中華佛學研究第4期 (p307-337)： (民國89年)，臺北：中華佛學研究所，http://www.chibs.edu.tw
Chung-Hwa Buddhist Studies, No. 04, (2000)
Taipei: The Chung-Hwa Institute of Buddhist Studies
ISSN: 1026-969X

The Nettippakaraṇa: Buddhist Hermeneutics?

“ Hāya ” and “ Naya ”

The Netti is organized around two principles: “ hāra ” and “ naya ” . Both offer a number of angles from which one can investigate the text for “ implications ” . For example, one of the hāras examines individual words

p. 313

as to their synonyms.　For a reader of the Buddhist scriptures it would be almost self-evident to take this hāra at face value as a technical rule.　In the introduction to his translation Ñāṇamoli stresses that the Netti is quite distinct from a commentary in that it does not intend to give rules that one can apply to difficult sections in the canon. Rather, it gives a “ number of contextuals departments within which the text can be considered for its particular implications ” .[7] On which ground does Ñāṇamoli distinguish the Netti from a simple manual of rules and procedures? Is the understanding that we gain through the application of a set of exegetical rules to a text really different from the same understanding obtained through considering the same text for its “ particular implications ” ?　Ñāṇamoli gives no reason why there should be a difference.　From the philosophical point of view Ñāṇamoli’s distinction is trivial if we leave it as it is.　Fortunately this need not be the case.　A detour via Schleiermacher show us why.　One of the important contributions of Schleiermacher to hermeneutics is that he was the first to discern the twofold a priori in the process of understanding.[8] Up to Schleiermacher the assumption was that understanding was an established fact and that the goal of the hermeneutical enterprise was to avoid “ misunderstanding ” .　Schleiermacher suggested that we may have reason to assume that “ misunderstanding ” is prior to “ understanding ” and that at all stages “ understanding ” must be actively sought.　The difference between the first and the latter approach is merely one of rigour.　We cannot prove that one of either positions is more adequate than the other.　The merit of Schleiermacher is that he has articulated a dilemma which any form of practising hermeneutics must acknowledge even if only by implication.　In this sense Ñāṇamoli’s distinction is hermeneutically relevant.　If we read the Netti as a set of

p. 314

rules which we have to apply, we presuppose that unhindered access to the actual meaning of the text is possible but that we have to clear the way ahead.　If, on the contrary, we read the Netti as Nanamoli wants us to, we reason the other way around.　We are sure that we never will be able to understand the text adequately but we do make an effort to do so.　The distinction between both options is fundamental but merely a matter of orientation.　The first perspective is negative; the latter is positive.　The first seeks to avoid misunderstanding; the second actively seeks to understand.　Ñāṇamoli follows Schleiermacher in that he would call the latter approach the more “ rigorous approach ” (die strengere Praxis in der [sc. Auslegungs] Kunst).

“ Hāra ” and “ naya ” are the two criteria to judge whether a “ sutta ” is “ yathā-suttam ” .　In the Netti “ sutta ” has a special technical meaning.　“ Sutta ” usually simply stands for the material text of the suttas, the sermons of the Buddha.　Sutta here connotes “ coherence ” .　A sutta of the Buddha is to be memorised because it represents the essence of his teachings.　It is valid in itself.　Another way of saying this is that it is coherent (as a “ thread ” ).　To make this clear Buddhism strictly distinguishes āgama or “ teaching ” of the Buddha from yukti (logic).　Teachings of the Buddha, laid down in the suttas, cannot be contradicted by any logical reasoning.　But the ideal of coherence does not only apply to the teachings of the Buddha as expressed in the words, the sentences and the texts they form.　Coherence is as much a matter of the reader.　Ideas are more likely to find a coherent expression in a certain text if we impose the criterium of coherence onto the text.　Coherence also relates to the way we present and re-present the teachings of the Buddha.　A discourse by the Buddha is more likely to last if we tie its ideas together into a coherent whole (a thread or sutta).[9] The Buddhist simile is that of flowers on an altar.

p. 315

Without being tied together they are soon blown away whereas tied together by a “ thread ” they remain longer.　Both aspects, the material text and the subjective act of understanding are equally important.　Understanding is never a unilaterally objective or subjective event.　A set of rules for the interpretation that would ideally cover all cases, does not exist.　The ambiguity of the term sutta/“ thread ” suggests that we cannot simply retrieve the meaning of a text “ as it was/is ” .　A text always has meaning also because it has meaning for us.　It seems that, for the Netti, the objective and the subjective pole of interpretation coincide in the act of the understanding.　The Netti expresses this insight by its use of the word “ sutta ” , a word that is half-metaforical, half-technical.　It is not until this century that this insight will receive systematic articulation in the science of hermeneutics.

The Netti does not define hāra and naya.　In its introductory verses it says that hāra relates to the wording of a text (byañjana), naya to its referent (attha).　This cannot be taken for a definition of hāra and naya for the Netti immediately adds that two of the three naya investigate the wording of the text instead of its referent.　The Netti clearly fails to circumscribe hāra and naya.　The body of the Netti will not help much to clarify the distinction.　Should we therefore leave it as it is?　Not necessarily!　The author of the Netti must have been aware that distinguishing between the words of a text and the many ideas corresponding to it is trivial.　We know that there are words that need to be interpreted and it is equally clear that their referents sometimes need clarification.　This formal distinction in itself does not suffice, however.　To the hermeneutician, who seeks to analyse the act of understanding, it is quite irrelevant to know that there are words and referents.　There can be words without referents (as in nonsensical speech) or referents without words.　So, too, must it be possible to develop rules for the interpretation of either of both separately.　Seeking to achieve a universal status for the

p. 316

act of understanding, the hermeneutican will try to surpass this level of formal distinctions.　Words and referents must be “ investigated ” (-vicaya), says the Netti.　What “ should be understood ” (viññeyyam) is not words and their referents, but “ teaching ” (desanā) and “ what is taught ” (desita).　“ Teaching ” slightly corresponds with wording and “ what is taught ” with what is referred to.　Both pairs of terms differ in that “ teaching ” and “ what is taught ” cannot be separated.　There is no teaching without something taught and vice versa.　We cannot reason the message away in favour of its contents and vice versa.　At the basis of every act of understanding for a Buddhist is this tension.　The distinction between words and referents is just a formalization of this tension.

What do these considerations of the hermeneutician contribute to the Buddhologist’s study of the Netti?　They may help him to see that the sixteen hāra and the five naya in the Netti are not two successive, isolated steps in interpreting Buddhist texts.

On the one hand, we cannot reduce “ teaching ” to “ what is taught ” by the Buddha.　The Buddha uses several techniques to adapt to his audience.　For example he recurs to several modes of discourse, to synonyms, etc.　Likewise, there must be a similar skill on the side of the reader of the sermons of the Buddha.　Examining the sermons of the Buddha is an ongoing, never-ending process that can be undertaken from different but complementary angles (the hāras).　Any of these many perspectives highlights another aspect of the teaching of the Buddha.　It would not be possible to examine the sermons in terms of hāras if it were possible to reduce the teachings of the Buddha to one single idea.　Not one aspect of the teaching of the Buddha is inferior or superior to another.　The very act of the Buddha’s teaching in different ways cannot be reasoned way in favour of what the Buddha “ really meant to say ” .

On the other hand, to fully understand the teaching we need to know “ what is taught ” .　Real understanding requires not only perspectivism. 　

p. 317

We would not even know what understanding means was there not “ something ” to understand.　But, for the reasons indicated, this “ something ” cannot be compressed into one or many discrete core-ideas.　How then does the Buddhist hermeneutician try to understand the sermons of the Buddha knowing that he cannot systematize the teachings of the Buddha in terms of scholastic rubrics?　If “ what is taught ” is no single “ something ” , or various of these, the only way out for the hermeneutician is to develop a number of guiding ideas which guide the work of the interpreter.　This is what happens in the second section of the Netti, on the five nayas.　Unlike in the first section on the hāras, we find not a single direct quotation from the sermons of the Buddha.　Instead, we find lists of types of mental “ corruption ” , of “ unprofit ” and “ profit ” , of “ persons ” , ……all of them organized in dyads, triads, and tetrads and divided under five nayas (called “ conversion of relishing ” , “ trefoil ” , “ play-of-lions ” , “ plotting-of-directions ” and “ hook ” ).　The nearest equivalent of the nayas in the Netti is not, as Bond suggests,[10] the system of nayas in Jainism.　Jainism has developed its naya-system as a corollary to its idea that reality is infinitely manifold.[11] Apart from its name there is nothing that fundamentally unites the Jain conception of naya and that of the Netti.　The proper context of the Netti is not a metaphysical presupposition about the nature of reality, as in Jainism, but the interpretation of the sermons of the Buddha in terms of the tension between “ teaching ” and “ what is taught ” , desanā and desita.　As I said, the naya-section does not illustrate the individual nayas by referring to specific portions from the scriptures where it sees its principles at work.　Yet each naya is composed of what

p. 318

undoubtedly are ideas culled from the sermons of the Buddha.　One way of dealing with the naya-section would be to deduce other ideas from it.　But this would do injustice to the Netti as a sample of Buddhist hermeneutics and as such it would be of no use to future generations of commentators.

Probably the Netti here applies a hermeneutical principle similar to the doctrine of the “ Loci communes ” .　This doctrine was introduced in 1521 by Melanchton through his book with the same title.　The “ Loci communes ” contains a summary of the main arguments of the Bible.　Its aim is not to enable commentators to “ draw the right conclusions ” when interpreting the Bible for this would be mere deduction.　The loci communes are a hermeneutical concept; not a tool.　The Bible explains itself (sui ipsius interpres), thus Luther’s maxim.　Loci communes are an attempt to show how it explains itself.[12] The reverence of Buddhism towards their canon of scriptures is not quite the sola scriptura principle operative in early modern hermeneutics (which was protestant).　However, both do share the idea of “ auto-interpretability ” .　In Buddhism it is the Dhamma which is its own interpreter (sui ipsius interpres).　The only criterion for interpreting the Dhamma is the Dhamma itself.　The Dhamma must be understood in terms of itself.[13] In light of a Dhamma that defies our human attempts to fully grasp it, the nayas can be nothing else than resources that help us to explain the “ Dhamma with Dhamma ” .

The Hermeneutical Nature of Each of the Hāras

Having considered the hermeneutical character of the Netti in general, I return to the two questions posed at the outset of this paper: “ Do interpretational theories or devices in Buddhism generate meaning where there is none? ” and “ Do they leave room for a ‘surplus of meaning’? ” I

p. 319

shall review the sixteen hāras of the Netti and try to assess to what extent they provide an answer to these questions.　Whether the Netti offers as Buddhist hermeneutics or not depends on this answer.

The first of the sixteen hāras is desanā-hāra (“ mode of conveying a teaching ” , in Ñāṇamoli’s translation).　The message of the Buddha cannot be understood unless we approach it as a teaching of some kind or another.　The being-taught of the message of the Buddha needs to be taken account in any attempt at explanation or understanding.　It is, so to say, one of the a priori’s of the act of understanding of the reader/interpreter.　Some texts should be approached in terms of “ escape ” , others as “ disappointment ” and still others in terms of “ gratification ” .　Some sections in the sermons of the Buddha may appeal to our desire for “ gratification ” and thus best be understood in this sense.　This is one of the aspects under which the material text of the Buddha’s sermons can be approached.　This perspective will be especially helpful for people of little intelligence.　People with a higher degree of intelligence may focus on the message of the Buddha in terms of “ disappointment ” .　For the more advanced reader it suffices to consider the message of the Buddha in terms of escape.　Such people need not be gratified by the text to find it “ meaningful ” as to their liberation from suffering.

One way of making sense of this first hāra would be to say that it merely stresses the exceptional skill-in-means of the Buddha.　This is unlikely, for another aspect under which one can consider the text is precisely “ skill-in-means ” (upaya).　Some sermons of the Buddha, or parts of it, may most fruifully be understood as a convenient fiction, suited and adapted to a specific audience.　According to the Netti the Buddha “ condenses ” in his sermons for people of high intelligence; he “ expands ” for people of lesser intelligence and for people of little or no intelligence he “ details ” .　In this latter case the Buddha tells stories (nidessa) or explains about the origin of certain words (nerutti).　If we take the Netti on its own

p. 320

not be a form of skilful means.　If skill-in-means really is a separate aspect to approach the message of the Buddha, there no longer can be sections in the text of the Buddha’s sermons that are “ merely ” skill-in-means.　Commentators have the tendency to know better what the Buddha really meant by saying or doing something than the Buddha himself.　What the Netti seems to be doing in this first hāra is taking our attention away from the Buddha.　All we possess to understand the Buddha’s message are a number of perspectives from which we can look at it.　Just as two looks at the same object from opposite angles do not contradict each other but instead help us to form a more complete idea of the object, so do the different perspectives in this first hāra.　The message of the Buddha is single.　For the Dhamma to be known to mankind, means to be taught.　Once it is taught it is apprehended by people in different modes.　Being one likewise means to be perspectivistic.　The best access to the teaching is in its being taught.　We have no direct access to it.　But that Dhamma is always “ taught Dhamma ” entails that we can understand it from many perspectives without one less adequately representing the Dhamma than the other.　If, for example, I do not right away understand how a particular sermon of the Buddha is conducive to my liberation (“ escape ” -hāra) I can have a second or a third look at it.　I can suspend my understanding in terms of “ escape ” and try to understand the text in terms of “ disappointement ” or see if I feel “ gratified ” through it.

The second hāra (vicaya-hāra or “ mode of conveying an investigation ” ) is an illustration of the idea that the Dhamma is sui ipsius interpres.　At this point the interpreter investigates “ terms, questions, answers and consecutivity (padam, pañham, vissajjanam, pubbāparam) ” .　Basically, the Netti does nothing more than to show how in a given section we need to analyse each of these four elements.　At first this seems trivial.　But questions presuppose more than we think.　By asking “ what is meant ” we can already subsume different question under one root-question.　In

p. 321

doing so the Netti makes explicit what is implicit.　It breaks though the literalness of the text to find meaning in the text.　In other words, the presupposition of the Netti is that meaning is achieved in spite of the literalness of the text.　That this is a specific option, and not the only viable one, is clear when we look at the Christian equivalent of the Netti, Augustine’s “ De Doctrina Christiana ” .　For Augustine obscuritas is not a negative idea.[14] In the case of trivial texts there is no problem: the question of understanding here does not pose itself.　In texts that contain obscure passages, however, we owe it to the literalness of the text that we can understand it.　We understand a text because it contains obscure passages.　Unambiguous passages throw light on the obscure passages.　But these first passages need to be understood as well.　Our understanding of these mainly springs from the tension they maintain with the obscure passages.　Obscure passages save our intelligence from “ boredom ” and keeps it active also when it has to understand clear passages.　For the Netti literalness is an obstruction that needs to be removed.　For Augustine it gives us access to the meaning of a text.　Both approaches are equally valid from a hermeneutical point of view.

The third hāra (yutti-hāra or “ mode of conveying a construing ” ) follows logically upon the preceding hāra (∫156).　I would suggest to change Ñāṇamoli’s translation “ construing ” (for yutti) into “ correctness ” or “ being in accordance with ” .　At this point the Netti contains an allusion to the “ Four Great Authorities ” (cattāro mahāpadese).　The Four Great Auhorities are four instances that the Buddha himself has appointed in the Mahaparinibbāna Sutta in case uncertainty would arise about certain important points of the Dhamma.[15]These four authorities are to of consulted in the proper order.　First, there is the authority of the Buddha

p. 322

himself.　What can be proven to be the word of the Buddha must be true.　In the absence of this authority the other authorities that are to be consulted are: “ a community of elders and distinguished teachers ” , if available; otherwise, a lose group of learned elders; and, finally, one such a learned elder.[16] What matters for the Netti is not so much each of these authorities independently.　Each of them has an authority that is guaranteed by the fact that it has been conferred upon them by the Buddha himself in the Mahāparinibbāna Sūtta─one of his major sermons.　What is important from the perspective of the interpreter is our appeal to each of these authorities.　An adequate understanding of the Dhamma does not depend on the level of “ learnedness ” of the “ elder ” consulted.　Reaching “ understanding ” of the Dhamma is entirely a matter of the individual interpreter.　His task consists in determining the “ correctness ” (yutti) of his appeal to the Four Great Authorities.　The Four Great Authorities are right (so the Buddha declared!) but we may be wrong in invoking their authority to support our interpretation of the text.　Before one of the Great Authorities can endorse our interpretation we have to make sure whether our interpretation is in accordance (yutta) with the (1) Suttas, the (2) Vinaya and the (3) “ Nature of Things ” (dhammatā).[17] A text consists of “ terms ” (pada) and “ expressions ” (byañjana).　Understanding these words is not merely a matter of saying what they refer to.　For example, the words and expressions in a sermon of the Buddha (sutta) must be “ descended into ” (otārayitabbāni).　When these terms and expressions occur in a text that has to do with Discipline (vinaya) one must be able to

p. 323

“ see ” them “ for oneself ” (sandassayitabbāni).　In a text that relates to the “ Nature of Things ” (3) (I suppose, a scholastic text) what we read must be “ be adaptable ” (sc. to the Nature of Things) (upanikkhipitabbāni).　But, when does all this happen?　In the case of a text that relates to the Nature of Things this is relatively easy to see.　We understand a scholastic text properly when we are able to relate it to the Nature of Things.　We ourselves understand the text but the text itself offers us the key: Texts on the Nature of Things are adaptable to the Nature of Things precisely because we approach them as texts on the Nature of Things.　A text of the Nature of Things is a text on the Nature of Things inasmuch as it “ is ” pratītyasamutpāda.　Here, as we encountered earlier in relation to the term “ sutta ” , the Netti uses Nature of Things/Dhammatā in a double sense.　It is both the material text as well as “ principle ” .　What does the interpreter do with the sermons of the Buddha (1)? He “ descends into them ” , says the Netti.　His interpretation of the suttas must be “ accordance with ” the principle of sutta-ness (or “ coherence ” ), just as our interpretation of scholastic texts (i.e. texts on the Nature of Things) had to be “ in accordance with ” the principle of the Nature-of-Things.　Texts that guide our conduct (2), finally, are well understood when they are related to our own situation, in other words to the principle that we are to be “ guided out ” (vi-naya): “ In which Vinaya are the terms and expressions ‘seen for oneself’? [answer of the Netti] in those Vinaya-texts that contain reference to raga, dvesa and moha. ”

In the Netti, “ sutta ” , “ vinaya ” and “ dhammatā ” (or abhidharma) are auto-interpretative.　They set themselves their own principles for interpretation.　Only when the interpreter acknowledges this can he invoke one of the four Great Authorities to endorse his interpretation of the text.[18]

p. 324

The fourth hāra analyses “ footings ” (padaṭṭhāna) or, paraphrasing, it searches for “ that which makes something possible ” .　The Netti explains itself by analysing the twelve-membered chain of dependent origination: member one of the chain is the “ footing ” of member two as two is to three, and so on (∫164-165).　As such this fourth hāra is utterly useless.　The modern hermeneutician does not need the Netti to know that ideas follow from one another.　Though the “ footings ” hāra is of little or no direct exegetical use to today’s reader, it has hermeneutical value.　The fourth hāra touches upon the act of understanding itself without itself explaining anything.　This “ understanding ” is, of course, in the first place the understanding of the modern (and not so modern) Buddhologist.　What does it mean to say that a Buddhologist understands a certain term in the texts? For example, what does paññā mean?　A common procedure adopted by Buddhist scholars is to point to a referent that in fact refers back to paññā.　Trying to understand paññā in Buddhology often is no more than recording a list of synonyms that explain each other mutually.　We need the synonym to explain paññā but at the same time need paññā to explain the synonyms.　The “ footings ” hāra proves its value here as warning against such a form of Buddhology.　“ Footings ” /padatthana does not mean anything in itself.　It always implies lakkhana.　Both relate to each other as “ forma ” and “ materia ” .　paññā is no longer an “ empty ” idea (as it is in much Buddhist scholarship) if we can show where it derives it “ materia ” or “ substance ” from, i.e. from the Four Noble Truths (saccani).　But this alone does not yet prevent paññā becoming something like one of the members of the twelve-membered chain when it is wrongly understood.　The materia-forma scheme belongs to the structure of the act of understanding itself.　paññā too has “ forma ” : pajāna.

The fifth hāra (lakkhana-hāra) takes up the “ lakkhana-mäβigkeit ” of the act of understanding.　Briefly, it says that “ when one idea is mentioned, by way of implication all other ideas of ‘like characteristic’ (eka-lakkhanā)

p. 325

too are intended ” .　One, again, might use this as an exegetical device, for example to reduce discrete terms to one basic idea (as in the case of the twelve-membered chain of dependent origination).　However, if the Netti is really a sample of Buddhist hermeneutics, this is insufficient.　Probably, lakkhana or characteristic has not so much to do with “ what is understood ” as with our “ understanding ” itself.　No understanding, and a fortiori no Buddhist form of understanding, is possible without some structure of “ lakkhana-mäβigkeit ” in it.　With an example of the Netti: If the Buddha says that the eye is “ impermanent ” , this same statement applies also to the ear, the nose and to all other ideas of like “ characteristic ” .　This may be so because the Buddha says so.　However, it is true in the first place because we make the association.　A seventh sense-faculty, discovered after the Buddha, would force us to subsume it under the group of six.　It is because we can make such associations according to “ likeness of characteristic ” that understanding the Buddha remains possible though never complete.

In the sixth hāra (catubyuha-hāra or “ Mode conveying a Fourfold Array ” ) we find that the interpreter should analyse the “ etymology ” (nerutti), the “ intention ” (abhippāya), the “ circumstances ” (nidāna) and the “ sequence ” of the words and phrases in the text.　To discuss this hāra we quote two sections from the Sutta Nipata, a statement by the cattle-owner Dhaniya, followed by the answer of the Buddha.　Together they are the first of a whole series of quotations under the heading “ circumstances ” :

A man with children finds relish through his children;

And a cattle-owner likewise through his cattle.

These essentials of existence are a man’s relish;

Who has them not will never relish find. (Sn. 33)

A man with children finds sorrow through his children;

And a cattle-owner likewise through his cattle.

p. 326

These essentials of existence are a man’s sorrow;

Who has them not will never sorrow find. (Sn. 34)[19]

Does it add anything significant to our understanding of the text to know under which circumstances and with what intention in mind the Buddha uttered his answer to Dhāniya? Does our understanding of the text depend upon on this knowledge or is just “ footnote-knowledge ” ? I believe that we do not understand this passage unless we ask for its “ intention ” and “ circumstances ” .　Our understanding of this text does not, so to say, accompany the text as we have it in front of us.　Meaning is something we have to recover from the text.　The second and third member of this hāra helps us in this respect.

Let us analyse the answer of the Buddha!　The Buddha’s statement that man finds sorrow through his children is not a statement that is valid by virtue of itself.　It would be un-Buddhist to believe that things exist which cause suffering in absolute way.　The Buddha says that children are a source of sorrow but he does not “ intend ” to say that one should consider them as the cause of sorrow.　What he “ intends ” to say is something different, namely that one has to abandon the idea that through and in children relish can be found.　Nonetheless, if we leave it to the inspiration of the reader to understand the meaning of the text from the Suttanipata the result will be something meaningless.　If we leave the text from the Suttanipata as it is, hoping that it will spontaneously release its meaning, the only thing that can happen is that we understand it wrongly.　We should address the text through such distinctions as the one between

p. 327

“ intention ” and “ circumstances ” .　Only then can we find meaning where there is none or where it threatens to become nonsensical.　Actually, in the dialogue between Dhāniya and the Buddha it is irrelevant to know that Dhaniya’s question was the “ circumstance ” that led the Buddha to his statement.　But the interpretational device of “ circumstance ” becomes important.　The dialogue has the right meaning because there is a difference between “ intention ” and “ circumstances ” .　

Something of the sophisticatedness of this reconstruction of the logic behind the Netti is explained by the fact that Buddhist texts do not necessarily subdivide themselves in sections relating to the “ intention ” of the Buddha and sections relating to the “ circumstances ” .　As I have tried to demonstrate with the Netti’s quotation from the Suttanipata the distinction between “ intention ” and “ circumstances ” helps us to put the right meaning into the text.　But this distinction, if we follow the Netti, is not something that is empirically observable.　We cannot list “ intentions ” and “ circumstances ” in a glossary appended to the sermons of the Buddha (as is possible for images and similes).[20]

The tenth hāra is simply called “ synonyms ” (vevacana).　Is there any point in stressing the fact that the interpreter should search for synonyms for the words he does not understand? Would not every reader do this spontaneously without the need of a rule which prescribes him to do so? As a rule of interpretation this hāra is trivial.　From the point of view of the hermeneutician it is not.

What does “ synonym ” mean? Two definitions of “ synonym ” are possible.　Either, we may say that a synonym is the equivalent of another expression that refers to the same idea: “ (...) The Blessed One demonstrates

p. 328

a single idea by means of many synonyms ” .[21] 　Or, in a somewhat weaker version, we may stress the mere interchangeability of these expressions.　This distinction is important for it entails two different conceptions of the idea to which the synonyms refer and, therefore, two different attitudes towards the Dhamma preached by the Buddha.　In each case, the single idea (dhamma/Dhamma) referred to has another meaning.　In the first definition, it is something abstract-indeterminate.　If, as happens in the second definition, we simply confine ourselves to the observation that synonyms are merely interchangeable expressions, we do not preclude the possibility that somewhere there might exist a vevacana which accurately renders the “ single idea ” , that we once even might find it and thus make the preaching of the Buddha redundant.

What is the relevance of this distinction for the Buddhist who is confronted with a text on which he wants to comment?　One major characteristic of many Buddhist texts is that they want to say a lot about things that actually can or may never be adequately described or qualified.　A good example is the idea of “ insight ” (paññā).　Buddhist commentary or exegesis becomes genuine hermeneutics when it tries to do full justice to the Middle Way in the way it comments upon texts.　Ideally, a Buddhist commentary ought to avoid to be both exclusively affirmative as well as exclusively negative.　A good Buddhist commentary does not postulate its definition of “ insight ” .　But neither does it end up being merely apophaticism.　This is a very delicate balance.　What, to give another example, will the commentator do with the eulogy of nirvāṇa found in the Netti?[22]　Let us assume that the Buddhist commentator has nothing but the bare text in front of him and that in some way or another he needs to make

p. 329

sense of it.　The question that the Buddhist commentator, and any commentator for that matter, ought to ask himself is as follows:

Do I not understand the text wrongly by reading it, thus reading it, saying either too much or too little on nirvāṇa?

This is the ambiguous situation in which the commentator finds himself.　How to find a way out of it?　One way would be to adopt the vevacana/ “ synonyms ” hāra.　This hāra does not help him to explain the text; a good dictionary would help him more here.　Rather, the vevacana/ “ synonyms ” hāra is a scheme which the commentator integrates in his act of understanding the text.　Understood in the first of two definitions we gave to “ synonym ” , it structures his understanding of the text.　It is only after this first step that the commentator may proceed to the actual explanation of the text.

From the level of “ words ” in the tenth hāra the Netti proceeds to the level of “ descriptions ” in the eleventh hāra.　Just as one single idea has several synonyms, it also may be expressed in several equivalent “ descriptions ” or paññatti.　To illustrate the hermeneutical value of this hāra let us take the following section:

How could a man to sensual desires stoop

Who pain has seen and that wherefrom it sources?

Who knows they make for clinging in the world

Should mindful train in guiding them away.[23]

We can easily recognize the four Noble Truths in this text.　Each of

p. 330

these four Noble Truths represents one “ single idea ” (eka dhamma).　As such an idea is not something we can have knowledge of.　To know what an abstract idea, such as “ suffering ” , means we need descriptions.　 “ Descriptions ” or paññattis mediate; literally, they “ make known ” (from paññāpeti).　 “ Who pain has seen ” (in the second line) is an attempt to describe the idea of suffering.　According to the Netti, the phrase “ who pain has seen ” does two things: it describes the abstract idea of “ suffering ” by being a “ synonym ” for it and by being an “ adequate analysis or diagnosis ” of it.[24] 　The fact that “ someone has seen pain ” is a good illustration of the abstract idea of suffering (= “ synonyms ” ).　And, if we want to have a good analysis of our existence in terms of the abstract idea of suffering, we will find it in the same phrase “ (...) who pain has seen ” (= “ adequate analysis or diagnosis ” ).　This phrase, however, is not the only possible one; we can imagine many other words, phrases and sentences in the Buddhist canon that all describe the abstract idea of suffering in their own way.　For each of these one or more different types of “ descriptions ” can be developed and each given a technical name.　It is clear why this is hermeneutics: we can never fully exhaust the meaning of the idea of suffering; speaking of suffering and other key ideas of Buddhism through the mediation of “ descriptions ” acknowledges and at the same time preserves this surplus of meaning.

The thirteenth hāra is, again, of a striking trivialness.　Ñaṇamoli (p. lxxii) explains the hāra called sodhana or “ clearing up ” as the hāra which shows “ how the subject-matter of a question must be covered by the answer ” .　Bond’s explanation is similar to Ñaṇamoli’s: The commentator understands the answer to a question correctly only if he can point to the “ spirit which motivates the question ” .[25]

p. 331

If we maintain that the Netti is about hermeneutics, questions of correct and incorrect interpretation are irrelevant.　A hermeneutician will not look for rules to guide his interpretation in the right direction.　Hermeneutics has no rules; it has only principles.　It surpasses the level of interpretation.　Hermeneutics relates to the act of understanding itself and tries to discover the principles according to which it operates.　This particular hāra, the “ clearing up ” , is hermeneutical for another reason─for its preserving a “ surplus of meaning ” .

What is the “ question ” meant by this hāra?　Within every question, we can distinguish between the padam or “ term ” , the material aspect of the question, and the “ instigation ” (ārambha) which causes someone to ask this question.　The implication of this is that answers are never related to a question as such.　There is no one-to-one correspondence between an answer and a question asked.　It is true that it helps to know what exactly a given portion in a Buddhist text is an answer to.　Everything the Buddha expounds in the sutras is relevant.　When a particular utterance of the Buddha gives the opposite impression, it will be the interpreter’s task to clarify why the Buddha uttered it.　One way of doing this, is to find some question to which the Buddha’s utterance could have been or could be an answer.　This would be “ explanation ” ─not yet hermeneutics─and it would not be what the “ clearing up ” hāra expects us to do.　We should not try to establish a causal link between the Buddha’s answer and some question.　The Buddha’s utterances have universal value.　They are true for a Buddhist even if the Buddhist is unable to identify one particular question to which the Buddha’s utterance might be an answer.　The Buddha’s teaching is universal and hence “ inexhaustible ” : It is valid even when we cannot immediately point to one specific question to which it is an answer.

What makes a portion in the scriptures understood by the commentator is not solely the fact that he can identify some question at the basis of it.　

p. 332

Surely, this is important but it is not enough.　For example, if we read the Buddha’s words “ By ignorance is the world shut in, ” we may simply refer to “ What is the world shut in by? ” [26]　If we read an utterance of the Buddha like “ suffering is the world’s greatest fear ” , we may likewise assume that someone asked a question “ what is the world’s greatest fear? ” and that the Buddha’s answer is an answer to this question.　But we have to realise that answers may answer more than the material contents of a question (its padam).　In this particular case the Buddha’s answer is also an answer to the situation from within which the question “ what is the world’s greatest fear? ” was asked.　The interlocutor asks the Buddha a question within a context of suffering, of which he seems to be unaware, and he receives an answer from the Buddha which makes him right away aware of the context from within which he asks his question.　In other words, a question is never a monadic entity.　In answering a question more is involved than exhausting the contents of the question.　For example, the answer “ My name is Philip ” is fully understood when we know that someone asked the question “ What is your name? ” .　Here the answer exhausts the question.　But the answers of the Buddha are manifold.　It is good for the commentator to realize that if the Buddha says that “ suffering is the greatest

p. 333

fear of the world, ” that he does more than answer a question posed by someone.　A person may have many conscious reasons and/or unconscious causes to ask a question which remain unexpressed in the words (padam) of the question.　A question means more than it says.　A commentator is a hermeneutician from the moment he acknowledges this surplus of meaning.

The fourteenth hāra introduces two important technical terms: “ unity ” (ekattatā) and “ diversity ” (vemattatā).　The Netti illustrates this hāra with a number of examples.　Each of these begins with the question “ What is X? ” (katamam X?).　If we, for example, ask what “ suffering ” means we expect to receive an answer which would allow us to gain a clair and distinct idea of suffering.　We want to know “ what ” suffering is.　But is it right to expect such an answer? The Netti answers the question “ what is suffering? ” with a whole list of things that are suffering: birth, ageing, sickness, the five khandhas, etc.─and adds that this is “ diversity ” .　If the commentator expects the Netti to be a help in his interpretation of the scriptures, it turns out to be doing exactly the opposite.

What then does the Netti want to make clear with this hāra?　It says that ideas can be explained as a “ unity ” or as a “ diversity ” without their meaning being altered (thus Bond, 1982, 92).　But is this all there is to be said about this hāra? It is true that “ the essential unity of the Dhamma can be expressed in diverse ways ” (ibid.) but do we need the Netti to convince us of this? Perhaps Dhammapala’s commentary to the Netti is more elucidating.　The root-verse belonging to this hāra reads as follows:

ekattatāya dhammā // ye pi ca vemattatāya niddiṭṭhā

te na vikappayitabbā //eso hāro adhiṭṭhāno (Hardy, p. 4)

Banamoli translates this as:

Ideas when demonstrated by // [both] unity and diversity　

p. 334

Need thereby suffer no disjunction:// This mode conveys Expression’s Terms (p.10, ∫18)

There is no reason to maintain “ expression’s terms ” as a translation for adhiṭṭhāno.　Apart from this, there is the hybrid expression “ suffer disjunction ” .　It is not at all clear what this could mean.　I would suggest the following paraphrase:

Ideas are demonstrated by unity or diversity

but may not be imposed upon the text, hence this hāra, called “ standing beyond ” .

In his commentary to this section Dhammapāla explains “ unity ” as “ sameness ” (sāmañña) and “ diversity ” as “ differentiation ” (visesa).　For Dhammapāla unity and diversity are concepts which we impose upon (vikappeti) the text and he rightly points out that the fourteenth hāra precisely warns us not to do so:

In the sermons of the Buddha ideas are taught in terms of “ sameness ” , for example when the Buddha speaks about the “ origin of suffering ” , or they are taught in terms of differentiation, for example in terms of “ birth, ageing, thirst consisting in desire, thirst consisting in becoming, etc. ” .　We are inclined to superimpose categories of “ sameness ” and “ differentiation ” by asking questions as “ What is sameness in this text? And what is differentiation?　 ” We must not do so.[27]

p. 335

“ Sameness ” and “ differentiation ” are fictions, literally: “ they have no position ” (anavaṭṭhāna). They do not correspond to something we could isolate from the text.　We cannot make one list of terms in the teachings of the Buddha that are expressions in terms of “ sameness ” and another, parallel list of terms which he used to express “ differentiation ” .　Dhammapāla compares the problem of “ sameness ” and “ differentiation ” in interpreting the sermons of the Buddha with our use of the terms “ today ” and “ tomorrow ” or “ east ” and “ west ” .　These are “ differentiations ” of the terms “ time ” and “ direction ” .　The relationship between “ sameness ” and “ differentiation ” in the Buddha’s sermons is similar.　The following example from the Netti might suggest that there are terms in the sermons of the Buddha that are by definition units ( “ sameness ” ) and others that exist only in order to differentiate:

Suffering is a unity.　Herein, what is suffering?─Birth is suffering, ageing is suffering, sickness is suffering, death is suffering, association with the loathed is suffering, dissociation from the loved is suffering, not to get one’s wish is suffering, in brief the five categories of assumption are suffering: form is suffering, feeling is suffering, perception is suffering, determinations is suffering, consciousness is suffering.　This is a diversity.　(transl. Ñāṇamoli, p.103, ∫424)

Is this a pattern of interpretation set out by the Netti which we can apply in other instances? Not if we follow Dhammapāla:

Although the expression “ This is suffering ” is “ sameness ” if seen from the point of view of the series beginning with birth; speaking from the point of view of the [four Noble] Truths, it is itself “ differentiation ” .[28]

p. 336

Dhammapāla seems to suggest that it is not enough to simply know which sections in the sermons of the Buddha are phrased in terms of “ sameness ” and which ones in “ differentiation ” , and then explain each of them accordingly.　 “ Unity ” (ekatattā) and “ diversity ” (vemattatā), “ sameness ” (samañña) and “ differentiation ” (visesa) are categories that belong to the text itself and that constitute its richness.　They are like a quick succession of alternating perspectives which never fully capture the over-all view of the object.　It is this richness which the interpreter must respect when he comments upon the sermons of the Buddha.

[1] Donald S. Lopez, Buddhist Hermeneutics, Honolulu, University of Hawaii Press/Kuroda Institute, 1988. Other relevant contributions include: John Maraldo, “ Hermeneutics and historicity in the study of Buddhism ” , in: The Eastern Buddhist, 19 (1986), 1, 17-43. José Ignacio Cabezón, “ Vasubandhu's Vyākhyāyukti on the authenticity of the Mahāyāna Sūtras ” , in: Jeffrey R. Timm (ed.), Texts in context: traditional hermeneutics in South Asia, New York, State University of New York Press, 1992, pp. 221-243.

[2] For an excellent survey of this quest for universalism in hermeneutics from the Alexandrine school onwards: Jean Grondin, L’Universalitué de l’herméneutique, Paris, Presses Universitaires de France, 1993.

[3] Friedrich Schleiermacher (ed. Heinz Kimmerle), Hermeneutik, Heidelberg 1974, pp.75: “ Die Hermeneutik als Kunst des Verstehens existirt noch nicht allgemein sondern nur mehrere specielle Hermeneutiken. ”

[4] This question is Ricoeurian par excellence. See Paul Ricoeur, Interpretation theory: discourse and the surplus of meaning, Fort Worth (Texas), The Texas Christian University Press, 1976.

[5] See introduction to the Pali Text Society edition: (Pali) E. Hardy, Oxford 1995 and (English) Ñānamoli, “ The Guide ” , Oxford 1977. Ñānamoli has nothing to say about the exegetical value of the Netti. See also: George D. Bond, “ The Nature and Meaning of the Netti-Pakaraṇa ” , in A.K. Narain (ed.), Studies in Pali and Buddhism (Delhi, B.R. Publications, 1979), pp. 29-39; George D. Bond, “ The Netti-Pakaraṇa: A Theravada Method of Interpretation ” , in S. Balasooriya (ed.), Buddhist Studies in Honour of Walpola Rahula, London, Gordon Fraser, 1980, pp.16-28. For the connection with Buddhaghosa's commentaries: George D. Bond The Word of the Buddha: the Tipitaka and its interpreters in Theravada Buddhism, Colombo, Gunasena, 1982. Id., “ The gradual path as a hermeneutical approach to the Dhamma ” , in: Lopez, pp. 29-45. In preparation: Marian Caudron, Templates for Wisdom: A Summary of the Method and Purpose of the Netti-Pakaraṇa (Ph.D. dept. of Religion, Northwestern University).

[6] Compare this with Geiger who concluded from the title of the Nettippakaraṇa (and that of the Peṭakopadesa) that its contents serves as “ an introduction to Buddhism ” (in: Pāli literature and Language, Delhi, Oriental Books Repr. Corp., 1968 (2nd ed.), p. 26.

[11] For a lucid exposition of the nayas and their classification in Jainism, see: Y.J. Padmarajiah, A Comparative Study of the Jaina Theories of Reality and Knowledge, Bombay, Jain Sahitya Vikas Mandal, 1963, pp. 303-330.

[12] Martin Leiner, “ Die Anfäge der Protestantischen Hermeneutik bei Philipp Melanchthon ” , in: Zeitschrift für Theologie und Kirche 94 (1997) 477.

[16] Ibidem: “ 1° sammukhā Bhagavatā sutaṃ / Bhagavato vacanam, 2° saṃgho satthero sapāmokkho, 3° sambahulā therā bhikkhū bahussuttā āgatāgamā dhamma-dharā vinaya-dharā mātikā-dharā, 4° eko thero bhikkhu ” (as preceding).

[17] Nature of Things/Dhammatā is absent from the Pali version of the Mahāparibbāna Sutta.　It does occur in the Chinese and Sanskrit version (as well as in the Netti), see: Έtienne Lamotte, “ La Critique d'Authenticité dans le Bouddhisme ” , in: Kern Institute (ed.), India Antique (Leyden: Brill, 1947), pp.213-222, esp. 221.

[18] For another attempt at interpretation of the four Great Authorities, see: Michael Pye, “ Comparative Hermeneutics in Religion ” , in Id. & Robert Morgan, The Cardinal meaning.　Essays in Hermeneutics: Buddhism and Christianity, The Hague/Paris, Mouton, 1973, p. 41.

[19] (Hardy, p. 34) “ Yatha so Dhāniyo gopālako Bhagvantaṃ āha: nandati puttehi puttimā, gopiko gohi that’ eva nandati, upadhī hi narassa nandanā, na hi so nandati yo nirūpadhī ti.　Bhagavā āha: socati puttehi puttim, gopiko gohi that’eva socati, upadhī hi narassa socanā, na hi socati yo nirūpadhī ti. ”

[20] as Ñanamoli has done in his translation of the Majjhima Nikāya (The Middle Length Discourses of the Buddha) (Boston: Wisdom Publications, 1995), pp.1407-1409.

[21] (Hardy p. 53) “ Ekam Bhagavā dhammaṃ aññamaññehi vevacanehi niddisati. ”

[22] (Hardy, p. 55) “ Asaṃkhatam, anantam, anāsavañ ca, saccañ ca, pāram, nipuṇaṃ, sududdasaṃ, ... ” (more than fifty epiteths are given).

[23] (Hardy, p.61) “ Yo dukkham adakkhi yato nidānaṃ//kāmesu so jantu kathaṃ nameyya//kāmā hi loke saṃgo ti ñatva//tesaj satimā vinayāya sikkhe ti. ”

[24] (Hardy, p. 61) “ Yo dukkhan ti vevacanapaññatti ca dukkhassa pariññapaññatti ca. ”

[26] The whole section from the Netti reads as follows (quoting from Sutta Nipāta, ∫1032-1033, P.T.S. ed. pp. 197-198):

　　(1) ti pañhe (1') ti Bhagavā padam sodheti no ca ārambhaṃ [and so on for (2)/(2') and (3)/(3')]

　　(4) ti pañhe (4') ti Bhagavā padam sodheti, suddho ārambho (Hardy, p. 70-71)

[27] (Nettipakaraṇa Aṭṭhakathā, Chaṭṭha Saṅgāyana CD-ROM, Myanmar page 28): Ye dhammā “ dukkhaṃ samudayo ” ti ādinā sāmaññena, “ jāti jarā kāmataṇhā bhavatanhā ” ti ādinā visesena ca sutte desitā, te “ kimettha sāmaññam, ko vā viseso ” ti evaṃ sāmaññavisesavikappanavasena na vikappayitabbā. ”

[28] (Ibidem): “ Idaṃ dukkhan ” ti vuccamānaṃ jāti ādi apekkhāya sāmaññampi samānam saccāpekkhāya viseso hoti. ”

The Nettippakaraṇa: Buddhist Hermeneutics?

Summary

“ Hāya ” and “ Naya ”

The Hermeneutical Nature of Each of the Hāras

《引導論》：佛教詮釋學？

提要