Necssity and sufficiency in the Buddhas causal schema

·期刊原文
Necssity and sufficiency in the Buddha's causal schema

By Jeffrey D.Watts
Philosophy East and West
32, no.4(October, 1982)
p.407-p.423
(C) by the University of Hawaii Press

--------------------------------------------------------------------------------
.

p.407

One of the most prominent features of the teachings
of the Buddha is his Twelvefold Causal Schema.(1) A
tentatively useful translation of the schema in
'anuloma' order (the order of 'arising') is the
following: On ignorance (avijjaa) depend
dispositional tendencies (sa^nkhaara) ; on
dispositional tendencies depends consciousness
(vi~n~naa.na) ; on consciousness depends the
psychophysical individual (naamaruupa) ; on the
psychophysical individual depend the six 'gateways'
(sa.laayatana [meaning the five faculties or organs
of sense perception plus the mind]); on the six
gateways depends contact (phassa); on contact depends
feeling (vedanaa) ; on feeling depends craving
(ta.nhaa); on craving depends grasping (upaadaana);
on grasping depends becoming (bhava); on becoming
depends birth (jaati); and on birth depends aging and
death (jaraamara.na). (2)
This schema has a number of important functions in
the context of the teachings. On the one hand, it
provides a causal explanation of the genesis of a
relatively 'new' psychophysical individual, conceived
not as an autonomous entity but as a congeries of
various causally arisen and interrelated phenomena.
(3) On the other, it often stands as a point of focus
for an illustration of the causal status of the
elements or 'nidaanas' with which it deals, in terms
of their relation(s)-individually and collectively-to
the human experience of suffering (dukkha). The
schema has still other functions in the teachings,
some of which will be mentioned in what follows.
A difficulty faced by scholars of Early Buddhism
has had to do with attempting to interpret the schema
in a way which would be consistent with the
possibility that suffering can be ended in one's life
and that the causal process culminating in rebirth
can be altered or thwarted in some way. That
suffering can be ended is a point regarding which
there has been no significant controversy, since two
of the Four Noble Truths of Buddhism assure us that
it can be ended, and that there is a way by which
this can be brought about.(4) The difficulty lies
only in squaring this possibility with the notions
that the schema is circular,(5) that each birth
somehow brings with it conditions which are causally
associated with suffering and rebirth, and that each
element of the schema apparently 'causes' the next to
arise.
In this article, I will reevaluate the Twelvefold
Schema. The importance of the schema is not in
question here. Rather, my main concern is with the
nature of the relations which the schema purportedly
expresses, both between contiguous nidaanas and also
between individual nidaanas, the arising of suffering
and the phenomenon of rebirth. It will be argued,
first, that there is an unfortunate ambiguity in the
textual expressions of the schema. Next, it will be
suggested that the ambiguity can be resolved, for the
most part, by means of introducing a useful
distinction between what I will call the 'normal' and
the 'nonnormal' causal processes. As I proceed, the
value of this distinction in terms of the development
of a relatively satisfactory interpretation of the
schema should become apparent.

p.408

The most common statement of the Twelvefold Schema,
in anuloma order, expresses the relationship between
two contiguous nidaanas by saying, for example
'avijjaapaccayaa sa^nkhaaraa, sa^nkhaarapaccayaa
vi~n~naa.nam,' and so forth. Now, 'paccaya' can have
such various meanings as "support," "requisite,"
"means," "reason," "grounds," "motive," and finally
"cause" or "condition." (6) While most scholars would
agree that, in the context of the schema, the term
should be rendered as 'cause' or 'condition,' there
seems to have been a failure to appreciate that it
may well make a difference whether we construe it in
one sense or the other.
This failure seems especially odd when it is noted
that there has been some further agreement that the
Early Buddhist causal theory seems most properly
analyzed in terms of a theory of necessary and
sufficient conditionship relations. (7) Surely it is
clear that, on such an analysis, it will be necessary
to distinguish between 'conditions' (construed in
terms of 'necessary' or sine qua non conditions) and
'causes proper' (construed in terms of groups of sine
qua non conditions which are, collectively,
'sufficient' for the arising of a given effect). (8)
In other words, it may well make a difference,
philosophically, whether we read the texts as saying
that a given nidaana in the schema is merely a sine
qua non condition for the arising of whatever nidaana
follows it, or if we instead understand each nidaana
to be sufficient for the arising of the next.
Unfortunately, the texts do not address this issue
directly.(9)
There is still another aspect to this ambiguity.
Whereas each of the relations in the schema is
expressed by means of 'paccaya' (or, in a few rare
cases, by a similar and similarly ambiguous term, for
example 'upanisaa' as at S.II.31), there is reason to
believe that not all of the relations so expressed
are in fact on equal footing. For example, there are
good reasons to think that the relation between
feeling (vedanaa) and the arising of craving
(ta.nhaa) is not of precisely the same nature as that
between grasping (upaadaana) and becoming (bhava).
More importantly, and in the context of that level of
explanation whereby the schema suggests relations
between various nidaanas and the presence of
suffering in one's life, it seems especially
important that we be able to distinguish between
those relations which are ones of sufficiency and
those which are merely of necessary conditionship.
But again, the texts do not provide us with any
direct information which would help to clarify
matters in this respect.
Neither of these features of the usage of 'paccaya'
in the schema seems to have captured the imaginations
of previous writers; and as a result, the tendency
has been to render the term across the board as
'cause-or-condition', or at least to use either of
these translations sometimes the one and sometimes
the other--rather indiscriminately and
inconsistently. In a sense, of course, there is an
easy explanation for this tendency. After all, the
Buddha could have used various terms to express the
relations between the nidaanas, and he could have
done so in such a way that at least some sort of
distinction might have been apparent between them.
Why he did not is open to speculalion. But the
important thing

p.409

to see for now is that the fact that the same term is
used throughout need not require us to assume that a
philosophical distinction between certain kinds of
relations cannot be applied legitimately to the
relations expressed in the scheme.(10)

For various reasons, it would appear that we could
say at the onset that each element in the schema must
be construed as at least a necessary condition for
the arising of whichever element is said to follow
it. First, we can consider the phrasing of the
cessation or'pa.tiloma' version of the schema. The
relations in this version are expressed in terms of
the cessation of one element following the cessation
of whichever element precedes it, for example,
'sa^nkhaaranirodhaa vi~n~naa.nanirodho' ('on the
cessation of dispositional tendencies, consciousness
ceases'). The idea here seems clear enough: The one
element (consciousness) cannot persist unless the
other (dispositional tendencies) is present. And this
is to say, at the very least, that the presence of
the first element is 'necessary' for the presence of
the second, that is, that it is a sine qua non
condition for it. Of course, this in itself doesn't
preclude the possibility that the first element might
be not only necessary for the second but sufficient
for it as well, since in either case, the removal of
the first would entail the removal (or 'cessation')
of the second. But at this initial stage of the
analysis, this possibility poses no serious
difficulties.
Next, we can consider some passages from the Diigha
Nikaaya. In the Mahaanidaana Suttanta (D.II.55 ff),
the Buddha gives a relatively detailed discourse on
the Twelvefold Schema to his disciple, AAanda. After
having stated the schema in the standard form, the
Buddha then explains to Aananda the manner in which
the relations between the nidaanas are to be
understood. Taking each relation in its turn, and
starting with that between birth (jaati) and old age
and death (jaraamara.na), the Buddha explains by
posing a formally similar question in each case. The
question regarding the relation between birth and old
age and death is essentially this: If there were no
birth of any kind whatsoever, would there be any
appearance (pa~n~aayetha) of old age and death?
AAnanda's not surprising answer is in the negative.
The Buddha then continues by saying, "Wherefore,
AAnanda, precisely that is the 'cause' (hetu), that
is the 'source' (nidaana), that is the 'origin'
(samudaya), that is the 'condition' (paccaya) of old
age and death, namely birth."(11)
Putting the matter in this way strongly suggests
that the relations between the nidaanas ought
provisionally to be construed as ones of necessary
conditionship. At the very least, by clarifying the
relations in the way that he did in this passage, it
would seem that the Buddha was assuming that it is
most consistent with a main purpose for which the
schema was formulated that each nidaana be understood
as a necessary condition for the next. It seems clear
that the Buddha wanted, among other things, to show
that the process of rebirth can be ended by bringing
at least one of the conditions expressed by the
nidaanas to a halt. Given this, it may well have
appeared to him to be sufficient to portray each
nidaana as no

p.410

more than a necessary condition for whichever one
followed it. After all, if what is wanted is the
cessation of suffering and rebirth, and if this can
be brought about by the cessation of any one of the
nidaanas, it could easily be thought to follow that
the question of whether the relations in the schema
are more than ones of necessary conditionship is
'practically' or soteriologically irrelevant.
However, when we turn to a discussion of the
relations apparently supposed to hold between
individual nidaanas and the experience of suffering
(dukkha), it becomes clear that the question is
relevant indeed, from even a purely practical point
of view. To illu -strate this, we can consider some
passages from the Suttanipaata.(12) The passages
speak of dukkha in relation to eleven conditions or
elements, seven of which have found their way into
the standard version of the schema. What is
interesting is that in each case, there is a phrase
which includes the term 'paccaya' and which
establishes a causal relation between the particular
element mentioned and dukkha. As should not be
surprising, the phrases are ambigous; but they do
seem to suggest that dukkha can be done away with
simply by the cessation of any one of the elements
being discussed. For example, we find the statement,
"Whatever dukkha arises is all in consequence of
avijjaa (ignorance)," followed by " .... but from the
complete destruction of avijjaa... there is no origin
of dukkha." (13) The same is said regarding the other
ten elements, those found in the standard version of
the schema being sa.nkhaara, vi~n~naa.na, phassa,
vedanaa, ta.nhaa, and upaadaana.
For our purposes, we can focus upon two of these
elements, phassa (contact), and vedanaa (feeling),
since these are two of the nidaanas regarding which
there seems to be some practical significance to a
distinction between necessary conditionship and
sufficiency in an analysis of the relations between
each of them and suffering. The significance is this:
If each of these elements is no more than a necessary
condition for suffering, then all that should be
required in order to bring suffering to an end is the
elimination of either of them. However, if they both
stand as sufficient for suffering, then it must
follow that both of them would have to be eliminated
in order for suffering to be destroyed.
Initially, it might appear that both phassa and
vedanaa could be sufficient, individually, for
dukkha. Indeed, the pa.tiloma form of the schema, by
speaking of the cessation of each one of the nidaanas
and by suggesting that it is by means of this that
the whole mass of suffering can be brought to an end,
appears to imply that nothing short of the cessation
of each of them will suffice. It is important to
note, however, that nowhere in the texts is it said
explicitly that dukkha can be destroyed if and only
if vedanaa and phassa are destroyed. In fact, if such
a strong claim were made explicitly, it would be in
direct conflict with other claims made elsewhere in
the texts.
Let us consider vedanaa and examine an important
passage in the Itivuttaka, wherein a distinction is
made between the nibbaana of an arahant (that is, an
enlightened one) who has not yet died and that of an
arahant whose present life has come to an end with
the break-up of the body. With regard to the former,
it is

p.411

said that he is one who has done what has to be done,
has laid aside the burden, achieved the noble goal,
destroyed the fetters, and so forth. At the same
time, however, he is said to retain his senses,
through which, since they are not yet destroyed, he
experiences sensations and has feelings.(14)
Furthermore, in the Diigha Nikaaya (D.II.157), there
is a passage which reproduces a statement made by the
monk Anuruddha, to the effect that when the Buddha
died, in his last moments he felt pain.
What these passages seem to be saying is that
enlightened ones are capable of feeling sensations, a
phenomenon described in the schema in terms of
relations between the three nidaanas of sa.laayatana
(the six faculties of sense perception, including the
mind), phassa (contact), and vedanaa (feeling). This
point is especially relevant in the context of the
present discussion when we consider certain other
textual passages, two of which will now be cited.
Both occur in the Sa.myutta Nikaaya, and the first
can be translated as "The mortal [or 'man'] who
delights in nibbaana is freed from all dukkha"
(S.I.38).(15) The second is found in a verse in which
it is said that one who is lax or of little strength
cannot win nibbaana, "... [which is] the emancipation
from all dukkha (sabbadukkhapamocana.m) " (S.II.
278) .(16) That 'sabbadukkhapamocana' is to be
understood as a characteristic of nibbaana-or, if we
prefer, a characteristic of one who has attained
nibbaana-is clear from the syntax of the verse. But
what is most important to see here is that if it is
true that one who has attained nibbaana can-and,
according to the evidence, does-experience feelings,
and if it is also true that attainment of nibbaana
entails at least a release from all dukkha, it
follows that the experience of feeling (vedanaa)
cannot be taken as sufficient for the presence of
suffering.

There is an unfortunate tendency, both in the texts
and in the writings of some contemporary Buddhist
exegetes, to speak as if there is something
ineluctable about the relationship between the senses
(and feeling that comes as a result of their
functioning) and the arising of lust, craving
(ta.nhaa), and most of whatever else is believed to
be a source of suffering. This tendency is exhibited
in the textual formula for the perceptual process
which says basically that what one perceives, one
reasons about; and what one reasons about, one
becomes obsessed with. The implication seems to be
that perceiving inevitably leads to obsessions. (17)
Similarly, passages such as those from the
Sutta-nipaata, discussed earlier, can easily be
construed as suggesting that such things as contact
and feelings necessarily entail (or are necessarily
followed by) the arising of suffering.
I want now to suggest, however, that a great deal
of confusion and unclarity can be forgone with regard
to passages such as these by means of introducing a
distinction which neither the Buddha nor the majority
of writers on Buddhism seem to have recognized as
useful in analyses of the Buddha's causal theory. The
distinction is between what can be called the
'normal' and the 'nonnormal' causal processes.(18)
The normal process would be that described in the
texts in terms of the anuloma form of the Twelvefold
Schema. What it means to say that this

p.412

process is 'normal' is just that it is that which, in
a general way, accounts causally for the 'normal'
human condition, the most salient feature of which,
according to the Buddhist teachings, is the
experience of suffering. The normal process is that
which involves such things as craving (ta.nhaa),
grasping (upaadaana), and becoming (bhava) and which
leads eventually to another birth (jaati), another
spin of the wheel of sa.msaara.
Now, we know that the complete cessation of
suffering is a possibility. Further, we know that the
cessation of suffering is something that comes about
causally. (19) Finally, since the evidence seems
clearly to show that, on the one hand, an involvement
in the causal process expressed in the anuloma
version of the schema leads to suffering and rebirth
while, on the other, some individuals have by some
means been able to abolish suffering and put an end
to the rebirth process in this life, it follows that
there must be a causal process which is perhaps
slightly but nevertheless importantly different from
the normal one. It is this process which I will call
'nonnormal'.
Admittedly, this distinction between a normal and
nonnormal causal process is somewhat artificial; but
this is only because the texts do not make the
distinction directly. I would argue, however, that
some such distinction is implicit in the Buddha's
teachings; for, without it, the analysis of the human
condition in terms of the Twelvefold Schema, coupled
with the Buddha's general causal theory, would leave
the individual stuck, as it were, in sa.msaara with
no apparent 'way' to escape. Indeed, I should say
that it is only by allowing for the possibility of a
disengagement from the normal causal process and an
engagement in a nonnormal one that the complete
cessation of suffering could itself be possible.
There is, in addition, a different sort of
advantage to be had by recognizing an implicit
distinction between a normal and a nonnormal causal
process in the Buddhist teachings. If we accept that
there are (at least) two rather different causal
processes in which one can be engaged, the one
leading to continued suffering and rebirth and the
other culminating in the cessation of both of these,
then we can begin to see how the ambiguities
discussed with regard to the usage of 'paccaya' might
be resolved, in specific instances. For, we can
entertain the possibility that the causal talk in the
texts cannot be understood properly as being
univocal. It would seem, in fact, that many of the
causal claims made in the context of the schema, as
well as claims dealing with the purported relations
between elements of the schema and suffering, are
best understood as expressive of only one of the
causal processes, namely, the normal one. If this is
so, then when it is said, for example, that whatever
suffering arises has phassa (contact) as its
'cause/condition' (paccaya), we can take this claim
as descriptive of what goes on in the normal causal
process, but not as expressing an ineluctable
relationship between the presence of phassa and the
arising of suffering.
Putting the matter in this way adds an interesting
new twist to the question about whether the relations
between specific nidaanas and suffering in general
ought to be understood as ones of necessary
conditionship or ones of sufficiency.

p.413

By distinguishing between a normal and a nonnormal
causal process, and by suggesting that passages
linking such things as contact (phassa) with
suffering should be understood as expressive of the
normal process, we are allowing that there may well
be something about that process which establishes a
certain ineluctability to such relations, whereas in
the nonnormal process this will not be the case. In
other words, it looks as though it could be argued
that in the normal process, such things as contact
and feeling (vedanaa) do stand as sufficient for
suffering, whereas in the nonnormal process they do
not. We know that vedanaa at least cannot be
sufficient for suffering under all conditions because
we have seen that there are those who have succeeded
in ending suffering while still being subject to
feelings. Concurrently, there are those troublesome
passages which do suggest that vedanaa is always
sufficient for suffering; the problem is how to
square these apparently divergent notions. It is
precisely here that the distinction between two
causal processes should be of much philosophical
value.

On the preceding analysis, it will be correct to
say that, given the normal causal process, suffering
will 'inevitably' follow phassa and vedanaa. It would
be misleading, however, to say categorically that
these nidaanas, either individually or together, are
'sufficient' for suffering. The point is that when we
say that X will be followed 'inevitably' or
'necessarily' by Y in the normal causal process, but
that Y will not necessarily follow X in the nonnormal
one, we are not speaking of X and Y and the relation
between them in the abstract. Rather, we are saying
something about the normal process itself, considered
on the whole. We are certainly speaking of
sufficiency; but we are saying no more than that
there are conditions present when the normal process
is functioning which set up a 'causal background', as
it were, against which the arising of certain
elements will necessarily be followed by the arising
of certain others. Also, we are saying that, given
that background, the presence of, for example, phassa
will necessarily lead to suffering, not because of
something intrinsic to phassa itself, but because of
a particular determinate causal role which the
phenomenon of contact comes to play in the normal
causal process.
If it is true that in the normal causal process,
phassa necessarily leads to suffering but that in the
nonnormal one it does not, the natural question to
ask next is just what is it that is present in the
normal process-but absent in the nonnormal one-such
that with the addition of phassa there will be
conditions sufficient for the arising of suffering.
Before attempting to answer this question, it is
worth noting that virtually all expressions of the
Twelvefold Schema in anuloma order deal with what I
am calling the normal process. While the patiloma
statements of the schema can be seen as reflecting-
albeit obliquely-that I have called the nonnormal
process, it is unfortunately the case that this
process is nowhere expressed as directly as is the
normal one. There are several passages in which
liberation is said to be 'causally associated' with
various conditions, for example, faith (saddhaa),

p.414

concentration (samaadhi), energy (viriya), and so
forth;(20) but there is no one standard statement of
specific causal elements involved in the nonnormal
process comparable to the schema in anuloma form. The
patiloma form does not parallel the anuloma in this
respect. It merely expresses how things go once the
alteration from the normal to the nonnormal causal
process has been accomplished.
Nevertheless, it is possible to analyze the texts
in such a way that the primary differences between
the normal and nonnormal causal processes can be
illustrated. The analysis begins with an examination
of the twelve nidaanas of the schema and then
proceeds by determing which, if any, of them can be
said to be unique to the normal process. Clearly, not
all of the nidaanas can be unique to the normal
process, for reasons which will be given shortly. But
it is equally clear that not all of the nidaanas in
the schema can be construed as causal elements in the
nonnormal process, which is to say that not all of
these elements can figure into a causal explanation
of the life and general condition of an enlightened
one.
Of the twelve nidaanas, those which clearly cannot
be unique to the normal causal process will be those
which must remain as causal elements in a relatively
exhaustive analysis of the life-process of an
arahant. Without any difficulty, we can see
immediately that there are at least four out of the
twelve which must so remain, and these would be
naamaruupa (the psychophysical individual) ,
sa.laayatana (the six sense faculties including the
mind); phassa (contact, both physical and mental),
and vedanaa (feeling, either pleasant, painful, or
neutral). Each of these nidaanas describes (or refers
to) a particular aspect of a human being in general,
whether enlightened or not. We have seen that an
enlightened one is said to be subject to phassa and
vedanaa, and both of these phenomena presuppose the
presence of naamaruupa and sa.laayatana. In other
words, neither phassa nor vedanaa could be present
unless there were a psychophysical individual endowed
with sense organs or faculties. And since both phassa
and vedanaa can figure into a description of what
goes on in the life -process of an enlightened one,
it follows that both naamaruupa and sa.laayatana can
as well.
One thing which follows from this is that none of
these four nidaanas can be understood, either
individually or collectively, as sufficient for the
arising of dukkha (suffering) . This is because
arahants, whose lives are analyzable partly in terms
of them, are said to be rid of dukkha altogether,
which is at least to say that no causal factors which
play a part in the determination of their
life-processes (once they are enlightened) are
sufficient for the product -ion or sustaining of
suffering. This is not to say that these four
nidaanas are in no way causally involved in the
arising of dukkha. On the contrary, they seem to be
very much involved in that, as many passages in the
texts attest; and it is for this reason, we may
assume, that it was thought to be important to
include them in the analysis of the normal causal
process, as stated in the anuloma form of the schema.
All that is being argued at present is that it must
be mistaken, for the reasons given earlier, to see
dukkha-and, we should add, rebirth-as following
necessarily from the presence of these four causal
factors.

p.415

Next, it must be asked whether any of the remaining
eight nidaanas can be unique to the normal process,
and the way to determine this will again be to
ascertain which, if any, of them can be said to
remain as causal factors in the life of an arahant.
Whichever of the nidaanas cannot apply in a causal
explanation of the life-process and general condition
of an arahant will, for that reason, fall necessarily
into the group of factors which are unique to the
normal causal process.
Limitations on space require that the following
analyses be brief, which is unfortunate since several
of the remaining nidaanas are complex and admit of
various interpretations. However, it is possible to
make a few general observations regarding each of
them which will suggest where they ought to be placed
in our categorization.
The first nidaana occurring in the standard
statements of the Twelvefold Schema is 'avijjaa,'
most often translated as 'ignorance'. In the context
of the schema, it is important to notice that
'avijjaa' does not mean 'ignorance' in a general
sense. Rather, it is said to be a certain kind of
ignorance, or ignorance of particular 'facts'. The
texts make it relatively clear as to what these facts
are. For example, at M.I.54, it is said that avijjaa
is 'nescience' (a~n~naa.na) concerning dukkha
(suffering), its origin (samudaya), its cessation
(nirodha) and the course leading to its cessation
(nirodhagaaminii pa.tipadaa) . In other words,
ignorance, in the context of the schema, is precisely
a lack of understanding of the Four Noble Truths of
Buddhism. The texts indicate that it is through a
failure to understand these Four Noble Truths that
individuals wander along the path to rebirth; but, it
is further said, when these are understood, the
craving for existence (or 'becoming' [bhaveta.nhaa])
is destroyed, that which leads to becoming
(bhava-netti) is exhausted and, therefore, there is
(or will be) no future birth (n'atthi
punabbhavo).(21)
Now, given that avijjaa amounts primarily to a
failure to understand the Four Noble Truths, and that
this failure is said to entail wandering along the
path to rebirth, it follows that one who has
succeeded in getting off this path cannot have
avijjaa as a causal factor in the determination of
his life and condition. And since an arahant is one
who has put an end to rebirth, (22) it follows that
avijjaa cannot be an element in a causal explanation
of his present condition. Of course, however, since
every birth is a rebirth, and since every arahant has
been born, there is a sense in which an exhaustive
explanation of the very presence of an arahant, qua
human being, must refer to avijjaa. This is because
avijjaa is a causal factor in every rebirth, as its
position in the schema would seem to suggest. But if
escape from rebirth is possible, then it must also be
possible for avijjaa to be abandoned in this life,
and it is because of this that we can say with
confidence that avijjaa cannot be a causal factor in
the nonnormal process. Therefore, avijjaa must be
seen as unique to the normal causal process.
The second nidaana in the schema, sa^nkhaara, is
difficult to analyze. 'Sa^nkhaara' has numerous
meanings in the texts and therefore requires
different translations in different contexts. For
various reasons which I cannot go into here, the term

p.416

seems best translated in the context of the schema as
'dispositional tendencies which involve or are
determined by raaga (lust or passion), dosa (hatred
or aversion), and moha (confusion)'.(23) Now, in
several places in the texts, there is the phrase
'sabbe sa^nkhaaraa dukkhaa': 'All sa^nkhaaras are
suffering (or sufferingfraught)'. (24) In addition,
the texts occasionally speak of an attainable state
called 'sabba-sa^nkhaara- samatha' which means 'the
pacification of all the sa^nkhaaras'.(25)
The contexts in which this state is mentioned
indicate that it is very close to, if not in fact
equivalent with, nibbaana.(26) I need not insist upon
an equivalency to make my point, however, for even if
nibbaana is something more than the pacification of
all sa^nkhaaras, it is clear that the sa^nkhaaras are
intimately connected with suffering. Given this, it
is reasonable to argue that they cannot correctly be
construed as causal factors in the life of one who
has brought all suffering to an end. Therefore, since
it is only the normal causal process which is
involved with the production and sustaining of
suffering, it follows that sa^nkhaaras must be unique
to that process.
Following 'sa^nkhaara' in the schema is
'vi~n~naa.na'. Again, this is a term which requires
different translations in different contexts. Most
generally, 'vi~n~naa.na' means 'consciousness'; but
not a few scholars have argued that in the schema,
the term must be understood as meaning 'rebirth
consciousness', which is a consciousness of a
specific and limited sort.(27) Briefly put,
'vi~n~naa.na' in this context can be understood as
consciousness marked by, and in a sense constituted
in terms of, ignorance (avijjaa). It is that which
provides for a continuity from one life to the next,
not as a perduring entity of any sort, but as a
causal process determined in part by factors present
in the last moments of one life-process and
determining, in turn, the arising of a 'new'
psychophysical individual.(28)
As with avijjaa, vi~n~naa.na is clearly a sine qua
non condition for each and every birth; and since
every individual-including an arahant-has been born,
it follows that a causal analysis of his coming-to-be
in this life will necessarily make mention of
vi~n~naa.na qua rebirth consciousness. Indeed, as
'rebirth consciousness', the presence of vi~n~naa.na
presupposes the presence of avijjaa and sa^nkhaara as
causal factors in its own arising.
It is clear from the texts that the attainment of
nibbaana carries with it a knowledge that there will
be no further births, that this particular life will
be the last. Both avijjaa and sa^nkhaara have been
eliminated as causal factors in the life of an
arahant, and with their elimination has come the
cessation of vi~n~naa.na qua rebirth consciousness,
the cessation of that mysterious process which
normally continues at the break-up of the physical or
material elements of the psychophysical individual.
Therefore, since the nonnormal process does not lead
to any further birth, and since as long as
vi~n~naa.na (in this special sense) is present a
future birth is sure to come, it follows that
vi~n~naa.na must be understood as a causal factor in
the normal causal process only, but not in the
nonnormal one.
The next four nidaanas in the schema-naamaruupa,
sa.laayatana, phassa, and vedanaa-have already been
discussed. Following 'vedanaa' are 'ta.nhaa',
'upaadaana',

p.417

'bhava', 'jaati' and 'jaraamara.na'. Of these, it is
quite easy to see that at least bhava and jaati
cannot be construed as causal factors in the
nonnormal process. 'Bhava' means 'becoming', and a
proper understanding of its explanatory function in
the schema depends upon the recognition that the
schema accounts not for one single life-process but
actually for three related ones.(29) The first three
nidaanas of the schema-avijjaa, sa.nkhaara, and
vi~n~naa.na- apparently refer to causal factors
present in an immediately past life, and they
'explain' the arising of a relatively new naamaruupa
or psychophysical individual. 'Vi~n~naa.na', as it
were, marks the transitional point from the past to
the present life. Likewise, 'bhava' marks the
transitional point from one's present life to another
one; and for this reason, it is followed in the
schema by 'jaati,' or 'birth', which is always a
rebirth.
The general idea is elear enough: Whenever bhava is
present, another birth is sure to follow. Indeed,
'bhava' is listed among the aasavas (influxes,
impediments), and in several passages arahantship (or
the attainment of nibbaana) is spoken of as
intimately connected with their destruction
(aasavaana.m khaya). (30) Given this, it is clear
that the nonnormal causal process cannot have bhava
as a causal element in its determination. That this
process cannot either have jaati as a determinant in
it should be obvious without considering anything
more than that the nonnormal process does not lead to
rebirth.
'Ta.nhaa' ('craving') receives special treatment
in the texts. Because of the nature of this
treatment, several scholars have remarked that
ta.nhaa appears to be an integral factor in the
production and sustaining of suffering.(31) Indeed,
in more than one place the cessation of ta.nhaa is
specifically linked with the attainment of nibbaana;
and at S. III.190, we find the straightforward
assertion that the destruction of ta.nhaa is
nibbaana: "Ta.nhakkhayo hi Raadha nibbaanan ti."
Further, there is an entire chapter of the Dhammapada
(chap. XXIV, "Ta.nhaavaggo" )which is devoted to a
discussion of how ta.nhaa affcets the life of an
unenlightened one, and how it leads to continued
suffering unless and until it is rooted out and
completely destroyed.
The point is clear: Ta.nhaa can-and must-be
destroyed if one is to attain nibbaana. Ta.nhaa is
often spoken of as a 'root' ('mula') of
suffering;(32) and as such, its presence is clearly
causally incompatible with the attainment of
nibbaana, which entails the complete cessation of
suffering. Therefore, since the nonnormal process is
not involved with the production or sustaining of
suffering, it follows that ta.nhaa cannot be
construed as a causal factor in that process.
'Upaadaana' ('grasping') requires little individual
treatment in this context, since it is generally
agreed that it is little more than ta.nhaa which has
developed to a particularly high degree.(33) Thus,
much of what ean be said regarding ta.nhaa will also
hold true for upaadaana; and in any case, because
upaadaana appears to be a sort of 'species' of
ta.nhaa and because ta.nhaa cannot be a factor in the
nonnormal causal process, it follows that neither can
upaadaana.
The final nidaana in the schema'jaraamara.na'
('old age and death')-is peculiar. Listing it as a
causal factor at all has seemed to at least one
scholar to

p.418

make about as much sense as calling day the cause of
night.(34) However, when we recall that the schema is
supposed to be circular and that it accounts for
three contiguous lives, we can begin to see why 'old
age and death' should have been included therein. In
the schema, 'jaraamara.na' immediately follows
'jaati', or 'rebirth'. Starting with 'bhava', the
final three nidaanas refer not to one's present life,
but to a future one; and so, insofar as the anuloma
version of the schema is concerned, these three count
as potential causal elements, not as actual ones in
one's present life. This, of course, is not to deny
that everyone who is presently living will undergo
aging and death; but it is to say that not everyone
will necessarily be reborn again. And if one is not
to be reborn againly virtue of having attained
nibbaana in this life--then a fortiori he will not be
subject to aging and dying in a future life.
Therefore, because the nonnormal causal process
entails no future birth, it follows that
'jaraamara.na', taken as meaning 'old age and death
in a future life', cannot be counted as a factor in
that process.

What the preceding anlaysis suggests is this:
Whereas in the normal causal process all twelve of
the nidaanas function as causal factors, the
nonnormal process is one in which several of them
have been eliminated, not only as causal factors but
as the effects of causal factors as well. Those of
the twelve nidaanas which do not figure into the
nonnormal process are avijjaa (ignorance), sa^nkhaara
('dispositional tendencies' in the special sense),
vi~n~naa.na (rebirth consciousness), ta.nhaa (craving
), upaadaana (grasping), bhava (becoming), jaati, and
jaraamara.na (a future birth and aging and death in a
future life). Those nidaanas which appear to be
common to both the normal and the nonnormal processes
are naamaruupa (the psychophysical individual) ,
sa.laayatana (the six sense faculties) , phassa
(contact), and vedanaa (feeling).
We've seen that the eight nidaanas peculiar to the
normal causal process are all intimately connected
with the arising and sustaining of suffering, and
that the four remaining nidaanas apparently have no
connection with suffering, at least in the nonnormal
process. However, this is not to say that they have
no connection with suffering at all. As long as the
normal process is still at work-unless and until the
alteration from the normal to the nonnormal process
has been accomplished-even these intrinsically
innocuous causal factors conduce to suffering.
The reason for this, I suggest, is because of the
presence of avijjaa in the normal causal process.
Avijjaa, as it were, colors the entire normal
process, establishing a causal background against
which these nidaanas take on special causal roles.
For example, as long as avijjaa remains, vedanaa
(feeling) gives rise to ta.nhaa (craving) , not
because of anything intrinsic to the phenomenon of
feeling itself, but because of a mistaken grasp of
the nature of the object which has given rise to the
feeling, as well as a mistaken understanding of the
nature of the individual and of the feeling itself.
While an analysis of the nature of these 'mistakes'
would be of value, we are not here concerned with
them perse as much as we are with the mere

p. 419

fact that they exist in the normal process, and that
they would appear to be reducible to avijjaa in one
form or another.
The sense in which these four nidaanas are involved
in the production and sustaining of suffering should
be clear. In order for there to be the experience of
suffering, there must first be the presence of the
psychophysical individual (naamaruupa), the six sense
faculties (sa.laayatana) , contact (phassa) with
objects-either physical or mental or both-and feeling
(vedanaa) arising as a result of that contact. In
other words, each of these nidaanas counts as a
necessary or sine qua non condition for the arising
and sustaining of suffering. None of them, however,
can be construed as sufficient for suffering, the
evidence for this being that all four are at work in
the life-process of one who has attained nibbaana and
who has, therefore, put suffering to an end.
In fact, it would appear that no one of the twelve
nidaanas can be understood as sufficient for the
arising of suffering, even though the texts-and
particularly the Sutta-nipaata passages discussed
earlier -seem to imply just the opposite. There is a
way to understand such strong implications of
sufficiency, however, and it has to do with
recognizing that the presence of certain crucial
causal factors in the schema 'presupposes' the
presence of certain others. For example, the presence
of ta.nhaa presupposes not only the presence of
naamaruupa, sa.laayatana, phassa, and vedanaa, but
avijjaa, sa.nkhaara, and vi~n~naa.na as well. In
other words, all of these count individually as
necessary conditions for the arising and sustaining
of ta.nhaa; and so, whenever we find an instance of
ta.nhaa, we can be assured that these other nidaanas
are also present. In addition, we can see that these
seven nidaanas count as sufficient, collectively, for
the arising of ta.nhaa; and because of this, when it
is implied that ta.nhaa is sufficient for suffering,
we must understand that it is not ta.nhaa itself in
isolation from its necessary and sufficient
conditions which is being spoken of.
Only by distinguishing between the normal and the
nonnormal causal processes, and by recognizing that
the majority of the causal claims in the texts
presuppose the normal process, can we see how it can
be possible for some of the nidaanas to be apparently
sufficient for suffering but actually not, when
considered in themselves. In essence, what has just
been said regarding ta.nhaa can also be said of each
of the nidaanas which precedes it in the schema. For
example, within the normal process, the presence of
vedanaa (feeling) presupposes the presence of avijjaa
; and in a certain sense, this will be true even
within the nonnormal process in that, for there to be
vedanaa at all, there must have been a birth; and for
there to have been a birth, there must have been
ignorance. But whereas avijjaa is in this sense
necessary for the arising of vedanaa-that is,
necessary for there to be a context in which the
arising of vedanaa is possible-it does not follow
that avijjaa is necessary for the arising of vedanaa
in specific instances, nor for its sustenance once it
has arisen.
The most significant nidaana in the normal process,
then, appears to be avijjaa.

p.420

Within that process, I have argued, avijjaa works in
conjunction with the other nidaanas and establishes
sufficiency relations between them and the arising of
suffering. Thus, whereas the individual subject to
avijjaa will experience suffering as an inevitable
result of contact and feeling, he who has done away
with avijjaa altogether will not. For the ignorant
one, feeling will always give rise to craving,
craving will grow to become grasping, grasping will
lead to becoming, a future birth and a repeat of the
cycle. But for one who has escaped the normal process
and entered into the nonnormal one (primarily by the
elimination of ignorance), contact and feeling become
innocuous. This is not to say that they no longer
have any causal function, but only that the function
has become importantly different. The main difference
with regard to vedanaa is that in the nonnormal
process, it no longer entails ta.nhaa nor suffering,
whereas in the normal one, it does. And this is
because vedanaa presupposes avijjaa in the normal
process, but not in the nonnormal one.
Given that the notion of the presupposition of
prior causal elements seems to be important in an
assessment of the causal relations expressed in the
schema, there are at least two things that can be
said about how to construe 'paccaya'. The first is
that when a relation between two contiguous nidaanas
is considered in the abstract, that is, apart from a
specification as to whether prior elements are being
presupposed or not, 'paccaya' must be taken as
indicating no more than a relation of necessary
conditionship. In no case does the presence of a
single nidaana alone stand as sufficient for the
arising of any other. However, when it is the normal
causal process which is being spoken of, that is,
when the presence of avijjaa as a causal factor is
being presupposed, there is a special sense in which
'paccaya' can be construed as indicating a
sufficiency relation.
In the case of relations between individual
nidaanas and suffering, the same thing can be said;
that is to say, when vedaana, for example,
presupposes avijjaa (as it does in the normal process
), its presence is sufficient for the arising of
suffering. However, when avijjaa has been eliminated,
the sufficiency relation no longer obtains. The case
with some of the nidaanas might appear to be
different in this respect. The presence of ta.nhaa,
for example, will appear always to be sufficient for
the arising and sustaining of suffering, as will the
presence of sa.nkhaara, vi~n~naa.na (qua 'rebirth
consciousness'), upaadaana, and bhava. But the reason
for this, once again, is because the presence of each
of these nidaanas always presupposes the presence of
avijjaa. Whereas naamaruupa, sa.laayatana, phassa,
and vedanaa are possible only by virtue of a birth
generated by avijjaa, these four nidaanas can remain
once avijjaa has been eliminated. The other nidaanas,
however, cannot.
It may be disturbing to some to find that such an
apparently simple formula as the Twelvefold Causal
Schema actually requires some effort in order to be
understood. It would be easy to argue that the
expression of the schema in the texts is woefully
imperfect and perhaps even inconsistent with other
central features of the Buddha's teachings. But if
one focuses first upon those aspects of the teachings
which clearly imply that what I have called the
normal causal

p. 421

process cannot be the only possible one, and then
proceeds to look again at the schema, it should
become apparent that this formula is not so
problem-laden after all. The interpretation which I
have given the schema here is by no means the only
possible one which could exhibit it as consistent
with the rest of the Buddha's teachings. But it is
hoped that the preceding analyses and arguments, if
not unassailable, will at least generate further
philosophical discussions of this important aspect of
Early Buddhist thought.

NOTES

1. Following the early (1844) lead of the French
scholar Eugene Burnouf, most writers on Buddhism have
referred to the Buddha's formula as the 'chain' of
causation (Burnouf's term was `enchainement' ). The
metaphor, however, is not entirely adequate,
primarily because it suggests that e various elements
in the formula are on equal footing, as are the links
in a chain, in a certain sense. More importantly, it
suggests that if one of the elements were eliminated,
the entire schema would in some sense collapse or
fall apart. There is some truth to this with regard
to the Buddha's schema; but at the same time, the
idea is misleading. As will be argued in what
follows, it is possible for some of the nidaanas in
the schema to remain and to retain some causal status
even when others have been eliminated. Given this, I
prefer to speak of the formula as 'the Twelvefold
Causal Schema', or simply the scheme' throughout.
2. The schema is stated in many places in the
texts, e.g., at Majjhima I.263 and III.63, Diigha
II.31, A^nguttara V.184, etc. (From this point on,
references to the Nikaayas will be indicated by upper
case initials, the volume number and then the page,
with periods in between. "M" will refer to the
Majjhima Nikaaya, "D" to the Diigha, "S" to the
Sa.myutta and ''A" to the A^nguttara. Thus. "M.I.
263" should be read as "Majjhima Nikaaya, vol. I,
page 263.")
3. Regarding this function of the schema, see David
J. Kalupahana's Buddhist Philosophy: A Historical
Analysis (Honolulu, Hawaii: The University Press of
Hawaii, 1976) (hereafter cited as Kalupahana,
History), p. 31, and K. N. Jayatilleke' s Early
Buddhist Theory of Knowledge (London: George Allen &
Unwin Ltd., 1963) (hereafter cited as EBTK), p. 451.
4. The Four Noble Truths of Buddhism (cattaari
ariyasaccaani) (S.V.420 ff) are (1) dukkha ([there
is] suffering), (2) dukkhasamudaya ([there is] the
arising or origin of suffering), (3) dukkhanirodha
([there is] the cessation of suffering ) and (4)
dukkhanirodhagaami.nii pa.tipadaa ([there is] a way
leading to the cessation of suffering).
5. Regarding the circularity of the schema, see
Kalupahana's History, p. 31.
6. See the Paali Text Society's Paali-English
Dictionary (London: Paali Text Society, 1972),suh
voce.
7. See, for example, Kalupahana's Causality: The
Central Philosophy of Buddhism (Honolulu, Hawaii: The
University Press of Hawaii, 1975) (hereafter cited as
Causality) , pp. 54, 56, 58, and 59. See also
Jayatilleke's EBTK, pp. 423, 446, and 447, and Mrs.
Rhys Davids' Buddhism (London: Williams and Norgate,
s.d.), p. 88.
8. Regarding this distinction, see Richard Taylor's
article titled "Causation" in the Encyclopedia of
Philosophy (New York: Macmillan, 1967), vol. 2, p.
63. See also A. J. Ayer's The Problem of Knowledge
(Harmondsworth: Penguin Books Ltd., 1956), p. 171.
9. Indeed, there is no technical distinction made
explicitly in the texts between necessary and
sufficient conditions. But an analysis of what has
been called the General Formula of Causation ("When
this is present, that comes to be; on the arising of
this, that arises. When this is absent, that does not
come to be; on the cessation of this, that ceases")
(D.II.31, A.V. 184, S. II.28, M.I.263, etc.) and the
Four Characteristics of the Causal Nexus (tathataa,
avitathataa, ana~n~nathataa and idappaccayataa)
(S.II.26) strongly suggests that the Buddha's causal
theory implies such a distinction.
10. The question whether, and to what extent, the
Buddha's teachings can properly be construed as
'philosophical' is controversial and interesting in
its own right. The point here is simply that

p.422

whatever the answer to this question turns out to
be, the teachings involve abundant claims which are
philosophically pregnant... claims which raise
questions that are philosophically interesting
Clearly, the Buddha neither anticipated nor attempted
to answer many of these questions. This. however,
does not entail that it is illegitimate or
misrepresents his teachings for us to pursue them.
Indeed, by studying the teachings on the whole, it is
possible to surmise how the Buddha would most likely
have answered several of these philosophical
questions, had he only considered them.
11. As is well known, it is standard practice in
Sanskrit and Paali literature to use several
apparently synonymous terms to express a single idea.
While the scare-quoted translations given here for
the terms in this passage are all legitimate possible
meanings, it is likely that they were all intended to
carry the same sense. But whether it would be best to
render them all by means of a single English term or
to give each a slightly different thrust, it seems
clear that a precise philosophical distinction
between 'cause' and 'condition' cannot be made on the
basis of passages such as these alone.
12. Sutta-nipaata 724 ff.
13. Ibid., 729.
14. Itivuttaka 38 f. For a translation and
discussion of this important passage, see
Kalupahana's History, p. 70.
15. The passage occurs in verse form, the stanzas
of which will be separated here by means of single
and double slashes: "Ki.msu dutiya.m purisassa
hoti/ki.msu c-ena.m pasaasati/kissa caabhirato
macco/sabbadukkhaa pamuccatiiti//Saddhaa dutiyaa
purisassa hoti/pa~n~naa c-ena.m pasaasati/
nibbaanaabhirato macco /sabbadukkhaa pamuccatiiti. "
16. Again, the passage is in verse: "Nayida.m
sithilam aarabbha/na yidam appena thaamasaa/
nibbaanam adhigantabba.m / sabbadukkhapamocana.m. "
17. The formula for the perceptual process is
stated most clearly at M.I.111-12. A nontextual
example of the tendency I am speaking of can be found
in Paravahera Vajira~naa.n Mahaathera's Buddhist
Meditation in Theory and Practice (Kuala Lumpur:
Buddhist Missionary Society, 1962): "Beings that
dwell in the sense-world are dominated by desire
which operates through the five senses and so weaves
the whole fabric of physical existence" (p. 166).
Given this, it seems to follow either that arahants
do not dwell in the sense-world- assertions that they
still retain their senses and have feelings
notwithstanding-or that they, like unenlightened
ones, are 'dominated by desire'.
18. Kalupahana seems to have anticipated this
distinction, although he does not develop it fully.
On page 139 of his Causality, for example, he argues
that "the elimination of ego-consciousness by the
development of insight can change the normal process
of perception." And further, "in an enlightened one,
perception simply does not generate obsessions and
the consequent suffering." While Kalupahana is
speaking here of the perceptual process, it is clear
that a distinction between a normal and nonnormal
causal process can be worked out along the same
lines.
19. In various passages (e.g., S.III.69 and S.II.
30), purity, impurity, and liberation itself are said
to be 'causally associated' or 'with cause and not
without cause'.
20. For example, at M.I.473ff, II.173ff; A.I.3, II
and 152; V.150; S.II.195; and V.200.
21. D.II.90.
22. S.I.174; V.422; M.I.184; etc.
23. For a useful analysis of the notion of
'sa^nkhaara', see Kalupahana's "The notion of
suffering in early Buddhism compared with some
reflections of early Wittgenstein" (Philosophy East
and West 27, no. 4. (October, 1967): 423-431). See
also S.IV.359 and 373, where a connection is made
between the sa^nkhaaras and raaga, dosa, and moha.
24. A.I.286 and Dhammapada 278, for example.
25. S.I.136, III.133, IV.362; A.I.133; D.II.36;
M.I.167; etc.
26. In fact, in the Mahaasudassana Jaataka
(translated by Rhys Davids in the introduction to his
translation of the Mahaasudassana Suttanta (Dialogues
of the Buddha) (London: Paali Text Society, 1971),
Part II, p. 195), we find the equation being made
explicitly.
27. Kalupahana (Causality, p. 116) and Mrs. Rhys
Davids in her Buddhist Psychology (London: G. Bell
and Sons Ltd., 1914) , p. 20, suggest that
'vi~n~naa.na' must here be construed in an eschato
logical sense.
28. For a helpful discussion of the part played by
vi~n~naa.na in the rebirth process, see Mrs. Rhys
Davids' Buddhist Psychology, pp. 22ff.
29. Regarding this, see Mrs. Rhys Davids' Buddhism,
p. 91.

p.423

30. D.I.156; S.II.29 and 214,III.57, IV.105; and
A.I.107 and 123, and II.6 (in addition to many other
passages) all speak of the connection between
arahantship and the destruction of the aasavas.
31. For example, Radhakrishnan, Indian Philosophy
(London: George Allen & Unwin, 1923),vol. I, p. 414
); Sir Charles Eliot, Hinduism and Buddhism (London :
Routledge & Kegan Paul Ltd., 1921), vol. I, p. 209;
A. B. Keith, Buddhist Philosophy (Oxford: The
Clarendon Press, 1923), p. 97; etc.
32. E.g., at Dhammapada 338 and elsewhere.
33. See Eliot, Hinduism and Buddhism, p. 208, and
Piyadassi Thera's Dependent Origination (Kandy:
Buddhist Publication Society, 1959), p. 27.
34. E. J. Thomas, The History of Buddhist Thought
(London: Routledge & Kegan Paul Ltd., 1933), p. 63.