Hemaraj Pandey as the copropounders of the so- called "Manadeva
Samvat". The Pants have been hitherto cashing in on this concocted
historical fraud. These distortions of facts were solely motivated by
the urge to prove all other scholars wrong and himself and his school
alone as right in everything they publish.

Pandit Nayaraj's Theory of "Mandeva Samvat"


While there is such a disappointing confusion among eminent scholars,
NR Pant boldly wrote in the Preface to their edition/transcription of
the Sumatitantra,

Sumatitantrasyopalambhadamsuvarmadibhih
prayuktah samvatsaro manadevasamvatsara iti naipalaka vidvamsah
svicakruh. (Pant et al. 1978:27). As attested in the Sumatitantra, the
Manadeva Samvat used by Amsuvarma and his successors is accepted by all
Nepali scholars. This is only a happy conclusion arrived at by a
mahavidyavaridhi.

While we were working on a facsimile
edition of the Gopalarajavamsavali in 1984, we realized the need of
verifying Manadeva II Samvat for preparing an acceptable chronology of
the Licchavis. (See Vajracarya and Malla, 1985: 235). Vajracarya
already had doubts about a Manadeva II, so that "it is somewhat
inconvenient to call this epoch era as Manadeva Samvat. But we have not
been able to ascertain what happened during the time between saka
samvat 498 and 512" (Vajracarya, 1973: 299-300). Although both the
editors had reservations on this issue, we let it go. In a couple of
later publications as well, we saw no need to check the sources,
particularly the Tibetan, because we took the words of a well-known
Tibetologist and those of N.R. Pant for granted. Now we realize how
wrong both of us were in assuming that there was a "Manadeva II" who
launched an epoch era in A.D. 576! This paper is only a small
prayascitta for this big mistake.

Historical Facts and Sumatitantra's Figures

The British Museum copy of the Sumatitantra has the following lines,


Jato duryodhano raja kalisandhyam pravartate/ Yudhisthiro maharajo
duryodhanastayopi va// Ubhau rajau sahasre dye varsantu sampravarttati/
Nandarajyam satastanca scandraguptastatopare/ Rajyankaroti tenapi
dvatrimsaccadhikam satam/ Raja sudrakadevasca varsasaptabdhi casvinou//
sakaraja tatopascadvasurandhra kritantatha/ Ityate bhasitammahyam jnaya
raja kramena tu// Sesa yutasca krta ambaragni 304 sri manadevbda
pryujyamana etani pinda kali- varsamahuh// (Folio 2b- 3a)

The
above text has been transcribed, translated, and interpreted
differently by different Nepali and foreign historians of Nepal,
depending upon how, for instance, one translates the word,
sampravarttate. Yet the fact remains that not a single of the figures
for the six epoch eras mentioned in the Sumatitantra--(Yudhisthira
2000, Nanda 800, Candragupta 132, sudraka 247, saka 498, and Manadeva
304) matches with the known historical facts. In the first place, it is
not clear from the various translations of the word sampravartte
whether these figures refer to the duration of the rule of a king (as
the text intends?), or of the dynasty (as Petech thought), or of the
use of the epoch year (as the Pants think). If the intention was to
specify the duration of a King's reign or rule (as suggested by
candraguptastatopare// rajyankaroti tenapi dvarimsaccadhikam satam//),
i.e., thereafter Candragupta will rule for 132 years, then it is
clearly a pious fabrication. Jayaswal, who first published this text in
1936, claimed,

It is clear that the author of the chronology
took chief reigns as landmarks, and not always eras. There were
Yudhisthira Era, Nanda, and saka Eras, but there was no Chandragupta
Era, there was no Sudraka Era. There is no trace of an Era of Manadeva
I (Jayaswal, 1936:42).

If the intention of the scribe was to
indicate the duration of an epoch year, implying that saka Samvat will
last up to 498 only, clearly saka Samvat continued to be used in
Licchavi inscriptions from Samvat 512 to 535 by sivadeva and to Samvat
536, side by side with the so- called "Manadeva Era" by Amsuvarma.


As known to modern historians, much of ancient Indian history before
the invasion of Alexander the Great (328-326 B.C.) is mostly a core of
hard facts surrounded by a thick pulp of disputable interpretations of
the extant literary and archeological sources. Mahapadma Nanda, the
founder of the Nanda dynasty, ruled between ca. 362- 320. B.C.
Candragupta Maurya, between ca. 321-298 B.C. The Satavahan dynasty,
founded by King Simuka, ruled between ca. 50 B.C. to 250 A.D. sudraka,
the founder of Andhra Dynasty, ruled between ca. A.D. 350-400. The
Scythian Kusanas or sakas, used at least three different eras. The
first one was founded in 123 B.C., using Macedonian months and Greco-
Chaldean method of date recording. In the first three centuries of
their rule, the sakas used the old era with hundreds omitted. But they
also began to use Indian months and Kharosthi script. The classical
saka era, starting from Tuesday, March 3, A.D. 78, is nothing but the
old saka era, starting from 123 B.C. with 200 omitted, so that the year
1 of Kaniska is year 201 of the old saka era, though the qualification,
salivahana, is attached to it much later. This epoch year itself is a
lokakala, an abridged one for saka 201. Jayaswal, who first published
the Sumatitantra chronology was not too sure of its historical
relevance because,

The year for the commencement of the
Nanda- Rajya was hopelessly wide off the mark. The dates for the
commencement of the Maurya kingdom and the Satavahana kingdom are short
by about 22 years. (Jayaswal, 1936:42)

In Licchavi Nepal,
saka era continued to be used, with the figure for the centuries
omitted, till the beginning of Nepala Samvat. (For early Indian
Chronology and Calendar, See R. Morton Smith, "Ancient Indian
Chronology," Journal of the American Oriental Society, Vol. 77 No. 2
(1957), pp. 276-280; Van Lohuizen de Leeuw, The "Scythian" Period
Leiden: E.J. Brill, 1949; M.N. Saha, "Different Methods of Date
Recording in Ancient and Medieval India, and the Origin of the saka
Era", Journal of the Asiatic Society, Letters Vol. XIX No 1, 1953;
Etienne Lamotte, Histoire du Buddhisme Indien. Des origins a l' Era
saka. Louvain: Institut Orientaliste de l' Universite de Louvain, 1958,
and David Pingree, "A Note on the Calendars Used In Early Indian
Inscriptions." Journal of the American Oriental Society, Vol. 102 No. 2
(1982), pp. 355-359).

From the Beginning of the Kali to the Composition of the Text


Although it is a tantra based on mathematical astronomy, the framework
of the Sumatitantra is puranic. M.R. Pant's excavation of a brahmarisi,
named Sumati from among the mourners lined up at the deathbed of
Bhismapitamaha in the Mahabharata War is laudable on its own, but its
use to defend the puranic veneer of the text is rather unfortunate. The
puranic bias of the text is evident, not only from the opening verses,
but also from its history, cosmogony, and geography.

The main
function of the chronology planted in the Sumatitantra appears to be to
add up to the total years lapsed between the beginning of the Kali Yuga
(i.e., Friday 17/18 February 3101 B.C.) and the imagined date of
composition of the text in question. The total comes out to be 3981
years, mainly to facilitate the computation of the ahargana. The
historical use of these figures is virtually nil because these figures
are based on the Puranic lists of various dynasties of the Kali Yuga.
Despite Pargiter's (1913) admirable patience in attempting to
reconstruct an acceptable chronology out of these Puranic lists, there
is a virtual chaos among various puranas and upapuranas. For example,
Matsya Purana says that Mahapadma Nanda ruled for 88 years whereas Vayu
Purana says that he ruled for 28 years. Pargiter, (1922:287) assigns 80
years to the Nandas, and Sumatitantra claims that the Nandas ruled for
800 years. No sensible historian can take these figures too seriously.
However, N.R. Pant was so credulous about the authenticity of these
figures that he quotes the Sumatitantra to give the date of Candragupta
Maurya ! (See his Hindu Siddhanta- Jyautisa ra Greek Siddhanta
Jyautisako Tulana (1990: 10). This is only an example of N.R. Pant's
sanguinity and complete uncritical faith in his sources. It is not for
nothing that R.G. Collingwood wrote in his classic, The Idea of
History, saying, "In so far as an historian accepts the testimony of an
authority and treats it as historical fact, he obviously forfeits the
name of historian; but we have no other name by which to call him."
(1941:252).

Nepala Samvat

Nepala samvat was
founded, not in the imaginary year 304, nor in saka 802, as implied by
Sumatitantra, but on Karttika sukla Pratipad of current Samvat 303,
i.e., Karttkadi current saka 803, i.e., on Tuesday, October 20, 879
A.D. On that day, the Moon was in the Anuradha Naksatra for 20 hours
and 23 minutes after the mean sunrise in Kathmandu, and sovana Yoga
continued between lunation 1481-1852. After the mean sunrise in
Kathmandu, the sukla Pratipad tithi lasted for 22 hours 8 minutes. This
date is 7 months later than the beginning of Caitradi expired saka 801.
Levi (1905, II:181-183) speculated that it was also a lokakala, with
800 left out because the Nepalese thought that the number 8 was
inauspicious.

The verses in the Sumatitantra, giving
different fictitious numbers for different rulers, or eras, or the
duration of a dynasty must have been an elaboration by a Nepali scribe,
with a smattering of Jyautisavidya, upon the original verse which
probably gave a plain formula to arrive at the time duration between
the beginning of the Kali and the composition of the given Karana, not
dissimilar to Chapter I Verse 23 of new Surya Siddhanta which
explicitly enunciates: atah kalam prasankhyaya sarikhyamekatra pindyet,
i.e., add up all the past years between the present and the beginning
of the Kali Yuga. At any rate, the figure 802, to be added or deducted
from Nepala Samvat to get an integer with saka Samvat, is only a round
figure for 801 year 7 months. This was necessary for all pancanga-
makers as well as for astronomers--the saka era being the standard era
in use among astronomers in India and Nepal.

A Likely Copyist: An Hypothesis


On the basis of a preliminary digital scanning of the manuscripts
copied by Jayasihamalla Varman and the British Museum copy of the
Sumatitantra, one is tempted to believe that these figures were added
by the copyist Jayasihamalla Varman of Nhola Vihara, in Patan because
he was also the author of Sumati- karana, the transcriber of
Khandakhadyaka (Cat III Vi 52, dated NS 470 Bhadrapada 2-3),
Brhajjataka (Cat. Vi 262, dated NS 471 savana sudi 4/5),
Bhojadevasamgraha (dated NS 472 Karttika Krsna 5/6), and probably also
of Jyotirajakarana in the National Library (No 699 Vi 30, dated saka
1304/ A.D. 1382, copied by a later hand in NS 543 Bhadra sukla 10/A.D.
1423.). Assuming such high-flown virudha as bhagnarajasthapanartha//
lubdharajagajankusa/ saranagatavajrapanjara he also copied some
literary texts such as Hariscandropakhyana, Mahiravanavadha, and
Dharmagupta's Ramankanatika.

The Equation of Kanyaadvipa with Nepalamandala


The late Pandit Nayaraj believed that the author of the Sumatitantra
gave the latitude of his country as 27 degrees and its longitude as 1
ghatika and 58 pala (47.2 minutes East of the Prime Meridian). If N.R.
Pant was right then he was so only if Kanyadvipa or Kaumaridvipa,
Kanyakhanda or Kanyakhyah was a mythical name for Nepal. However,
nowhere ha the text the name of the country Nepala is mentioned. On the
contrary, in a chapter on triprasnadhikara (Time, Place, and Direction)
Sumati says that of the nine divisions of Bharatavarsa,


Kanyadvipa lies to the south of Mount Meru and on its southern end lies
the Malaya mountain; there flows the fiver Lavanasru, i.e., salty
waters. It is ruled by King Bhujagasana. At its foot lies Lanka, the
home of demons, with walls and archways made of gold, studded with
various wonderful metals. (Folio 48b verses 6-7 in the British Museum
copy) Vatesvara (b. A.D. 880) writes in his Siddhanta, Lanka (then
northwards) Kumari, then Kanci, Manata, Asvetupuri, then northwards,
the sveta mountain, thereafter Vatsyagulma, the city of Avanti, then
Gargarata, Asramapattana, Malvanagara, Pattasiva, Rohitaka,
Sthanvisvara, the Himalayan Mountains and lastly Meru, ... (these are
situated on the, prime meridian). For these places correction for
longitude is not needed. (VS 1.8. 1-2 in Kripa Shankar Shukla's
translation).

sripati, (ca. A.D. 999) in his Siddhanta-sekara (II:995-996) lists the following places as lying on the prime meridian,


Lanka, Kumari, Kanci, Panata, Sitadri, sadasya, Vatsagulma, Mahismati,
Ujjayini, Pattasiva, Gargarata, Rohita, Sthanisvara, Sitagiri, and
Sumeru.

Do these geographical descriptions fit Nepal?
Certainly not, nor is Nepalamandala ever considered as the Prime
Meridian in Indian astronomy.

What N.R. Pant had done with
the Stumati's text, in the name of emendation, is an indefensible act
of convenient rewriting: N.R. Pant had rewritten the text to suit his
theory,

Kanyakhandaikabhagasya dhrtavarnaaasramasthiteh/
Nepaladronyabhikhyasya desasyaasya visesatah// (Pant et al. 1978:14;
1987:128).

For the following in the Sumati's text where there is no mention whatsoever of the Nepal Valley :

Visesena pravaksyami kanyadvipasya niscayam.


The Sanskrit-Nepali Comprehensive Dictionary published by Mahendra
Sanskrit University, defines Kumarikakhandam (based on the
Skandapurana) as the terrestrial division which was given as a share to
the daughter of King satasrnga, after having divided the earth among
eight sons of his (Pandey, 2000:306).

No known text or
dictionary of place- names, puranic or otherwise, calls Nepal the
Kanyadvipa or Kaumaridvipa, Kumarikhanda or Kumarikhyah. According to
Apte's Practical Sanskrit-English Dictionary, Kumarikah is the name of
the southern extremity of the Indian peninsula, cf., the modern name
Cape of Comorin (1957: 583).

If this is so, then the whole
point of N.R. Pant's argument that Kanyadvipa, Kumaridvipa, Kanyakhanda
is the Nepal Valley and that its latitude is 27 degrees North of
Equator and 1.58 ghatika East of Ujjayini stands as unfounded.


Why would acarya Sumati call his country Kanyadvipa or Kaumaridvipa, a
son of euphemism for a country when Nepal was an already known
placename by the time of Samudragupta (A.D. 335-374) in India, or
Vasantadeva A.D. 504-530) in Nepal, not to mention the references in
the classical Indian texts of great antiquity such as Krsna- Dvaipayana
Vyasa's Mahabharata, Kautalya's Arthasastra, or Bharata' Natyasastra or
Varahamihira's Brhatsamhita. Nepala, certainly was not an unknown
place- name at that time, nor was it a taboo word among the astronomers
in the sub- continent (Malla, 1984b:63- 69). If by any chance,
Kanyadvipa or Kaumaridvipa is Nepal, why cannot, for instance, the
eastern region of Tibet be Kanyadvipa because a 7th century Chinese
source, Fa-iouen- tchou- lin, the famous Encyclopedia of Buddism
compiled and completed by Tao- cheu, in 668 A.D., notes the following
about Strirajya,.

Lately the orders of the (Chinese) Empire
faded out of the Kingdom (of Nepal), and spread elsewhere in distant
lands. Now, it is dependent on Tu- fan (Tibet). On the east, the
Kingdom of Women is adjacent to Tu- fan." (Levi, 1900/1987: 60-61)


The Latitude and Longitude are of Kanyadvipa, not of Nepalamandala
There are scores of cities which lie 27 degrees North of Equator or
47.2 minutes East of a given Prime Meridian. In calculating the
longitude of so-called Kanyadvipa, the NR Pant resorts to an equally
odd expediency : he mixes up the figures from two different
authorities, Varahamihira (A.D. 550) and Laksmipati (A.D. 1758-1831) to
arrive at the desired longitude of 47.2 minutes East of Avanti.
Varahamihira's longitude for Varanasi, as computed from Alexandria, is
9 ghatika, and for Avanti/Ujjaini, 7 ghattika and 20 pala (9.00-7.20=
1.40=100 pala.). To these figures, N.R. Pant amputes Laksmipati's
longitude for the Nepal Valley (19 pala for the Nepal Valley) to arrive
at 2 ghatika or 120 pala. All this quibbling with figures to suit one's
theory was totally unexpected of 'a renowned mathematician,' considered
by some of his admirers as "the Socrates of Nepal." Not dissimilar is
his misleading calculation of the colophon of Samvat 301, where he
added 1 to Samvat 301 and made it Samvat 302 by a stroke of the pen for
Vaisakha Saptami in a Karttikadi year (Pant et al. 1987:19; 125-135).

Sumati was not a Nepali


Nayaraj Pant claimed that Acarya Sumati composed his tantra in Nepal.
In the wake of the nationalistic fervor of the 1960s, similar
"nationalist" claims were also made by Pandit Buddhisagar Parajuli in
his introduction to Brhatsucipatram, Vol. I (1960: gha) He wrote that
both Gavastidipa and Tamradvipa were in Kasthamandapa because the
cara-prana (addition or subtraction of ascensional differences, between
the meridian and a given latitude) for Mesadi 6 rasis in the
Sumatitantra are identical with those given by astrologers in Nepal. He
also claimed that the Sausruti- samhita- Sahottaratantra of Samvat 301
was copied during the reign of Manadeva I (Parajuli, 1966:19-21).
Similarly, a year later, Lamsal (1967), an acarya in Indian astronomy,
wrote that the Sumatitantra is the pride of Nepal and that it was
written at the time of Manadeva I and that the Mesadi cara-prana i.e.,
ascensional differences or corrections for Mesadi 6 rasis given in the
text are identical with those used by Nepalese astrologers. However,
Lamsal carefully compared the astronomical constants given in the
Sumatitantra with those given in Varahamihira and other authorities,
including the modern ones, mainly based on Sengupta's Introduction to
Burgess' translation of the Surya Sihhhanta (1935/1997:VI-LI). He noted
that the corrupt copy we have in the National Archives cannot be the
original one as it is so full of linguistic lapses. He also says that
its astronomical parameters are based on ancient Surya Siddhanta, and
that the text is of uncertain age as Sumati has ignored the precession
of equinoxes. Notwithstanding his wild claim that the original text was
written in Nepal after the founding of Manadeva I's epoch era, Lamsal's
paper is a simple concise introduction available in Nepali to the
theoretical base of Sumati. In less than 5 pages he succeeds in
conveying his views on the text for which Nayaraj Pant took 60 odd
pages! If that was the intellectual situation in the 1960s in Nepal,
what could one expect of the copyists of the Bhasa Vamsavalis in the
19th century or of the Sumatitantra- scribe in the 14th Century
Nepalamandala?

We have given these details also because
modern- minded specialists in Indian Astronomy are equally misled by
false clues based on hearsay. For example, Kripa sankar sukla, in his
edition of the Surya Siddhanta with Paramesvara's commentary, contends
that

The old Surya Siddhanta continued to be studied in
certain parts of India in some form or other till the end of the tenth
century. About 800 A.D. an astronomer named Sumati of Nepal wrote two
works on astronomy one entitled Sumati-tantra and the other entitled
Sumati- karana. In one of the opening verses of the first- mentioned
work, Sumati writes: "This work, called Sumatitantra, has been
extracted from the Surya Siddhanta like clarified butter from milk." It
shows that Sumati based his Tantra on the Surya Siddhanta. The
astronomical constants used by Sumati agree with those ascribed by
Varahamihira to the Surya Siddhanta. It, therefore, seems that Sumati
based his work on the same Surya Siddhanta as was available to
Varahamihira. The other work, as the name implies, is a calendarical
work. The works of Sumati show that in the end of eighth century A.D.
the old Surya Siddhanta was considered by the astronomers of Nepal to
be an important work on astronomy and its elements were used by them in
the construction of the Hindu calendar. Sumati received wide publicity
and his works traveled to the south as far as Travancore.
Shankaranarayana, who belonged to Quilon in Travancore, in his
commentary on Laghubhaskariya of Bhaskara I mentions the name of Sumati
and quotes a verse from his work. This commentary of Shankarnarayana,
it may be mentioned, was written only 69 years after the composition of
the Sumatitantra. (K.S. Shukla, 1957:27-28).

Yet another
specialist in Indian sciences, H.J.J. Winter of Exeter University in
the UK has contributed a chapter to A.L. Basham's A Cultural History of
India (Delhi: Oxford University Press, 1975). The following paragraph
by him is obviously based on K.S. Shukla, 1957,

According to
Sumati (A.D. 800) whose work was known both in Nepal and in Kerala, and
who wrote his Sumatitantra and Sumati-karana on the basis of an earlier
version of the Surya siddhanta, it provided the essential elements used
by Nepali astronomers in their construction of the Hindu calendar.
Evolving during the period between A.D. 628 and 966, the later version
gained greatly in popularity, especially in the twelfth century when
Bhaskara II quoted from it and Mallikarjuna Surii wrote commentaries on
it, first in Telegu then in Sanskrit. (Winter, 1975:153.)


While this quotation ably summarizes sukla, neither of them seems to
have really gone critically to the texts by Sumati. The date given by
both for Sumati, i.e., A.D. 800 is based on the misreading of the
colophon of the British Museum copy. Bendall (1902:193-194), not only
misread vasu randhra krta as vasu candra krta, i.e., 418 for 498, so
that saka 418+78--A.D. 496+Manadeva Samvat 304--A.D. 800. He had also
misinterpreted the title of the text as Sumata Mahatantra.

Latadeva's Surya Siddhanta


The Surya Siddhanta that forms the theoretical basis of Sumati is the
ardharatrika paksa of Aryabhata I as it was modified by his immediate
pupil Latadeva in the light of Aryabhata's ardharatrika system. In his
Indika, Al-Beruni (A.D. 1030) says that Latadeva was the author of the
Surya Siddhanta (Sachau: Vol. I: 153).

Sengupta, too,
believes that, "the old Surya Siddhanta was made up- to-date by
replacing the old constants in it by new ones from Aryabhata's midnight
system.... The date of original Surya Siddhanta becomes 384 A. D. It
came from the asura or Babylonian sources. (Sengupta, 1935, p.xl).
Nearly all of its astronomical constants are shared by Brahmagupta's
Khandakhadya, I,1, 8- 13; II, 1-5, and Bhaskara I's Mahabhaskariya (See
Kuppana Shastri's edition, Madras, 1957). The epoch of this
ardharatrika version of Surya Siddhanta is midnight of 20/21 March 505.
However, in Chapter IX of Varahamihira's Pan casiddhantika, there is an
evidence of an earlier Surya Siddhanta using noon epoch, and slightly
different parameters for the mean motion of the Sun, the Moon, lunar
apogee and ascending node. Neugebaur and Pingree think that it is the
work of Latadeva- -- the sarvasiddhanta guru, i.e., the teacher of all
scientific astronomy (O. Neugebaur and D. Pingree, 1970:13).

Sumatitantra: A Contaminated Text


The Sumatitantra, in the form we have it today, cannot be the original
text. Otherwise, how can "Manadeva Samvat" 304 or A.D. 880 be mentioned
in a text that is nearly three hundred years old in origin? Its
language betrays the fact that it has already passed through the hands
of generations of not- too-learned scribes and copyists. It is not an
unusual process for anonymous Sanskrit texts to become an inclusive or
"composite text" with contaminated deposits at different layers. The
best example is the present Surya siddhanta which "became a composite
growth from about 400 A.D. to 725 A.D. from the evidence of its star
tables" (Sengupta, 1935: xxix). It has been subject to correction,
emendation, and modification from time to time, and the present Surya
Siddhanta is the latest redaction of the work which assumed the present
final shape and size between A.D. 628- 966. In fact, the earliest date
given for the text is A.D. 285. The dates of its three substantive
revisions, as it were, are A.D. 285, 500 and 570 (Saha and Lahiri,
1992/1995).

Petech (1984:12 footnote No. 3), however, claims
that the Sumatitantra, in its main portion, (was) "compiled not after
850 A.D." He does not explain what he meant by "the main portion" from
among its five chapters and a sixth incomplete one. We do not know how
he had come to that conclusion.

Can Sumatitantra be Dated?

Is it, then, possible to date the Sumatitantra on any scientific basis?


The late Pandit Nayarij Pant believed that it can be dated on the basis
of whether or not it gives a rate for the precession of equinoxes
(ayanamsa). The Earth is not a perfect sphere, nor is its axis a
perndicular one. A continuous receding of the sidereal Zodiac, at the
rate of 1 degree in 72 years, is caused by the gravitational pull of
the Sun and the Moon on the central bulge of the Earth with a declined
axis. It means that if the Vernal Equinox in 1997 is on March 21st,
seventy- two years later it will be on March 20th N.R. Pant claims,
"how much precession is given in a text, or whether it is mentioned or
not, enables us to determine the age of the text" (Pant et al,, 1978b:
137). Or, "There is no mention of the precession of equinoxes in the
Sumatitantra. Therefore, it was composed in the Licchavi age before the
knowledge of the precession of equinoxes became common among the
astronomers" (Pant et al., 1983:19).

Some ancient astronomers
in India, including Varahamihira, believed that the sidereal Zodiac was
moveable and that its movement was oscillatory and that such a movement
was also ominous and inauspicious.

However, no ancient Indian
astronomer before Munjala (A.D. 932), makes an explicit mention of the
rate, nor of the concept of precession as such. Though the original
work containing his views on the topic is lost there is only a
reference to it in Bhaskara II's Vasanabhasya - - a self-commentary on
Siddhanta- siromani (A.D. 1150). In that text, he writes that
precession is what was known as krantipata (i.e., the intersection of
the ecliptic and the equinoctial circles) in earlier works. He says
that the earlier acaryas did not mention it because it was too
negligible during their time. (Commentary on the verses 17, 18 and 19
of SS in the Golabandhadhikara). Similarly,

From a stanza of
Visnucandra (A.D. 578), quoted by Prthudakasvami (A.D.864) in his
commentary on the Brahmasputasiddhanta, it appears that this subject
was dealt with in the old Surya Siddhanta. (p. 38). According to the
(new) Surya Siddhanta, the precessional motion is likened to liberation
or oscillation of the equinoxes about a fixed point. According to this
theory, the equinoxes, like the pendulum, at first move eastward, reach
the maximum amplitude and then move westward. The maximum eastward
deviation is 27 degrees whence the annual rate works out to 54 seconds
as against the modern value of 50.25 seconds. (Jaggi, 1990:75-76).


David Pingree, one of the foremost Western authorities on Indian
Mathematical Astronomy, published a terse treatment on the topic of
ayanamsa with the title "Precession and Trepidation in Indian Astronomy
before 1200 A.D," in the Journal for the History of Astronomy, Vol 3
(1972), pp. 27-35. He shows that there were three different accounts of
it in the astronomical works before A.D. 1200: 1. trepidation with 27"
x 2= 54" seconds, 2. trepidation with an amplitude of 59.9" seconds,
and 3. the third account gives simple precession.

The Year of Zero Degree Precession and its Annual Rate


There is no unanimity of scholarly opinions on the exact date of zero
degree precession, nor on its annum rate. The initial year of zero
degree precession is considered to be A.D. 285 by the Calendar Reform
Committee of India (1955). Cyril Pagan, a Western advocate of the
Sidereal Zodiac as against Western Tropical Zodiac, believed that the
Zero degree precession coincided with A.D. 213. K.S. Krishnamurti, an
Indian specialist in Oriental Astronomy, thought that in the year 291
A.D., the precession of the equinoxes was Zero. The current Sarya
Siddhanta gives A.D. 499 as the year of zero degree trepidation;
Munjala in his Laghumanasa gives A.D.527 as the year for zero degree
precession, Bhoja (A.D. 1042) gave A.D. 522, Damodara, A.D. 420, and
Pillai, A.D. 533.

Not only that there is such a diversity of
opinions on the initial year, there is also no uniformity of opinions
on the rate of precession. Munjala thought that it was 59.9" seconds or
nearly 1 minute of the arc whereas the modern Surya Siddhanta gives an
oscillation rate of 54" seconds, i.e., 27" eastwards and then 27"
westward. Modern astronomers since Newton give 50.25" seconds as the
annual rate of precession.

We cannot, therefore, date an
astronomical text simply on the basis of the presence or absence of a
reference to the precession of equinoxes.

The Incidence of the Word ayanamsa in Sumatitantra


Pandit Nayaraj Pant believed that the Sumatitantra was written some
time during the three centuries between A.D. 576-879, between the
so-called "Manadeva Samvat" and Nepala Samvat because Sumati does not
mention the precession or its rate explicitly anywhere in the text. The
first mention of ayana-calana (the shifting of equinoxes) occurs in
Vatesvara-siddhanta (A.D.904) but he says that the astronomer who knows
his spherical trigonometry should calculate his own precession rate.
(VS II:25). Thus, in the British Museum copy, on several folios of the
Sumatitantra, there is a mention of the word, ayanamsa. In the Chapter
on the True Motions of the Planets, there occur the following lines

Desantaranadi yojana 160 bunajya yojana 9600 bhaga 28812 viksepakranti ayanansanca tena karayet (folio 24b)

In the Chapter on Lunar Eclipses, we have the following lines

aksajyakarme guna 1561 bhaga 3438//ayanamsncakarme guna 1398 bhaga 3438// (folio 85b)

In the Chapter on Solar eclipse, we have the following lines:


Sparsa madhyanta purvena ayanamsanca sadhayet/ Aksasutrantathaivanca
parilekhani karayet// (folio 107b) grasadi madhyanta divakaresu/
kramena rasitryasamyutesu sadabhuttaresatkhapareca yamyam
krantijyamaniya yathaiva purvvam. bhanoyanamsammiti laksayitva/
sasparsa madhya grahananta kale// pratyaya nadimasu pinda kuryat
sumnyakhatirthena vibhakta labdham (folio114b) grasadi madhyanta
kramayanamsa//aksayanamsau samadik yutamtau yojitam (folio 124b)


The longitude of a planet for any given moment can be calculated by
using a standard pancanga. Nirayana (sidereal) longitude plus ayanamsa
gives the sayana (tropical) longitude of the planet. However, the moot
question here is: how come that the word is undeniably there in the
text of Sumatitantra if there were no related concept embedded in it?
Can we have the word without the related concept in any discourse? No,
certainly, not; unless the word is, like the words in nonsense verse or
Alice in the Wonderland, semantically vacuous.

Do We have the Whole of Sumatitantra Text?


Both the copies of the Sumatitantra end abruptly; the last 13/14 lines
in the British Museum copy are missing from the copy in the National
Archives, and in their stead there is a stray folio with only 7 lines,
and it is numbered 162a (?) which follows folio 143b. This indicates
the plain fact that there are at least 20 folios misplaced in the
National Archives. However, it has some tables (folios162a-166b),
giving the gigraha (epicycle) figures for each degree of the
Planets-Mars, Mercury, Jupiter, Venus, and Saturn. On folio 165b we
have a table for Krantikhanda (Sine for Declination), Lagna (Rising
Point of the Ecliptic), and Trigavadi (Three Transformations) for each
of the 90 degrees of the quadrant. There are also 3 additional folios
without page numbers, which give the summary of mean motions etc. as
enunciated in the main text. Then comes the colophon. It dates the
text, copied for the second time, on Nepala Samvat 495 Pausa sukla 13.
All the above Tables are missing in the British Museum copy. In that
copy, the last 18 lines end abruptly with

Evante kathiyinatra saptapatalakani ca/yakramova ...


after which comes all of a sudden the colophon. The copyist seemed to
have thought that the following descriptive section on the saptapatala
(seven circles of the Hell) is not of day-to-day relevance. The first
colophon has the following date: //svadadricatvah saha vatsareva masa
bhava pausa sita navammvam i.e., NS 476 Pausa sukla 9....

In
the National Archives copse, the supplementary folios 11a,-b, 12a,-b,
13a- and b appear to represent what are folios 121a,-b, 122a-b,
123a,-b, 124a and ban the British Museum copy. In both copies these
folios have little direct connection with the main running text, but
they try to show the saka Samvat calculations in the context of Nepala
Samvat. As such, these portions appear to be unsuccessful attempts at
textual modification, amputation or interpolation.

Although
all the topics enumerated at the beginning of the text in the Chapter I
on the Mean Motions of the Planets are covered by the copies we have,
we do not really know whether the copies we have contain the whole of
the Sumatitantra. Above all, there is no star--table or a Chapter on
Naksatras (asterisms), had there been one it would have made the dating
of the text a less serious problem. However, the Sumatitantra (folio
20a-b) gives the mandocamsa i.e., planetary longitudes for the higher
apses in degrees, pertaining to the Equation of the Centre. These
astronomical parameters are identical with those of Mahabhaskariya,
VII:25-28a and Khandakhadyaka. The patamsa or (longitude of ascending
nodes of the Planets in degrees since the beginning of the Kali) given
in the Sumatitantra (Folio 48a) are identical with those given in
Mahabhaskariya (VII. 9-10). Similarly, Viksepa (the Celestial Latitude
of the Planets) is given in folio 48a. All these astronomical constants
are of about A.D. 505.

The following verse cited by
sankaranarayana (A.D. 825-900) in his Vivarana on Laghubhaskariya (IV:
15-16) from Acarya Sumati, does not occur in the chapters on eclipses