Comparative Study and Buddhist Works in Chinese Translation

·期刊原文

Comparative Study and Buddhist Works in Chinese Translation

Biswadeb Mukherjee

佛教與中國文化國際學術會議上輯

1995出版 P.259-283

--------------------------------------------------------------------------------

P.259

I
For a period of about l000 years the Buddhist
scriptures written in Sanskrit and Prakrt were
translated into Chinese. We are not concerned with
all such texts. In the present article while
discussing certain aspects of the comparative study
of Chinese and Indian materials we will confine
ourselves to those of the translated texts whose
originals, or if not originals, parallel versions
are still extant in India. Such texts, it is obvious
, offer a most suitable field for comparative study
. Here again will deal only with certain problems
of linguistic and historical interests which have
not yet been discussed in a thorough and systematic
manner they deserve.
The languages of India and China belong to two
completely different language groups. Except for a
few expressions and names which are phonetically
reproduced in Chinese, the Chinese translations
virtually provide us with word-by-word commentary on
the original Indian texts. Once translated the text
became frozen, and thus remained free from the
additions and alterations which the texts in India
underwent. Sometimes the Indian text got so changed
that a new translation of the same text became
imperative. The Chinese translations thus constitute
an important milestone in the history of a gradual
development of a text or of the philosophy contained
in the text. The point would be clear if we compare
the availab1e

P.260

enlarged version of the Lankavatara-sutra with
several Chinese translations of this text done in
different periods.
I would like to draw the attention of the
scholars to another problem towards the solution of
which the Chinese translations because of their
commentarial nature can be of much help. I am
speaking of the original meaning of an expression
which has become obscure due to the passage of time.
The statement will be vindicated when we, by way of
example, discuss the connotations of the two terms,
gama and esika. It is obvious from what we have said
that such comparative studies can be equally helpful
in understanding the ancient Chinese language. In
cases where both the languages have preserved an
identical version of a text, the obscure expressions
in the Chinese version can be greatly illumined with
the help of a corresponding Indian text. As it
explains the Chinese translations, the Indian
original can help us to understand the ancient
Chinese language in its various aspects. If we pay
attention to the time when the translation was done,
or the region from which the Chinese translator
hailed, we would be able to throw light on some of
the regional peculiarities and the period wise
developments of the Chinese language. Unfortunately
the contributions which the Indian texts can make
toward the study of ancient Chinese language has not
been properly recognized so far, and no comprehensive
and methodical study covering this important field
is known to me. It is a desideratum that we can
hardly afford to go on neglecting.

II
The present paper is not intended to be a
detailed treatment of all these problems. We will
discuss certain specific cases by way of
illustration of some of the points mentioned above.
We may start our discussion with the term gama which
occurs as a part of the formulation of the second
Parajika rule. This rule states that if a monk takes
anything from a gama or from aranna with the
intention of stealing what is not given to him, he
is guility of a parajika offence. i.e., an offence
entailing expulsion from the community. The rule
(註1)may be quoted here:

P.261

yo pana bhikkhu game va aranna va adinnam
theyya-samkhdtam adiyeyya yatha-rupe
adinnadane rajano coram gahetva haneyyum va
bandheyyum va pabba-jeyyum va-----talharupam
bhikkhu adinnam adiyamano ayam pi parajiko
hoti asamvaso'ti.
This phrase gama va aranna va occurs in all the
Vinaya versions of the second Parajika rule, and so
must have belonged to the earliest strata of the
Vinaya tradition current before the Sthavira-
Mahasanghika schism; there is no reason to
disbelieve the tradition that the Buddha himself
formulated the second Parajika rule. This expression
has always been translated by scholars as "either
from village or from forest." The term grama or gama
and its modern equivalents as gram, gam, or gaon
are universally known to the people in Northern
India as signifying a village. The Indians as well
as the foreigners conversant with the North Indian
languages will automatically translate the phrase as
'either from village of from forest.' Before we
express ourselves on the interpretation of gama as
village, one point needs to be emphasized here. The
Pratimoksasutra or the Suttavibhanga which contains
this rule represents the legal code of the Buddhists
and in a law-book every implication of the law is
clearly conveyed through unambiguous expressions and
nothing is left to imagination. There is not much
scope for speculation here. The acceptance of the
interpretation of gama as village would be
tantamount to saying that though stealing in a
village or forest is recognized as an offence of
utmost gravity, stealing in a city remains beyond
the scope of Buddhist law. Thus this interpretation
of gama as village lands us in an absurd situation
and so cannot be accepted. During the lifetime of
the Buddha powerful kingdoms have come into
existence, and their capitals were certainly
different from a mere village. It was in this period
that different types of settlements with different
appellations were already in existence.(註2) It is
reasonable to expect that the Buddha or the early
Buddhists must have taken this situation into account
while formulating the Vinaya rules. The term gama
was certainly used to mean more than a village only.
From the rule itself it appears that the terms
gama and aranna have been

P.262

used as mutually exclusive terms, and these two
contrary terms were juxtaposed to cover the entire
area where theft could be committed. This
juxtaposition of the two terms seems to have formed
a stock expression which is as old as the Rgveda
where these two terms occur in opposition to each
other, and also indicates the entire area where
night descends. It is obvious that 'gama' refers to
any area which is not aranna. This is confirmed by
the commentary embedded in the Suttavibhanga or in
the parallel versions preserved in Chinese
translations of other Vinayas. (註3)On the other hand
'aranna' has been translated in Chinese as 空地(kung
ti) (註4) or empty space, which
means the area where no body continuously stays. In
other words, aranna means an uninhabited area.
Consequently "gama" stands not only for village but
for any inhabited area viz city, market town,
village, etc. This conclusion of ours is fully
supported by the different Vinaya versions. The Th,
Mi, Dh, Ma. etc. hold gama to be the opposite of
aranna in connotation, and while enumerating
different types of settlements which fall under that
category, mention also city (nagara) as one of them.
The city according to the Ma (註5) version is
surrounded by a wall or encircled by a moat. We have
a similar description of a city on the border in the
Angnttra Nikaya or chunga-han-ching. (註6) The account
preserved in the different Chinese versions of the
Vinaya leaves no doubt that the term gama stands for
any inhabited area.
The Pali Suttavibhanga account is confusing, as
it has recorded earlier and later traditions
contradictory to each other. Here we find that a
gama is defined as a place which contains one to
four huts, a resting place for caravans, etc. This
definition does not mention any city, market town,
etc. This is a later tradition which is not found in
any other Vinaya including those of the Dh and Mi
which are otherwise so similar to the Pali Vinaya.
The old definition of gama finds its reflection in
the Suttavibhanga commentary which explains
'pabbajeyum' or should banish as "they would banish
from village, small town (or market town), city
(nagara), province, etc." This explanation of
pabbajeyum is in conformity with the earlier
tradition and contradicts the definition of gama.

P.263

This mistake on the part of the Pali Vinaya
throws interesting light on the growth of urban
civilization in India. We have seen that these terms
go back to the time of the Rigveda when only one
type of settlements was current among the Aryan
people. This type of settlement was called grama
while uninhabited area was known as aranna. These
two terms became a part of the legal code of the
country, and was known to the Buddha in their
original sense. As a current legal term he included
them in his formulation of the Vinaya rule. This
inclusion shows that the original implications of
these two terms were still widely understood by the
people. Gradually with the growth of powerful
kingdoms in North Indian, the expansion of trade and
accumulation of wealth, urban culture grew and
different types of settlements came into existence.
Gama came to mean the smallest rural unit which
still retained some of the characteristics of the
original settlement while 'nagara' referred to
comparatively bigger, urbanized units. This
nomenclature covering different types of settlements
had become quite familiar to the people. The wider
legal implication of the term gama was no longer
automatically understood, and conmmentaries had to
be composed to explain such terms. It may be said
that the original combination of gama and aranna was
the outcome rural culture while the commentary
reflects the growth of urban civilization.
In Chinese translations of Vinaya the term gama
or grama has been accurately rendered as 聚落 (chu
luo). According to the 漢語大字典, (註7) chu means
an assembly, a group of people while one of the
meanings of luo (註8) is 居處, place of residence, an
inhabited area. So Chu Luo means an area inhabited
by a group of people, which is exactly the meaning
of grama. Unfortunately under the influence of the
misinterpretation of gama as 'village', chu luo has
also been wrongly translated as village.
The other term which I am going to discuss is
esika. The difficulty of this term lies in its
obscurity, and not in its familiarity. So far as I
know, this obscure word has always been translated
as 'pillar'. For the time being we will leave the
word untranslated and proceed to examine some of the
passages containing this expression
. In Digha Nikaya(註9) passage the word

P.264

esika is used to depict the permanency of a soul
(atta), e.g. esikatthayithito, staying permanently
like an esika. This passage shows that esika was a
strongly built structure of high durability. To
understand what esika really was we must refer to a
passage from the Anguttara Nikaya (Vol. iv.p.106)
which gives us some extra informations about esika,
and sheds light on the purpose for which it was
built. This text speaks of seven requisites which
will provide a nagara, i. e. city against outside
enemies. The passage in question runs as follows:-:
Idha bhikkhave ranno paccantime nagare esika hoti
gambhiranema sunikhata acala asmpavedhi....imina....
suparikkhitam abbhantaranam guttiya bahiranam
patighataya...E. M. Hare (註10) translates this
passage: "Monks, there in a raja's citadel in the
pillar, deeply embedded immovable, and unshakable.
With this...the rajas citadel in the marches is well
provided for the protection of the inmates and for
the warding of outsiders".
Following the Pali commentaries the Pali Text
Society Dictionary (註11)gives the meaning of esika
as pillar, and further points out that esika may be
a by-form of isika(Skt. isika) which means a reed.
E. M. Hare in his translation (The Book of Gradual
Sayings, Vol. IV, p.70)also renders esika as pillar.
In note no.1,p.70he further points out that esika is
a symbol of stability.
We have to take note of 3 points regarding this
interpretation: i) esika means a pillar, ii) esika
might have been derived from isika whose meaning is
reed, iii)esika is a symbol of stability. All these
surmises seem to be baseless speculations and cannot
be accepted. The very fact that isika means 'reed'
goes against this speculation. It is difficult to
imagine how esika which stands for a durable,
strongly constructed and big structure can be
derived from isika which stands for a frail thing
like 'reed'. Again it is obvious that esika had a
positive role to play in the defence of border city
by repulsing the external enemies. A pillar cannot
be expected to fulfill such a role. In the end we
may mention that there is no tradition about the
pillar being a symbol of stability.
The term esika can be reasonably derived from
the verbal root √ is

P. 265

(Skt. √ is) which means "to search". Such cognate
words as esa, esana or esana are all derived from
this root, and mean "searching". Consequently the
term esika can be translated as a search-tower or
watch-tower. This interpretation is in full
agreement with the Anguttara Nikaya passage quoted
above. The passage in question may be translated as
follows: -"Monks, in a king's city on the border
there is a watch-tower deeply embedded, well dug in,
immovable and impenetrable, with this... the king's
city on the border is well provided for the
protection of the inmates and for the repulsion of
the outsiders".
The same passage occurs in the chung-a-han-ching
, and there the translation of the term esika as lou
lu (樓櫓) also supports our conclusion. Lou stands
for a tall structure while lu means a look-out. Thus
the Chinese translation supports our rendering of
esika as watch-tower.(註12)
In Indian originals and Chinese translations
there are many such words whose meanings are not
properly understood. They may yield their true
meaning when studied with the help of both the
Indian and Chinese versions. Attention may be drawn
to another area which has scarcely been explored.
The Chinese is a non-inflectional language while
Indian languages are highly inflectional. The
parallel Skt. texts compared with Chinese renderings
would be of much help in understanding the various
means which the Chinese translators employed in
expressing different verbal tenses and moods found
in Indian originals. We may, for example, refer to
the rendering of perfect passive participle forms
with yu (有). Samskrta or sankhata has translated as
有為 yu wei, and upatta as yu shou chih (有執受).
(註13) A thorough and comprehensive study in this
field would help to understand some aspects of old
Chinese better.

III
It is towards the study of ecclesiastical history
that the texts presserved in Chinese translation can
make the greatest contribution. When compared with
the Theravada scriptures preserved in Pali language,
they yeild a rich crop of valuable data regarding the
origin and development of traditions,

P. 266

reconstruction of various earlier versions of the
texts now available, the interrelation between the
different versions of a text belonging to different
Buddhist schools, etc.(註14)
Regarding this aspect of comparative study the
different versions of the Vinaya Pitaka beloging to
different schools offer us a unique opportunity. Any
of these versions, if studied in isolation, will
not be able to tell us anything decisively on these
points. On the other hand, a comparative study of all
these versions results in bringing to light valuable
internal evidences bearing upon these problems.
Six complete versions of the Skandhaka and
Vibhanga portions of the Vinaya Pitaka have come down
to us. Five of these Vinayas, i.e. the Vinayas of
the Thervadins, Mahisasakas, Dharmaguptas,
Sarvastivadins and Mulasarvastivadins (henceforth
abbreviated as Th, Mi, Dh, Sa and Mu respectively)
developed out of the original Sthvira Vinaya. Out of
these Vinayas only the Th can be found in Pai while
the Mu, either wholly or partially , has been
preserved in Sanskrit, Tibetan and Chinese languages
. The other Vinayas viz, the Mahasanghika Vinaya
(abbreviated as Ma) and the Dh, Mi, Sa Vinayas are
mainly found in Chinese translations.
The materials which are common to all the six
versions should have been drawn from a source which
is prior to the Sthavira-Mahasanghika schism. So the
materials belonging to this period will at least
tell us about the earliest traceable version of the
Buddhist Vinaya current before the Second Buddhist
Council, and belonging to the undivided Buddhist
community.
There are some other traditions which are found
only in the five Vinayas that developed out of the
original Sthavira Vinaya. These traditions can be
reasonably assumed to have belonged to the Vinaya
Pitaka of the Sthaviras. These traditions are not
found in the Ma. This opens before us two
possibilities. Either these traditions belonged to
the earliest Vinaya and the Ma has lost it, or they
represent later additions and elaborations by the
Sthaviras. Which of these two possibilities is more
probable can only be decided if there are some other
traditions also

P. 267

belonging to the earliest strata of tradition, and
supporting or rejecting these traditions mentioned
in the Sthavira Vinaya.
There is still another group of traditions which
are mentioned either in the Vinayas of Th, Dh, and
Mi or in the Vinayas of the Sa and Mu. But often
both these two sets of materials betray later
additions, so they did not form a part of the
original Sthavira tradition. On the other hand they
do not represent disorganised, loose mass of oral
traditions. They are organised in a continuous
narration following a fixed sequence. This shows
that these two sets of traditions must have formed a
part of two different fixed versions - one of which
was inherited by the three schools while the other
by the two schools. This would suggest that these
five Buddhist schools (i.e. the Theravadins,
Mahisasakas, Dharmaguptakas, Sarvastivadins, and
Mulasarvastivadins) originated out of two Buddhist
communties which owned these two versions. The
identification of these two Buddhist communities
would throw further light on the origin of Buddhist
sects, but that would be out of scope for the
present article.
We have seen that none of these two versions of
Vinaya can be identical with the original Sthavira
Vinaya. But the two communities which owned these
versions must have descended from the Sthaviras, for
their Vinayas were inherited by the five schools
which developed out of the Sthaviras. So it would
not be unreasonable to describe these two Vinaya
versions as modified versions of the Sthavira Vinaya.
What we have discussed until now may be shown in
the following table:
Vinaya of the undivided Buddhist community
Sthavira Vinaya Mahasanghika Vinaya Modified
Version I Modified Version II
Th Dh Mi Sa Mu
A few specific cases may now be disussed in
order to illustrate the thesis propounded above.
A: - The case of the Sanghabhedakarani vatthuni:
- The Cullavagga(註15) of the
Th Vinaya records a list of 18 schismatic matters or
eighteen types of misinterpretations of the teachings
of the Buddha which can cause a schism

P. 268

in the Order. The list runs as follows. -1)
adhammam dhammo ti dipenti, 2) dhammam adhammo ti
dipenti. 3) avinayam vinayo ti dipenti, 4) vinayam
avinayo ti dipenti, 5)abhasitam alapitam tathagatena
bhasitam lapitam tathagatena ti dipenti, 6) bhasitam
lapitam tathagatena abhasitam alapitam tathagatena
ti dipenti, 7) anacinnam tathagatena acinnam
tathagatena ti dipenti, 8) acinnam tathagatena
anacinnam tathagatena ti dipenti, 9) apannattam
tathagatena pannattam tathagetna ti dipenti, 10)
apannattam tathagatena apannattam tathagetena ti
dipenti, 11) anapattim apatti ti dipenti, 12)
apattim anapatti ti dipenti, 13) lahukam apattim
garuka apatti ti dipenti, 14) garukam apattim lahuka
apatti ti dipenti, 15) savasesam apattim anavasesa
apatti ti dipenti, 16) anavasesam apattim savasesa
aptti ti aipenti, 17) dutthullam apattim adutthulla
apatti ti dipenti, 18) adutthullam apattim dutthulla
apatti ti dipenti.
The Dh(註16) and Mi(註17)
Vinayas have preserved accounts similar to that of
the Th.They have recorded only 9 schismatic matters.
These three account may be tabularised as follows:-
Th Dh Mi
No 2 same (No 1) same (No 1)
No 4 same (No 2) same (No 2)
No 6 same (No 9) same (No 8)
No 10 same (No 8) same (No 9)
No 12 Explains apatti as same as No 12
anapatti whether it is of Th (No 3)
laghuka or garuka
(No 3 & 4)
No 14 X same as No 14
of Th (No 4)
No 15 same (No 5) same (No 5)
No 17 same (No 6) same (No 6)
X Explains krtya as By karma one can get
aktya(No 7) rid of an offence (is misinterpreted as) one
can get rid of an offence
without karma (No 7)
No 7＆8 X X

P. 269

A different stream of tradition has been
preserved in the Vinayas of the other two Sthavira
schools, the Sarvastivadins and Mulasarvastivadins.
The Sa records two slightly different versions (註l8)
of the same tradition, one in the Devadattavastu,
and the other in the Sanghavasesa section enumerating
14 schismatic matters.
Sanghavasesa List
1)adharma explained as dharma
2)dharma explained as adharma
3)avinaya explained as vianya
4)vianya explained as avinaya
5)anapatti explaned as apatti
6)apatti explained as anapatti
7)laghuka explained as garuka
8)garuka explained as laghuka
9)savasesa explained as anavasesa
10)anavasesa explained as savasesa
11)krtyata explained as akrtyata
12)akrtyata explained as krtyata(註19)
13)asasana explained as sasana
14)sasana explained as asasana

Devadattava List
1)adharma explained as dharma
2)dharma explained as adharma
3)akusala explained as kusala
4)kusala explained as akusala
5)anapatti explained as apatti
6)apatti explained as anapatti
7)laghuka explained as garuka
8)garuka explained as laghuka
9)savasesa explained as anavasesa
10)anavasesa explained as savasesa
11)krtyata explained as akrtyata
12)akrtyata explained as krtyata
13)abhasita explained as bhasita
14)bhasita explained as abhasita
The Mu (註20) also states that the schismatic
matters are 14 in number. However the list of these
matters I failed to trace in the Chinese translation
of this Vinaya. As both the Sa and Mu have shared a
common tradition for their account of the schism
caused by Devadatta, it may be surmised that the Mu
account might have been closely similar to, if not
identical with, the Sa account.
According to the third stream of tradition
recorded in the Mahasanghika Vinaya the occasion for
a schism may occur if a monk explains 1) dharma as
adharma, or 2) vinaya as avinaya, or 3) apatti as
anapatti, or 4)guruka apatti as laghuka apatti, or
5) that which is punishable as that which is not
punishable, or 6) dharma-karma as adharma-karma, or
7) (sangha- ) samagri-karma as not (sangha-) samagri
-karma, or 8) karaniya as akaraniya.

P. 270

Reconstruction of the earlier versions of the
schismatic matters
In our introduction to this section we have
already given a brief sketch of the method followed
in this article in comparing the different versions
of Vinaya. We would like to point out a few more
points that deserve attention.
i) A tradition, though not mentioned in all
the six Vinayas, also should be regarded
as belonging to the earliest strata if it
finds mention in the majority of sources
belonging to all the three streams.
ii) A tradition mentioned only in the original
Sthavira Vinaya or only in the Ma can still
be regarded as being a part of the earliest
strata of tradition if it is confirmed by
another tradition of equal antiquity. In
this connection we may take note of another
reason specially applicable to the problem
of schismatic matters. We may take it for
granted that the list of schismatics matters
was formulated with the intention of
preventing all preventing possible
distortions of unwarranted additions to
what was regarded as the instructions
imparted by the Buddha. So we may be
justified in accepting a matter recorded
only in the undivided Sthavira stream as
belonging to the earliest tradition if it
covers an aspect of the Buddha's instruction
which has not been guarded by any other
matter in the Ma account. The same applies
to a tradition mentioned only in the Ma
Vinaya but covering a part of the teachings
of the Buddha missed in the Sthavira
tradition. On the other hand if such a
tradition is taken care of by any other
matter mentioned in all the three streams,
we have to reject it as a later addition.

An earlier version as the source of the Th, Dh, and
Mi accounts
What immediately draws our attention is the fact
that while the Th records 18 schismatic matters and
also gives the number of such matters as 18, the Dh
and Mi have noted only 9 matters leading to the
schism in the Order. To state the difference more
precisely, Th devotes two matters to each of the
topic (e.g. dharma) against one matter in Dh and Mi.
Which of these two ways of formulation represent the
original tradition?

P.271

We have pointed out that the Buddha and the
Sangha had to be constantly on guard against
possible distortions of the true teachings. A number
of rules were so formulated that any heterodox view
could be immediately detected and eliminated. Now
how the introduction of a heterodox view may take
place?Let us by way of illustration take the case of
dharma. Here the heterodox view may be introduced
in two different ways: 1) either when adharma i.e. a
view contrary to the dharma is accepted as dharma,
or 2) When the dharma is misinterpreted as adharma.
The Th formulation of the possible situations under
which a schism can place covers both these types of
deviations. The first type of deviation (adharma
explained as dharma) was formulated to detect the
introduction of heterodox elements into the body of
the true Law (dharma) while the second type (dharma
explained as adharma) was formulated to prevent the
misinterpretation and consequent rejection of the
words of the Buddha. Here any one way formulation is
not in a position to cover all the deviations from
the true teachings, and thus fails to serve the very
purpose for which it is formulated. The two way
formulation recorded in the Th, therefore. appears
to represent faithfully the original intention. The
conclusion is strengthened by the evidence of the Sa
which records the same two way formulation.
However the recording of the nine matters cannot
be regarded as a lapse on the part of Dh;this rather
betrays a clever device by which all the 18 matters
have been preserved while actually mentioning only 9
of them. From the principle of formulation
enunciated above, as well as on the analogy of Th,
which is but an illustration of this principle, it
is evident that given a rule, its counterpart can be
deduced by formulating it anew in a reverse order.
For example, if the rule "dhammam adhammo ti dipenti
" is given, we can deduce the opposite of it in the
form of" adhammam dhammo ti dipenti." So each of
these 9 rules in the Dh actuall implies the
existence of its counterpart, and the total number
comes upto 18 here also. This surmise is fully borne
out when we find out that this very Dh Vinaya gives
the total number of schismatic matters as 18 before
listing the above

P.272

mentioned 9 matters.
The same principle of deduction should, no doubt
be applied to the Mi list containing 9 schismatic
matters which, excepting three rules, is identical
with that given in the Dh account. The Mi account
also shows striking similarities with that of the Th
. So it may be concluded that the Mi account really
gives information about 18 schismatic matters.
Thus we find that all the three Vinayas have
followed a common tradition according to which there
are 18 potential causes of schism. Moreover these
three Vinayas have preserved almost identical items
in the lists of schismatic matters except the
following minor differences
1) The two matters regarding "tathagatena
acinnam and tathagatena anacinnam" mentioned
in the Th are conspicuous by their absence
in the accounts of the Dh, Mi, Sa. and Ma
Vinayas, and represents a later development.
2) Between the krtya of the Dh list(No.7) and
the Mi accnunt of "karma absolving one from
offence" (No.7), the former seems to have
formed a part of the undivided Sthavira
version owing to the following
considerations:
i) The matters regarding the krtya have also
been recorded in the Sa (No.11 and 12); ii)
krtya in its Unmodified scope, as in the
case of the Dh version, fully covers all
kinds of Formal Acts, and consequently the
Dh formulation is in a position to prevent
any distortion of any Formal Act. The Mi
formulation is limited in its scope and
takes into account only one type of Formal
Act.
3) Except Dh, the other four Vinayas (Th, Mi,
Sa, Ma) representing all the three streams
of tradition, have recorded two separate
matters, one combining apatti and anapatti,
and the other dealing with garukapatti and
laghukapatti. The formulation which combines
all these four concepts, and of which the
Dh is the sole repository, is surely of
later origin.
When we exclude the schismatic matters which
have been to be additions and distortions (Nos.7 & 8
of Th, No.7 of Mi; Nos.3 & 4 of Dh)

P.273

the remaining matters also 18 in number may be taken
to be the earlier account on which the Th, Mi, and
Dh versions are based.
This earlier version contains the following
rules, viz. rules nos.1 to 6 and nos.9 to 18 of the
Pali list,altogether 16 rules. To this we have to
add rule No.7 of the Dh list together with its
opposite formulation.
Another earlier account on which the Sa version is
based
As noted above (see, p. 13) the Sa records two
slightly different versions of the schismatic
matters, each of which suffers from omissions and
later additions. For example, the matters regarding
"vinaya" preserved in the Sanghavasesa version, but
omitted in the Devadattavastu list, surely formed a
part of the undivided Sthavira tradition, as these
also find mention in all the other Vinayas. Likewise
, kusala mentioned only in the Devadattavastu list
is a later addition. Again the Devadattavastu list
records "bhasita" in place of "sasana" of the
Sanghavasesa version. "Bhasita" and not sasana finds
mention also in the Th, Dh, and Mi accounts. Moreover
sasana may stand for both dharma and vinaya (cf.The
staock phrase: ayam dhammo ayam vinayo ayam
satthusasanam). Therefore sasana cannot be taken as
something separate from dhamma and vinaya. Hence
bhasita is a better reading and could be taken to
have belonged to the original Sthavira tradition.
Original Sthavira tradition
The Sa has recorded only bhasita and not
prajnapta. The question whether bhasita and
prajnapta belonged to the earliest tradition is
irrelevant here. What we have to consider is that
according to the Buddhist tradition which most
probably belonged to the earliest period, bhasita,
and prajnapta stood for dharma and vinaya which
constituted, as it were, the two wings of the words
of the Buddha. The fact that the Sa has recorded
only one member of the group, suggests that the
other member which was most probably included in the
list of schismatic matters in an earlier period,
got lost to the Sa list afterwards. The three
Vinayas (Th, Dh, and Mi) in recording both these
members (i.e. bhasita and prajnapta) haven given a
more trustworthy account of the earlier version.
Taking this conclusion as a basis

P.274

for further argumentation we can proceed a step
further. We can justifiably add these two schismatic
matters regarding prajnapta to the Sa list (also
probably to the Mu list) and consequently hold that
these two lists are based on a tradition which
contained at least 16 schismatic matters.
On the other hand the two schismatic matters
concerning dutthulla (註21) and adutthulla mentioned
in Dh, Th, and Mi but omitted in the Sa and Ma (i.e.
mentioned in only one stream of tradition but omitted
in the other two streams) might well be discarded as
later addition. The dutthulla stands for the four
Parajika offences and thirteen Sanghavasesa offences
which form a part of the Pratimoksasutra; they cannot
form a category of their own and must be included in
the category of vinaya (see, P.). These two matters
are certainly superfluous later additions.
We find that these two traditions differ in only
two cases. viz. 1) mention of dutthulla and
adutthulla in Th, Dh, and Mi and 2) omission of
prajnapta in Sa, and contain too many identical
matters to be accidental. This state of things
forces us to postulate another tradition of still
earlier date of which these two traditions are but
slightly modified later versions.
The number of the potential causes of schism
included in this earlier tradition can be fixed with
a fair amount of certainty. Whether we add two
matters regarding prajnapta to the Sa list or deduct
two matters concerning dutthulla and adutthulla from
the accounts of the Th, Dh, and Mi, we get a total
of 16 matters.
The arbitrary change in the total number of
schismatic matters as evidenced by the recording to
two different numbers in the Vinayas (18 in Th and
Dh; 14 in Sa and Mu) suggests that this earlier
tradition did not expressly mention the total number
, as it is still the case with the Ma version, and
the Sanghavasesa rule dealing with the schism in the
Order.
This earlier tradition differs, as it will be
shown, from the Ma version and has been utilised by
the five schools which arose from the Sthaviras. So
it is evident that this earlier tradition is to be
recognised as the original Sthavira tradition which
gained currency after the cessation of the
Mahasanghikas but before the rise of the above
mentioned five schools

P.275

(Th, Dh, Mi, Sa, and Mu). The original Sthavira
tradition runs as follows:-1) dharma explained as
adharma, 2) adharma explained as dharma, 3)vinaya
explained as avinaya, 4) avinaya explained as vinaya
, 5) bhasita explained as abhasita, 6) abhasita
explained as bhasita, 7) prajnapta explained as
aprajnapia, 8) aprajnapta explained as prajnapta,
9)anapatti explained as apatti, 10) apatti explained
as anapatti, 11) laghuka apatti explained as guruka
apatti, 12)guruka apatti expiained as laghuka apatti,
13) krtya explained as akrtya, 14) akrtya explained
as krtya, 15) savasesa explained as anavasesa, 16)
anavasesa explained as savasesa.
Each of the two traditions on which the accounts
of the five schools are based, is identical with the
original Sthavira tradition except in two points (
mention of dutthulla and adutthulla in the Th, Dh,
and Mi, and the non-mention of prajnapta and
aprajnapta in the Sa). These deviations show that
the accounts of the five schools are not directly
dependent on the original Sthavira tradition, but on
the two modified version of it. (註22)

Reconstruction at the earliest version of the
schismatic matters
The Ma list of 8 schismatic matters differs from
this reconstructed Sthavira version only in a few
cases. We find that the Ma account like the Dh and
Mi versions has recorded only one-way formulation.
Here also each of the matters surely indicates its
counterpart. Thus the Ma version actually gives
information about 16 potential causes of schism.
When we compare the Ma version with the reconstructed
original Sthavira account, we are immediately struck
with the close similarity existing between these two
traditions. The eight matters connected with dharma,
adharma, vinaya, avinaya, apatti, anapatti, garuka
apatti and laghuka apatti are common to both these
versions. Again the dharma-karma of the Ma version
means legally valid Formal Acts while adharma-krama
means those Formal Acts which are not legally valid.
These two matters are synonymous with the krtya and
akrtya of the reconstructed Sthavira version. So far
the other schismatic matters are concerned (i.e.nos.
5,7,8 of Ma; 5, 6, 7, 8, 15, 16 of the reconstructed
of the Sthavira list), the two versions differ from
each other. The schismatic matters regarding savasesa
and anavasesa (Nos. 15 &

P.276

16) are included in the Sthavira account only. As it
will be shown later, the misinterpretation regarding
these two matters are covered by the schismatic
matters concerning laghuka apatti and garuka apatti
mentioned in both the Sthavira list and the Ma
account. These two potential causes of schism can,
therefore, be discarded as later additions.
Again the Sthavira list alone contains four
schismatic matters regarding bhasita, abhasita,
prajnapta, and aprajnapta (Nos. 5, 6, 7, 8). The
exact purport of these terms becomes clear when we
read them in the light of the account of the First
Council. (註23) Here we read
that Mahakassapa asked Upali questions on Vinaya.
The first question he asked Upali runs as follows:
"Pathamam avuso upali parajikam kattha pannattam ti".
We find that the first parajika has been brought into
association with the verbal form pannatta. In an
identical form questions were asked about the other
parajikas. In the end we read that Mahakassapa asked
questions about the two"vinayas" and this shows that
in the technical language of the early Buddhists the
rules of vinaya was connected with the verbal form
pannatta. On the other hand Ananda was asked about
"dharma," the questions on which were formulated in
the following manner: "Brahmajalam avuso Ananda
kattha bhasitam ti." In an identical manner Ananda
was asked about all the sutras which came to be
included in the Agamas or Nikayas, This shows the
early association of the term "bhasita" with the
sutra dealing with "dhamma." So we may conclude that
the term "bhasita" refers to the words of the
Buddha on dhamma, while the term "pannatta" refers
to the instructions of the Buddha on "Vinaya."
It is clear that the terms "bhasita" or
"pannatta" in the reconstructed Sthavira list do not
refer to anything distinct from dhamma or vinaya
respectively. The non-mention of these four matters
in the Ma version therefore reflects the earliest
tradition current before the Sthavira-Mahasanghika
division.
On the other hand the matters regarding(sangha-)
samagri-karma(註24) mentioned only in Ma stands for
one type of Formal Acts, and consequently should be
included in the category of "krtya" (dharma-karma).
Hence it has

P.277

to be rejected as a superfluous later addition.
Likewise the punishable offences mentioned only in
Ma, and referring to parajika and other offences
included in the vinaya should be taken as later
elaborations. Again the Ma formulation combining
karaniya and akaraniya should be excluded from the
earliest version of the schismatic matters, as
karaniya (註25) in a general sense refers to any
proper conduct of a monk or nun, and does not form
a special category of its own.
To sum up, the Ma and the original Sthavira
versions closely resemble each other and differ only
where they have incorporated later elements. As
both the lists contain later elements, none of them
can be taken to be the earliest version. This state
of things is a proof in itself of the existence of
another still earlier tradition on which these two
versions are based. This earlier version must have
been current long before the Sthavira Mahasanghika
schism. as it has been faithfully preserved both in
the Sthavira list and the Ma version. Moreover in
the very creation of an Order the problem of schism
is inherent. In fact we know that the Buddha himself
had an experience of schism when Devadatta broke
the Order. Under such circumstances we can
reasonably surmise that the list of the schismatic
matters originated with the Buddha himself.
The reconstructed earliest version of the
schismatic matters current before the Sthavira-
Mahasanghika schism runs as follows. 1) dharma
explained as adharma, 2)adharma explained as dharma,
3)vinaya explained as avinaya, 4)avinaya explained
as vinaya, 5)apatti explained as anapatti, 6)anapatti
explained as apatti, 7) krtya explained as akrtya,
8) akrtya explained as krtya, 9) laghuka spatti
explained as guruka apatti, 10)guruka apatti explained
as laghuka apatti.
Possible objections refuted
One may object to the inclusion the laghuka and
guruka apattis in the original version of the
schismatic matters on the ground that they are not
outside the scope of vinaya and apatti. For
laghukapatti may either stand for the pacittiya or
for the apatti, while gurukapatti means parajika and
other serious offences. Still the combination of
such contrary terms as

P.278

laghuka and guruka in the formulation of schismatic
matters is not superfluous; for the combination of
such contrary terms is a safeguard against those
deviations from the accepted version of the
teachings of the Buddha which cannot be checked by
the combination of contradictory terms like dharma
and adharma, vinaya and avinaya, etc. in case of such
combinations the negative members like adharma,
avinaya, anapatti, etc, signify only the negation of
their positive counterparts and are themselves
devoid of any valid content. On the other hand in
case of such combinations as laghuka-guruka, none of
the members is a mere negation of the other.
Consequently this type of combination can deal with
deviations which involve two concepts of positive
nature. For example, if one explains an anavasesa
offence (i.e. parajika offence) as a savasesa
offence (e.g. sanghavasesa offence), then it is not
a case of misinterpretation of vinaya as avinaya,
for both the parajika and sanghavasesa refer to
positive concepts in the vinaya. Here is a case of
misinterpretation of a more serious offence as a
less serious offence, for parajika is more serious
offence than sanghavasesa.
Earliest classification of Satthusasana
One significant fact emerges from the list of
ten schismatic matters reconstructed by us. These
ten numbers actually deal with four different
categories of the teachings of the Buddha. When we
think that these ten schismatic matters were
formulated to prevent any distortion of what was
regarded as the teachings of the Buddha, we come to
realize that these matters must jointly cover the
whole of the Buddha teachings. In other words one
list of the schismatic matters reveal the fourfold
classification of the words of the Buddha, viz.
dharma, vinaya, krtya and apatti. This fourfold
classification though nowhere explicitly mentioned
in the canonical literature now available, seems to
have been recognised by the early Buddhist community.
B: Fixing the earliest meaning of a technical term
Again it is only through a comparative study of
the materials preserved in Chinese translation with
the Pali account that we can in some cases

P.279

determine the earliest connotation of a technical
term. Let us, by way of example, study the concept
of "vinaya" as known in the earliest period.
Generally the term "vinaya" is taken to mean any
ecclestiastical rule controlling the conducts of
individual monks and nuns or the community life. But
this was not the meaning ascribed to the term
"vinaya" in the beginning. We have seen from the
different list of schismatic matters as well as from
the reconstructed earliest version of such matters
that vinaya was not only different from the "dharma,
" but also from the krtya and apatti. That means
that the term vinaya means only the rules contained
in the Pratimoksasutra or the Suttavibhanga. The
traditions contained in all the six vinaya versions,
i.e. the earliest traceable tradition, not only
confirm this but also helps us to decide more
precisely what was meant by the term "vinaya".
The Dh account of the First Council is most
illuminating. (註26) It states that in course of the
recitation of the vinaya Upali was first asked about
the Parajikas, Sanghavasesas, Aniyatas, Patayantikas,
Pratidesaniyas, and Saiksa-dharmas; and then he
was asked about the different Skandhaka sections.
Immediately after this follows a significant statement
(註27) that the two vinayas (i.e.the Bhiksu-vinaya
and the Bhiksuni-vinaya)together with the Skandhakas
and the appendices constitute the Vinaya Pitaka. It
is evident that according to the Dharmagupta tradition
the vinaya meant only the disciplinary rules from
the Parajikas to the Saiksadharmas, but not the
Adhikaranasamathas, the Skandhas, etc. The Th (註29)
also supports this conclusion. Here we read that
Upali was first asked the four Parajikas and then it
is stated that he was asked in the same manner about
both the vinayas. What is meant by vinaya here
becomes clear when go through the discussions on
khuddanukhuddakani Sikkhapadan (minor vinaya rules).
(註30) The relevant portion may be summarised as
follows: "Some theresa said thus:Leaving aside the
four Parajikas,the remaining rules are minor vinaya
rules ( cattari parajikani thapetva avasesani
khuddanukhuddakani sikkhapadani ti). Some said that
leaving aside the four Parajikas and the thirteen
Sanghavasesas, the remaining rules are minor vinaya
rules; others

P.280

said that save those, and the two Aniyatas; others
all save those and the thirty nissaggiyas; others
that all save those and the ninety-two Pacittiyas,
and others that all save those and the four
Patidesaniyas are the minor vinaya rules."
From the above account it is clear that the
khuddanukhuddakani sikkhapadani means certain vinaya
rules which are now found in the Pratimoksasutra
and the Vibhanga. From the concluding lines of the
account given above, it follows that the vinaya also
included other rules in addition to the four
Patidesaniyas. If we pay attention to the sequence
of vinaya rules as recorded in all the versions of
the Vibhanga and the Pratimoksasutra we find that
the sekhiya (saiksa) rules occur inevitably after
the Patidesaniyas. So we may conclude that according
to the Th account the vinaya included also the
sekhiya rules. However the Adhikaranasamathas do not
appear to have formed part of the vinaya. Had the
Adhikaranasamathas also been included in the vinaya,
we would have expected a statement like this:
"Leaving aside the four parajikas.... the sekhiyas
the remaining rules are the minor vinaya rules."
The Mi (註31) and the Ma (註32) account likewise
includes all the disciplinary rules upto
the Saiksadharmas in the vinaya, but omits the
Adhikaranasamathas. The Mu (註33) account of the First
Council included all the rules up to the
Adhikaranasamathas. However as the inclusion of the
Adhikaranasamathas in the vinaya is not supported
by any other version, we may conclude that this is a
later addition. The same is the case of exclusion of
the Saiksa dharmas from the Sa account, as it has
been included in all the other Vinayas. Moreover
in the different versions of the Vighanga, the
Saiksadharmas, like the other rules in the vinaya,
have been fully discussed with reference to the
vastu, nidana and pudgala.
We may therefore conclude that by the term
vinaya the Buddhists of the earliest period meant
the disciplinary rules from the Parajikas to the
Saiksadharmas, but not the Adhikaranasamathas. That
the Adhikaraasamathas were not regarded as a part of
the Vinaya is also suggested by the manner of their
treatment in the Vibhanga. (註34) Unlike the

P.281

other vinaya rules (Parajika, etc), there is no
mention of the nidana, vastu and pudgala with
reference to the Adhikaranasamathas. The different
versions of the Vibhanga just make a bare mention of
the seven types of Adhikaranasamathas. Quite
significantly the account here is even shorter than
that given in the Pratimoksasutra. In short, the
section on the Adhikaranasamathas has the appearance
of being appended to the Vibhanga at a later date.
Adhikaranasamathas were later added to the
Pratimoksasutra after the death of the Buddha when
the danger of schism increased. Originally they as
krtyas belonged to the category of krtya.
We can throw light on many other similar
problems by means of comparative study. But it is
not necessary to multiply examples anymore. The
previous discussions have already demonstrated that
one can arrive at reliable conclusions bearing upon
certain aspects of the history of Buddhism only
through comparative studies. It is also obvious that
for such comparative studies materials preserved in
Chinese translations are absolutely indispensable.

NOTES

(註 1) Vinaya Pitakam, Vol. Ill, p.46.

(註 2) We may refer to Kusinara where the Buddha
was going during his last journey. Nara of
Kusinara no doubt stands for nagara which
means city, and has been accordingly
rendered into Chinese as cheng ( 城). During
the lifetime of the Buddha Kusinara was in
a decadent condition, but according to the
Buddhist tradition it was a prosperous city
in ancient times. We may also refer to the
constuction of the city of Pataliputra by
Ajaiasairu.

(註 3)Vinaya Pitakam. Vo1.III, p.46: arannam nama
thapetva gaman ca gamupacaran ca avasesam
arannam nama, also see the commentary on
pabbajeyyum and bhindeyyum in the same page
; see, Taisho, 22, p.6a20-21, p.244al7.

(註 4) For example, see Taisho 22, p.244a5.

(註 5) T.22, p.244a11ff.

(註 6) Anguttara Niksya, Vol.Iy , p.106 (PTS.
1958); Taisho 2, p.422c14ff.

(註 7) Han Yu Ta Tzu Tien (漢語大字典) Vo1. 4, p.
2791.

(註 8) Ibid, Vo1. 5,p.3255.
P.282

(註 9) Digha Nikaya, Vo1.I,p.14(PTS.1975).

(註 10)The Book of Gradual Sayings, Vo1. IV, p.70
(PTS, Ed)

(註 11) Pali-English Dictionary (PTS,1986), p.162.

(註 12) Han Yu Tzu Tien, Vo1.2, pp.1279,1315.

(註 13) Taisho Vo1.29. p.2c11ff.

(註 14) Much of the discussion in the present section
is taken from my article " The
schismaticmatters and the early Buddhist
literature", Journal of Research, Visva-
Bharati, Vo1.1,Part I, Humanities and Social
Sciences (1977), pp. 81ff.

(註 15) Cullavagga VII,5.2(PTS Ed.).
(註 16) Taisho 22, p.595a 17-19.

(註 17) Taisho 22, p.154a15-19.

(註 18) Taisho 23. p.25a25-29; p.266b18-22.

(註 19)Chang-so-hsing-fa has been rendered as krtyta
(Pali kiccayata); cf. the Sa rendering of
krtya as Ch'ang-so-hsing in krtyadhikarana,
Taisho, 23, p.251b5-6. In the Sanghavasesa
list of the same Vinaya we read Ch'ang-so-
yung is place of Chang-so-hsing. Evidently
these two terms are synonymous. V. Rosen's
restoration of the termas "dharmata" is
wrong (see, V. Rosen, Der Vinayavibhanga
zum Bhiksu-pratimoksa der Sarvastivadins.
Akademie Verlag. Berlin. 1955, pp. 69,70).
The Ma also reads Ch'ang-so-hsing for krtya.
Note also the occurence of the term krtyam
in the Dh list.

(註 20) Taisho, 24. p.155b4-5.

(註 21) See, under dutthulla in the Pali-English
Dictionary, PTS, also see, Sacred Books of
the East, Vo1. XVII p. 316, note No. 2,
Motilal Banarasidas 1965.

(註 22) This point can be demonstrated more
convincingly in such cases where all the six
versions of Vinaya have preserved detailed
account.

(註 23) See, Cullavagga, Pancasatika-khandhaka,
pp.408-9 (Nalanda Ed.)

(註 24) This term refers to those Formal Acts which
are performed for the re-establishment of
the concord of the Order (see, Mahavagga, X.
5. 14; X. 6).

(註 25) Ying-tso has been rendered as karaniya and
pu-ying-tso as akaragiya. Both ying-tso and
karaniya literally mean "what should be done
, duty". In Ma, for example. ying-tso has
been used in connection with the procedure
of a Formal Act, or in connection with the
rules, the non-observance of which leads to
Nissayakamma (see, Taisho, 22,p.424c20ff.).
Similarly. the word akaraniya has been used
with reference to sissayakamma in the
Cullavagga (see, Kammkhandhaka) We also
note that "cattari-akaraniyani" can refer to
four Parajikas. It appears that these two
terms in a general way refer to proper and
improper conducts of a monk and they do not
form separate
P.283
categories of their own.

(註 26) Taisho 22, p. 968a2-b11.

(註 27) Ibid. p.968b10-11.

(註 28)Two appendices are called T'iao-pu and pi-ni-
tseng-yi-ni-tseng-yi(t.22, pp.971ff; 990ff.)

(註 29) Cullavagga XI. 1.7.

(註 30) Ibid.XI. 1.9.

(註 3l) Taisho 22,p.190c29-191a14.

(註 32) Ibid. p.492c8-12.

(註 33) Taisho 24. p.407c1-408b1.

(註 34) The Pacittiyas (Nalanda Ed.), 1970. p.490,
Taisho 22, p.412b; Taisho, 22, p.713c;
Gustav Roth, Bhiksunivinaya (Patna, 1970)
p.300.