Asoka and the Buddha-Relics
By T.W. Rhys Davids.
Journal of the Royal Asiatic Society
1901, pp. 397-410
Our oldest authority, the Maha-parinibbana
suttanta, which can be dated approximately in the
fifth century B.C., (1) states that after the
cremation of the Buddha's body at Kusinara, the
fragments that remained were divided into eight
portions. These eight portions were allotted as
1. To Ajatasattu, king of Magadha.
2. To the Licchavis of Vesali.
3. To the Sakyas of Kapilavastu.
4. To the Bulis of Allakappa.
5. To the Koliyas of Ramagama.
6. To the brahmin of Vethadipa.
7. To the Mallas of Pava.
8. To the Mallas of Kusinara.
1. That is substantially, as to not only ideas, but
words. There was dotting of i's and crossing of
t's afterwards. It was naturally when they came to
write these documents that the regulation of
orthography and dialect arose. At the time when
the Suttanta was first put together out of older
material, it was arranged for recitation, not for
reading, and writing was used only for notes. See
the Introduction to my " Dialogues of the Buddha,"
Drona, the brahmin who made the division,
received the vessel in which the body had been
cremated. And the Moriyas of Pipphalivana, whose
embassy claiming a share of the relics only arrived
after the division had been made, received the ashes
of the funeral pyre.
Of the above, all except the Sakyas and the two
brahmins based their claim to a share on the fact
that they also, like the deceased teacher, were
Kshatriyas. The brahmin of Vethadipa claimed his
because he was a brahmin; and the Sakyas claimed
theirs on the ground of their relationship. all ten
promised to put up a cairn over their portion, and to
establish a festival in its honour.
Of these ten cairns, or stupas, only one has been
discovered--that of the Sakyas. The careful
excavation of Mr. Peppe makes it certain that this
stupa had never been opened until he opened it. The
inscription on the casket states that " This deposit
of the remains of the Exalted One is that of the
Sakyas, the brethren of the Illustrious One." It
behoves those who would maintain that it is not, to
advance some explanation of the facts showing how
they are consistent with any other theory. We are
bound in these matters to accept, as a working
hypothesis, the most reasonable of various
possibilities. The hypothesis of forgery is in this
case simply unthinkable. And we are fairly entitled
to ask: "If this stupa and these remains are not what
they purport to be, then what are they?" As it stands
the inscription, short as it is, is worded in just
the manner most consistent with the details given in
the Suttanta. And it advances the very same claim (to
relationship) which the Sakyas alone are stated in
the Suttanta to have advanced. It does not throw much
light on the question to attribute these coincidences
to mere chance, and so far no one has ventured to put
forward any explanation except the simple one that
the stupa is the Sakya tope.
Though the sceptics-only sceptics, no doubt,
because they think it is too good to be true--have
not been able to advance any other explanation, they
might have brought
forward an objection which has so far escaped notice.
It is alleged, namely, in quite a number of Indian
books, that Asoka broke open all the eight stupas
except one, and took the relics away. This is a
remarkable statement. That the great Buddhist emperor
should have done this is just as unlikely as that his
counterpart, Constantine the Great, should have
rifled, even with the best intentions, the tombs most
sacred in the eyes of Christians. The legend
deserves, therefore, investigation, quite apart from
its reference to the Sakya tope. And in looking
further into the matter I have come across some
curious points which will probably be interesting to
the readers of this Journal.
The legend might be given in my own words,
filling out the older versions of it by details drawn
from the later ones. We might thus obtain an easy
narrative, with literary unity and logical sequence.
But we should at the same time lose all historical
accuracy. We should only have a new version--one that
had not been current anywhere, at any time, among
Buddhists in India. The only right method is to
adhere strictly to the historical sequence, taking
each account in order of time, and letting it speak
Now it is curious that there is no mention of the
breaking open of stupas in any one of the twenty-nine
canonical Buddhist writings, though they include
documents of all ages from the time of the Buddha
down to the time of Asoka. Nor, with one doubtful
exception, is such an act referred to in any book
which is good evidence for the time before Asoka. But
in the canonical books there is frequent reference to
the man who breaks up the Order, the schismatic, the
sangha-bhedako. And in the passages in later books,
which enlarge on this thesis, we find an
addition--side by side with the sangha-bhedako is
mentioned the stupa-bhedako, the man who breaks open
the stupas. The oldest of the passages is the
exception referred to. It is in the Mahavastu,
certainly the oldest Buddhist Sanskrit text as yet
edited, and most probably in its oldest portions
older than Asoka. Whether this isolated verse
belongs to the oldest portions of the work is
doubtful. It says (i, 101):
Sanghan ca te na bhindanti na ca te stupa-bhedaka
Na te Tathagate cittam dusayanti kathancana.
We find these gentlemen, therefore--the violators of
tombs, tomb-riflers--first mentioned in a way that
may or may not, and probably does not, refer to
Asoka. In the same connection, that is with the
schismatics, they are also mentioned in the Netti
Pakarana, p. 93. The editor of this work, Professor
Edmond Hardy, dates it about, or shortly after, the
beginning of our era. And he was the first to call
attention to the mention in these passages of the
`tomb-violators' as a test of age.
The next passage will seem more to the point,
inasmuch as it mentions both Asoka and the Eight
Topes. It is in the Asokavadana, a long legend, or
historical romance, about Asoka and his doings,
included in the collection of stories called the
Divyavadana. These stories are by different authors,
and of different dates. The particular one in
question mentions kings of the Sunga dynasty, and
cannot therefore be much older than the Christian
era.(1) The passage is printed at p. 380 of Professor
Cowell and Mr. Neil's edition. The paragraph is
unfortunately very corrupt and obscure; but the sense
of those clauses most important for our present
purpose is clear enough. It begins, in strange
fashion, to say, a propos of nothing:-
"Then the King [Asoka], saying, 'I will
distribute the relics of the Exalted One,' marched
with an armed force in fourfold array, opened the
Drona Stupa put up by Ajatasattu, and took the
There must be something wrong here. Ajatasattu's
stupa was at Rajagaha, a few miles from Asoka's
capital. The Drona Stupa, the one put up over the
vessel, was also quite
close by.(1) Whichever is the one referred to, it was
easily accessible, and the time given was one of
profound peace. Asoka's object in distributing the
relics, in the countless stupas he himself was about
to build, is represented as being highly approved of
by the leaders of the Buddhist order. What, then, was
the mighty force to do?
Then the expression Drona Stupa is remarkable.
What is probably meant is a stupa over the bushel
(drona) of fragments (from the pyre) supposed to have
been Ajatasattu's share. But it is extremely forced
to call this a Drona Stupa; and Ajatasattu's stupa is
nowhere else so called. Burnouf thinks(2) this is
probably a confusion between the name of the measure
and the name of the brahmin, Drona, who made the
division. The story goes on:
"Having given back the relics, putting them
distributively in the place [or the places] whence
they had been taken, he restored the stupa. He did
the same to the second, and so on till he had taken
the seventh bushel [drona];(3) and restoring the
stupas, he then went on to Ramagama."
Here again the story-teller must have
misunderstood some phrase in the tradition (probably
in some Prakrit or other) which he is reproducing.
Asoka did not want to get these relics in order to
put them back into the place, or places, they had
come from. He wanted, according to the Divya- vadana
itself, to put them in his own stupas. We shall see
below a possible explanation. The story goes on:-
"Then the king was led down by the Nagas into
their abode, and was given to understand that they
would pay worship [puja] to it [that is, to the stupa
or the portion of relics] there. As soon as that had
been grasped by the king, then the king was led up
again by the Nagas from their abode."
1. See Yuan Thsang, chap. vii; Beal, ii, 65.
2. "Introduction, etc., p. 372.
3. Bhaktimato is omitted. The discussion of its
meaning, irrelevant to the question in hand, is
here unnecessary. It is of value for the very
important history of bhakti in India.
Their abode, of course, was under the sacred pool
at Rama- gama, the stupa being on the land above.
After stating how Asoka then built 84,000 stupas (in
one day!) and distributed the relics among them, the
episode closes with the statement that this was the
reason why his name was changed from Candasoka to
Dharmasoka. Burnouf adds to the confusion with which
this part of the story is told through translating
(throughout) dharmarajika by 'edicts of the law.' It
evidently is an epithet of the stupas. Can we gather
from this any hint as to a possible origin of this
There is namely a very ancient traditional
statistical state- ment-so ancient that it is already
found in the Thera Gatha. (verse 1022) among the
verses attributed to Ananda--that the number of the
sections of the Dhamma (here meaning apparently the
Four Nikayas) was 84,000, of which 82, 000 were
attributed to the master and 2,000 to a disciple.
Dvasiti Buddhato ganhim dve sahassani bhikkhuto
Caturasiti sahassani ye 'me dhamma pavattino.(1)
Could it have happened that after the knowledge of
the real contents of the Asoka Edicts had passed
away, and only the memory of such edicts having been
published remained alive, they were supposed to
contain or to record the 84,000 traditional sections
of the Dhamma? And then that by some confusion, such
as that made by Burnouf, between epithets applicable
equally to stupas and 'edicts of the law,' the edicts
grew into stupas? We cannot tell without other and
earlier documents. But this we know, that the
funniest mistakes have occurred through the telling
in one dialect of traditions received in another; and
that the oldest form of the legend of Asoka's stupas
is in so late a work that such a transformation had
had ample time in which to be brought gradually
Such a solution of the mystery how this amazing
proposition could have become matter of belief is
1. Quoted Sumangala, i, 24.
by our next authority, the Dipavamsa (vi, 94-vii,
18), which says distinctly that the number of Asoka's
buildings was determined by the number of the
sections of the Dhamma. But the legend here is quite
different. There is no mention of breaking open the
eight old stupas. The 84,000 viharas --they are no
longer stupas--are not built in one day; they take
three years to build. It is the dedication festival
of each of them that takes place on the same day, and
on that day Asoka sees them all at once, and the
festivals being celebrated at each. This was the form
of the story as believed at Anuradhapura in the early
part of the fourth century A.D.
The next book, in point of date, which mentions
Asoka in connection with the eight original stupas is
Fa Hian (ch. xxiii). The passage runs, in Legge's
translation, as follows:-
"When King Asoka came forth into the world he
wished to destroy the Eight Topes, and to build
instead of them 84,000 topes. After he had thrown
down the seven others he wished next to destroy this
tope (at Ramagama). But then the dragon (1) showed
itself, and took the king into his palace. And when
he had seen all the things provided for offerings, it
said to him: 'If you are able with your offerings to
exceed these, you can destroy the tope, and take it
(2) all away. I will not contend with you.' The king,
knowing that such offerings were not to be had
anywhere in the world, thereupon returned.
"Afterwards the ground all about became overgrown
with vegetation; and there was nobody to sweep and
sprinkle about the tope. But a herd of elephants came
regularly, which brought water with their trunks to
water the ground, and various kinds of flowers and
incense which they presented at the tope."
1. Chinese-English for Naga.
2. "It" must be wrong. What he wanted to take away
was the relics. Beal translates, " Let me tale you
out," a more likely rendering, and one that would
harmonize with the Divyavadana legend as given
A group of elephants behaving precisely in this
way is sculptured on one of the bas-reliefs in the
Bharhut Tope (plates xv and xxx in Cunningham).
The pilgrim goes on to say that in recent times a
devotee, seeing this, had taken possession of the
This will probably represent the tradition at the
place itself about 400 A.D., or a few years earlier.
For Fa Hian left China in 399 A.D., and when he heard
this tale at Ramagama it was no doubt already current
there. It is good evidence of Ramagama having been
very early deserted. Incidentally, its distance east
of the Lumbini pillar is given as five yojanas, say
Only twenty or thirty years later is
Buddhaghosa's version of the story in the
introduction to the Samanta Pasadika, his commentary
on the Vinaya, in the portion edited for us by
Professor Oldenberg.(l) The story is well told, but
we need not repeat it, as it reproduces the Dipavamsa
version. In both versions the story is used merely as
an explanation of the way in which Asoka's son,
Mahinda, came to enter the Order. For it is on seeing
the glory of the 84,000 festivals that Asoka boasts
of his gift. But he is told that the real benefactor
is one who gives his son to the Order; and then he,
too, has both his son and his daughter initiated. All
this is said to have happened after the ninth year of
Asoka's reign had expired. We see there is nothing at
all in this version about the original eight stupas,
or rather seven of them, having been broken open.
But Buddhaghosa has another account in the
Sumangala Vilasini, a little later than the last, and
in that he introduces an entirely new factor. Here it
is not Asoka, but Ajatasattu who gets the relics out
of all the eight stupas (except that at Ramagama,
which is protected by the Nagas). This he does
(twenty years after the Buddha's death, according to
Bigandet, ii, 97) on the advice of Maha-kassapa, who
was afraid-it is not stated why--for their safety.
The king agrees to build a shrine for them, but says
it is not his
1. Oldenberg's Vinaya iii, 304 foll.
business to get relics. The thera then brings them
all, and the king buries them in a wonderful
subterranean chamber. In the construction of this
underground shrine Sakka, the king of the gods, or
rather Vissakamma, on his order, assists. And it is
there that Asoka, after breaking into all the seven
stupas in vain (the Nagas protecting the eighth),
finds the relics.(1) These he takes, and restoring
the place where he had found them, establishes them
in his own 84,000, not stupas, but viharas. It is
incidentally mentioned that Rajagaha is 25 yojanas,
say 190 miles, from Kusinara.(2)
The text of this part of the Sumangala has not
yet been published. It will appear in the forthcoming
edition for the Pali Text Society; and meanwhile an
English version of a very late Burmese adaptation of
the Pali can be consulted in Bigandet, ii, 131 foll.
The legend is here very well and clearly told, and
suggests possible explanations of several of the
obscurities and inconsistencies in the oldest version
in the Divyavadana.
The Mahavamsa (chap. v), which is again a very
little later, gives the episode of the 84,000 viharas
on the same lines as the Dipavamsa, omitting all
reference to the breaking open of the stupas. But it
agrees with the Divyavadana in stating (p. 35 of
Turnour's edition) that this building of the 84,000
viharas was the reason why the king's name was
changed from Asoka(3) to Dhammasoka.
The form of the legend, as thus given in almost
identical terms by the Dipavamsa and the Mahavamsa,
is no doubt derived by both from the older Mahavamsa,
in Simhalese, then handed down in the Maha Vihara at
Anuradhapura, and now lost.
About the same age (412-454 A.D.) is the Chinese
1. Is it possible that this idea can lie behind the
enigmatic expressions given above, p. 401, from
2. This harmonizes with the distances given in the
Jataka. See my "Buddhist Birth Stories," p. 87.
3. So the text. We ought perhaps to read Candasoka.
which Mr. Beal translated in vol. xix of the "Sacred
Books of the East," and which he calls a translation
of Asvaghosa's Buddha-Carita. Were this so, it would
be of the first importance for our point. But it is
nothing of the kind. There are resemblances, just as
there would be if two Christian poets had, in
different times and countries, turned the Gospels
into rhyme with poetical embellishments. There are
still closer resemblances, as if a later poet had
borrowed phrases and figures from a previous writer.
But there are greater differences. Taking the first
chapter as a specimen, the Chinese has 126, the
Sanskrit 94 verses. Of these, only about 40 express
the same thought, and this is often merely a thought
similar because derived from the same old tradition.
More than half the verses in the Sanskrit have no
corresponding verse in the Chinese, More than two-
thirds of the verses in the Chinese have no
corresponding verse in the Sanskrit. And even when
the verses do, in the main, correspond, there are
constant differences in the details and in the
wording. It is uncritical, even absurd, to call this
The blunder of dating the Lalita Vistara in the
first century on the ground of a 'translation' into
Chinese of that date, rests on a similar misleading
use of the word. We know of no such translation in
the exact and critical sense. Twenty years ago
(Hibbert Lectures, 198 foll.) I called attention to
this. But Foucaux's conclusion still sometimes
repeated as though it were valid. We must seek for
the date of the Lalita Vistara on other and better
grounds. Beal's so-called Dhammapada is also a quite
different and much later work than the canonical book
of which he calls it a version. See the detailed
comparative tables ibid., p. 202. Mr. Rockhill, "Life
of Buddha," p. 222, says that Beal's Chinese text
"could not have been made from the same original " as
the Tibetan version of the Buddha-Carita.
It was necessary to point this out as the Chinese
book has two verses, of interest in the present
discussion, which are not in the Sanskrit. If Beal
were right we should have
to ascribe them to Asvaghosa.(1) As it is we are in
complete ignorance of the real name and author and
date of the original of Beal's Chinese book. We must,
therefore, take the opinions expressed in the verses
referred to as being good evidence only for the date
of the Chinese book itself, only noting the fact that
they are taken from some Sanskrit work of unknown
date. The verses run, in Beal's words:-
"Opening the dagabas raised by those seven kings
to take the Sariras thence, he spread them
everywhere, and raised in one day 84,000 towers.
"Only with regard to the eighth pagoda in
Ramagrama, which the Naga spirit protected, the king
was unable to obtain those relics." (2,298.)
We see from Yuan Thsang's Travels, Book vi (Beal,
ii, 26), that this curious story still survived in
the seventh .century of our era. It is interesting to
notice how the legend had, by that time, become
rounded off and filled in. Thsang naturally has
nothing of the second Ajatasattu episode. He was
never in Ceylon, and we have no evidence that this
part of the legend was ever current in North India.
But he also drops the absurd detail of the 84,000
stupas built in one day; and he fills out the Naga
episode, making a very pretty story of it, turning
the Naga, when he comes out to talk to the king, into
a brahmin, and giving much fuller details of the
conversation. He mentions also the interesting fact
that in his time there was an inscription at the spot
"to the above effect."
Finally, when we come to the Tibetan texts, which
are considerably later,(2) we find an altogether
unexpected state of things. We have long abstracts of
the account, in the Dulva, of the death and cremation
of the Buddha and of the distribution of his relics,
from two scholars whose work can be thoroughly relied
on, Csoma Korosi(3) and
1. There are six Asvaghosas mentioned in Chinese
works quoted by Mr. Suzuki in his translation of
the " Awakening of Faith," p.7.
2. About 850 A.D.: see Rockhill, pp. 218 and 223.
3. "Asiatic Researches," xx, 309-317.
W.W. Rockhill.(l) According to both these authorities
the Tibetan works follow very closely, not any
Sanskrit work known to us, but the Maha-parinibbana
Suttanta. Where they deviate from it, it is usually
by way of addition; and of addition, oddly enough,
again not from any Sanskrit work, but on the lines of
the Sumangala Vilasini.
However we try to explain this it is equally
puzzling. Could they possibly, in Tibet, and at that
time (in the ninth century A.D.), have had Pali
books, and have understood them? In discussing
another point, Mr. Rockhill (p. ix) thinks that the
Tibetan author had access to Pali documents. M. Leon
Feer has a similar remark ("Annales," vol. v, PP. xi,
133), and talks at pp. 133, 139, 143, 221, 224, 229,
408, 414 of a Tibetan text as though it were a trans-
lation from a Pali one. And the translations he
gives, in support of his proposition, certainly, for
the most part, show that the texts are the same.(2)
Strange as it may seem, therefore, it is by no means
impossible that in our case also the Tibetan depends
on a Pali original, or originals. We have at least
good authority for a similar conclusion as to other
Tibetan writings. And we now know, thanks to
Professor Bendall, that a similar conclusion would be
possible in Nepal.(3)
If, on the other hand, our Tibetan texts are
based on Sanskrit originals, the difficulty arises
whence, at that date, could the Tibetans have
procured Sanskrit books adhering so closely to the
Rockhill has not even a word about Asoka; Csoma
Korosi has only a line, added like a note, at the end
of the whole narrative, and saying:--
1. "Life of Buddha," pp. 122-148, and especially
2. M. Leon Feer has not been able always to give
volume and page of the originals of these Tibetan
texts, often because they had not been edited. It
may be useful, therefore, to point out that his
page 145 = Anguttara, 5. 108.
" 222 = Ang.5. 342, Jat. 6.14.
" 231 = Ang. 4. 55 (which gives better
readings), comp. 2. 61.
" 293 = Divy. 193, Itiv. 76.
3. J.R.A.S., 1899, p. 422.
"The King Mya-nan-met (Asoka), residing at
Pataliputta, has much increased the number of
Chaityas of the seven kinds."(1)
What, then, are the conclusions to be drawn from
our little enquiry?
1. That the breaking open of stupas is not
mentioned at all in the most ancient Buddhist
2. That Asoka's doing so is first mentioned in a
passage long after his time. This passage is also so
curt, self- contradictory: and enigmatic, that we
probably have to suppose a confusion arising from
difference of dialect. It is of little or no value as
evidence that Asoka did actually break open seven of
the eight ancient topes.
3. The number of the stupas he is supposed to
have built --84,000--is derived from the traditional
number (which is about correct) of the number of
sections in the Four Nikayas, that is, in Buddhist
phraseology, in the Dhamma. This suggests a possible
origin of the whole of the legend.
4. In any case the eighth, that at Ramagama, was
untouched. The site of it can be determined within a
few miles, as we know, from the passages quoted
above, its distance from Rajagaha on the one hand and
the Lumbini pillar on the other; and we have,
besides, the details as to distance given by the
Chinese pilgrims. There was an inscription there,
presumably put up by Asoka's orders. It will be most
interesting to see if it lends support to, or could
have given rise to, the legend.
5. The greatest circumspection must be used in
dating any Indian work by the date of an alleged
translation into Chinese. Even when a Chinese book is
said to have the same title, and even similar
chapter-titles, as a Sanskrit or Pali one, it does
not follow it is really the same.
6. The Indian pandits who assisted in the ninth
century in the translation of Indian books into
Tibetan knew not only classical Sanskrit as well as
Buddhist Sanskrit, but also Pali. It would be a great
service if Tibetan scholars would
1. "Asiatic Researches," xx, 317.
But we must stop. We are here brought face to
face with some of the most debated of those larger
questions on the solution of which the solution of
the problem of the history of Indian thought and
literature must ultimately depend. We can only hope
in an enquiry like the present to lay one or two very
unpolished stones on the foundation of the Dhamma
they did not go to Ceylon.
ascertain exactly which Pali MSS. they had. They
certainly had the Paritta; and certain Suttantas
from, if not the whole of, the Digha; and certain
Suttas from, if not the whole of, the Anguttara and
the Samyutta. These books must have been handed down
all the time in India; for we know enough of the
journey of the emissaries from Tibet to be certain
Pasada of history, in which the scholars of a future
generation will, we hope, have the good fortune to