駒澤大学佛教学部論集 14 019松本 史朗「Lankavatara on itaretarasunyata」

pmes

k

_{eqpmetgzzthfiee14utngM581FIOfi}

LafakavataraonitaretaraStinyata

ShiroMatsumoto

There

are

two

types

of

Sitnyata

in

MahayEna

Buddhism.

One

is

lalesa4a-Sanyata,

the

other

is

itaretara-Sanyata(t).

These

two

types

of

jtinyatd

are explained

in

contrast

by

KamalaSila

in

his

Madhyamakalolea

(MA,

D.

ed.

No.

3887)

as

follows;(2)

[1]

gafi

yafi

don

gshan

gyis

dben

pahi

phyogs

la

sogs

_pa

bgah

shig mfion sum

gyis

bzufi

nas

de

las

_gshan

_pa

stofi

_pa

mtion sum

_gyis

rtogs

_par

bya

bar

hgyur

shes

bya

la

sogs

_pa

smras

_pa

de

_yah

rigs

_pa

ma

_yin

tel

tshul

bdi

ni

gcig

gis

gcig

stoh

_pa

fiid

_yin

_gyil

mtshan

fiid

stoh

_pa

fiid

ni ma

_yin

no!1

mtshan

fiid

stofi

_pa

fiid

kyis

ni

don

darn

_par

cho')'

thams

cad

ho

bo

fiid

med

par

bdod

de/

don

dam

par

de

dag

gi

rah

daft

spyibi mtshan

fiid

ji

ltar

rnam

par

gshag

pa

stofi

pahi

phyir

ro/1

dfios

po

rnams

kyi

gclg

gis

gcig

stofi

pa

fiid

gaft

yin

pa

de

ni

don

dam

_pa

ma

yin

gyi/

bon

kyah

tha

sfiad

bden

pa

la

brten

pa

yin

nol!

(sa

169b6--170al)

(tr.)

The

statement(3)

[of

the

]Vbga-ca'ras](4)

is

not true that, after

having

_percieved

by

perception

some place etc.

devoid

of other objects, one will unclerstand

by

perception

that

the

place

etc. are void _{of other objects.}

This

opinion advocates

itaretara-Sanyatd,

and not

lafesapta-Stinyata-.

According

_to

lalesapa-Stinyata",

all

perties

are considered

to

have

no essence

from

the

viewpoint of

paramdrtha,

because

they

are

devoid

of

the

special and

the

universal characteristics, which

are established,

from

the

viewpoint of

Paramartha.

The

itaretara-Stinyata

of

objects

is

founded

not on

Paramartha-satya

but

on vyavahiira-satya.

Itaretara-Stznyata

is

the

term

given

in

the

Lahleavatdrasditra

(LAS,

Nanjio

ed.,

75,

10,

17,

19)

to

a

kind

of

interpretation

of

Stinyata,

which

can

be

formulated

by

the

sentence "vad

yatra

ndsti

tad

tena

Stznyam"

(when

A

does

not exist

jn

B,

B

is

void of

A)(5).

The

characteristic

feature

of

this

interpretation

i$

that

the

locus

of

Stinyatd

(B,

stofi

_gshi)

-(2)

Lahkavatfira

on

intaretaragUnyata

(Matsumoto)

is

always regarded as real

(sadbhdra)(6)

in

contrast

to

the

unreality

(asadbhdva)C6)

of

the

super-locus of

Stinyata

(stofi

rgyu).

On

the

contrary,

the

rnain

_point

of

the

laksa4a-Stinyata

interpretation,

which

is

asserted

by

Mddhyamiha

_{philosophers,}

i$

that

the

Iocus

as well as

the

super-locus

of

Sitnyatd

is

unreal(7).

Therefore,

these

two

kinds

of

interpretation

concerning

Stinyatd

can

be

distinguished

from

one another according

to

their

opinions concerning

the

reality

of

the

locus

of

Stinyata.

In

my

previous

paper,

I

postulated

a

hypothesis

for

the

sake of understanding

more simply

the

historical

development

of

Mahayana

Buddhisrn

in

India(8).

According

to

my opinion,

in

Mahayana,

there

are

two

systems

of

_philosophy,

which are radically opposed

to

and must not

be

identified

with each other.

One

is

Dhatuvada

(the

theory

of

locus)

which asserts

that

the

locus

is

real, and

the

other

is

Sttnyatavdda

(the

theory

of

void-ness), according

to

whose

theory

all

is

unreal.

Therefore,

itaretara-Stinyatd

is

the

Stinyata

of

Dhatuvdda,

while

lalesa4a-Stinyatd

is

that

of

'Stiiayatdvada.

From

my viewpoint,

the

so-called

tathagatagarbha

theory

and

the

Yogdcdra

_philosophy

are

both

the

typical

forms

of

Dhatuvada,

although

the

tathagatagarbha

theroy

is

the

more

_genuine

one.

Then,

'

is

the

Lahkavatam

a work of

Dhatuvacla

or of

Stinptatjvada?

This

is

a

_question

somewhat

difficult

to

answer,

because,

although

the

Lahledvatara

is

famous

for

its

tathagatagarbha

theory

and

for

its

identification

of

alayavi]-fiana

with

tathdgarbha,

there

are many

_passages

in

the

Lafikdvatara

which explain

the

Mddhyamika-like

Sitayata

theory

and which are

_quoted

by

many

Mad]tyamika

_{philosophers.}

The

most

important

fact

which seems

to

contradict

the

contention

that

the

Laitleavatara

is

the

work of

Dhatuvada

is

that

itaretara-Stinyata

is

criticized

in

the

sutra as

follows:

[2]

esa ca rnahamate

itaretaragtinyata

sarvajaghanya sa ca

tvaya

_{parivarjayitavya}

1/(LAS,

75,

18-19)

(tr,)

Oh

Mahfimati,

this

itaretara-stiayatd

is

the

worst of all

[the

seven

Stitryatas].

Lafik"avatara

on

intaretaraSitnyat-a

(Matsumoto)

(3)

If

this

criticism of

itaretara-jtinyata

agrees with

the

main

therne

of

the

Lahleavatdra,

this

sutra should

be

regarded not as

the

work of

Dhdtuvdida

but

as

that

of

Stinyatdvado.

Then,

is

itaretara-Stiayata

really

criticized

in

the

Lahkavatara?

I

do

not

think

so.

Firstly,

the

commen-tators

either

_give

no comment on

the

passage

quoted

above or,

if

anything, vague comment.

Jfianasribhadra,

in

his

commentary

on

the

Lafika"uatdra

_(P.

ed.

No.

5519),

_gives

no explanation on

the

_passage,

although

he

expounds

the

idea

_of

itaretara-Stiayata

itself

(ni

124b2-7).

Jfianavajra's

comment

in

his

Tathagatagarbhala72tkara(9)

(TGA,

P.

ed.

No.5520)

on

the

_passage

is

as

follows:

[3]

de

_bdra

bahi

stofi

pa

fiid

ni

byis

_pabi

skye

bo

thams

cad

kyis

(orig.

kyi)

rtogs

gifi

_goms

kyah

kun

_nas

fion

mofis

_pabi

_phyogs

spofi shinlskyes

bu

la

iie

bar

mkho

bahi

don

thob

_pa

la

rni

phan

pas

de

bdra

ba

gtan

la

phab

(orig･

phan)

pa

ni

dgos

pa

med ces

pabo!!(pi

149a4-5)

(tr.)

Although

this

form

of

Su'nyatd

[i.e.

itaretaraSu-nyatd]

is

understood and

ticed

by

all

the

foolish

_people,

it

is

useless

to

ascertain

the

form

of

Sitnyafa

because

it

is

not useful

for

rejecting

the

defiled

_portion

and

for

obtaining

the

objects

desirable

to

[good]

persons.

This

comment can

hardly

be

regarded as

the

_{philosophical}

one,

because

no

theoretical

defect

of

itaretara-Stinyata

is

_pointed

out

here.

Therefore,

we can conclude

that

the

explanation of

the

theoretical

difference

between

the

two

types

of

Sanyata

mentioned above

is

not

_given

by

the

commentators on

the

Lanleavatdra.

The

more

important

fact

is

that

there

is

in

the

sutra a

_passage

which seems

to

support

itaretara-Stiayata.

This

passage

reads:

[4]

yad

apy. uktarp maya niratmanab sarvadharma

iti

tasyapy

artharp

yarp

mahamate, niratmabhavo _{rnahamate nair-atmyam,} svatmana sarvadharma

vidyante _na

_paratmana

_{go'gvavat!tadyathti}

rnahamate na

_gobhavo

'gvtitmako,

na _caSvabhavo

_gavatmako,

_{saCiO) san} nasan, na ca

tau

svalak$a4ato na,

vidyete eva

tau

svalaksarpatalj, evam eva mahamate sarvadharrnfi na ca

svalak$a4ena na sarpvidyante, vidyanta eva

(LAS,

188,

7-14)

(tr.)

Oh

Mahamati,

the

meaning of my _statement

_that

_all

_properties

have

no self

should

be

also

[correctly]

understood.

Having

no _self

(naira-tmya)

is

_the

state

-(4)

Lafikavat5ra

on

intaretaragtinyata

(Matsurnoto)

of

having

no essence

(atman).

All

_properties

exist with

_their

own essences,

and without

the

essences of other _objects, as a cow and a

horse.

Namely,

the

being

(bha-va)

of a _cow

has

no essence of a

horse,

and

the

being

of a

horse

has

no _{essence of} a cow.

Indeed

this

(the

being

of a

horse)

exists.

They

[the

beings

of a cow and a

horse]

certainly exist with

their

own

characteristics.

In

this

way, all

_properties

certainly exist

_possessing

their

own characteristics.

What

is

the

meaning of

this

_passage?

Here

it

is

decleared

that

all

objects exist

because

they

have

their

own essences, although

they

have

no essence of other objects.

This

interpretation

of nairdtmya

is

nothing other

than

the

itaretara-Sanyata

interpretation,

because

it

is

maintained

in

both

interpretations

that

an object

_possesses

an essence of

its

own

in

spite

of

lacking

the

essence of

other

objects.

As

for

my argument

that

itaretara-jtiayatd

is

advocated

in

Passage

[4],

it

seems

to

be

supported

especially

by

Jfianavajra's

commentary.

He

comments

on

the

_passage

as

follows

:

[5]

debi

don

yah

ji

ltar

snafi shifi myofi

babi

chos

tharns

cad

don

dam

_palji

dfios

po

la

sogs

_par

brtags

_pahi

bdag

_gis

stofi

_pa

fiid

bdag

med

_pabi

don

_yin

gyi!snafi

ba

tsam

_gyi

(orig.

_gyis)

rart

bshin

_gyis

_stofi

_pa

ni ma

_yin

telde

ltar

na _snafi

ba

dari

_{stofi pa zufi}

du

hjug

_pahi sgyu ma

fiid

chos

bdag

med

pahi

don

yin

shes

_pa

stel snafi shih myoh

bafpi

chos

dafilbrtags

_pabi

bdag

fiid

kyah

phan

tshun

gcig

gi

bdag

fiid

gcig

ma

_yin

lalsnafi

ba

tsam

_po

rari

gi

fio

bor

ma

_grub

_pa

ma

_yin

teldebi

_phyir

snari

ba

tsam

gyi

chos can

hid

dgag

bya

ma

_yin

_par

shar

bstan

to!1(pi

249b4-7)

(tr.)

The

meaning of

this

[passage]

is

the

following

one.

The

meaning of

having

no self

(nairatmya)

is

that all the

_properties,

which _{are appeared} and

_perceived,

are void of the essences constructed

imaginarily

as real

beings

a'rthabhava?) _{and so on, and} not

_that

_they

[i.e.

_perceived

_properties]

_are

void of

the

essences which are only appeared.

Thus,

"property's

having

no

self"

(dharmanaira-tmya)

means

_the

illusion

_where appearance and voidness

are combined as a pair.･･･-･･The property, which

is

appeared and

perceived,

and

the

essence, which

is

constructed

imaginarily,

are

different

frorn

each

other,

for

what

is

merely appeared

is

established

to

be

existing with

its

Lafikavatara

on

intaretaraSanyata

(Matsumoto)

(5)

the

locus

(dharmin)

which only appears must not

be

denied.

First

of all,

it

should

be

noted

that,

according

to

Jfianavajra's

comment,

the

mutual

difference

between

a cow and a

horse

stated as an example

in

the

Lafikdvatara

means not merely

the

difference

between

particular

objects,

but

the

difference

between

the

locus

appearing

only

and

the

super-locus costructed

imaginarily

on

the

locus.

It

is

clear

that

Jfianavajra's

comment

here

is

relied on

the

same

_{philosophical}

trend

of

'

the

three

Madhptamika

masters,

Jfianagarbha,

Santarakeita

and

KamalaSila,

who accepted

the

trisvabhjva

theory

in

conventional sense

(ayavahartztas).

For

_{Jfianagarbha,}

what

is

appeared only

(1'i

ltar

snaft

ba)

and

is

capable of

_producing

effects

(don

byed

nzts

Pa)

is

the

Paratantra

and

the

locus

on

which

is

constructed

imaginarily

something realistic such as

the

produc-tion

of objects

in

highest

sense

(don

dam

Pa

Pabi

shye

ba,

tattvikotPat-ti)(i2).

_{Jfianavajra's}

comment

_quoted

above, although

he

is

completely

influenced

there

by

the

_{philosophical}

system of

JfiAnagarbha

and so on,

seems

to

be

useful

in

interpretating

Passage

[4]

of

the

Lahkdvatara.

It

is

stated

in

the

_passage

that

an

object

possesses

the

characteristic

of

its

own

(Svalaksa4a).

This

idea

clearly contradicts

KamalaSila's

explanation

of

lalesarpa-Sanptata

that

all objects

do

not

_possess

any charactericstic

special

[i.e.

of

its

own] or universal.

To

say

that

an objeet

(B)

exists

with

its

own characteristic _{and without}

the

_{characteristic of other objects}

(A)

is

not

different

from

to

state

that

the

characteristic of other objects

(A)

does

not exist

in

the

object

_possessing

its

own

characteristic

(B)

or

in

its

ewn charaeteristic

itself

(B).

Thus,

it

seems

that

in

Passage

[4]

itaretara-Stinyatd

is

not

denied

but

asserted, and

Dhatuvada

is

advocated.

It

seems very

difficult

_to

determine

_what

_the

_{philosophical}

_position

_of

the

Lahleduatara

is.

To

use my

terminology,

this

sutra seems

to

be

a

'

rough composite of

Dhatuvdda

and

Sitayatdvdda,

at

the

first

sight,

However,

if

we

inquire

into

its

content

more

deeply,

we will

be

taught

that

the

main

theme

or

the

real

import

of

this

sutra

is

_{nothing other}

-(6)

Lafikavatara

on

intaretaragtinyata

(Matsumoto)

than

Dhatuvada.

For

example,

in

the

following

sentence, which succeed

directly

to

Passage

_[4],

the

_general

idea

of

Dhdtuvdcla

ean

be

recognized

that

the

teachings

of voidness _{and non-production} and

having

no essence must not

be

understood

in

the

literal

sense of

the

words,

because

they

have

provisional

meanings

(abhiprayika).

[6]

tena

ca

balaprthagjanair

niratmArthatavabudhyate vikalpam upadaya na

tu

avikalparn, _evam

gtinyanutpadasvabhtivyarp

sarvadharrnacam

tavyaml(LAS,

188,

14-16)

(tr.)

Therefore,

the

meaning of

having

no self

is

understood

by

foolish

_people

relying _on

imagination,

_and not on

the

cognition

bereft

of

imagination.

In

this

manner,

the

meanings of voidness and non-production and

having

no

sence ef all

_properties

should

be

understood.

Moreover,

there

are many

_passages

in

the

Lahledvatara

where

Dhatu-vadu

is

advocated clearly.

For

example,

in

the

following

verse, short as

it

is,

the

theoretical

structure of

Dhatuvada

is

fully

expounded :

[7]

yady

ava$tu na

Suddhih

_syat sarpkleSo napi

kasyacit!

_yasrnac

_ca

gudhyate

cittam sarpklegag capi

drSyate/

tasmat

tattvarp

bhaved

vastu viguddham

aryagocaramll

(LAS,

sag.

250,

298,

2--4)

(tr.)

If

the

locus

_(vastpt)

does

not exist,

_purification

and

defilement

have

no

locus

[or

subject].

But,

in

fact,

it

is

_perceived

that

rnind

is

_purified

and

defiled<i3).

Therefore,

the

locus,

which

is

_purified

and

is

the

object of saints, exist as

real.

Here

the

reality of

the

locus

is

proved

by

the

following

argumentation.

If

there

is

no

locus,

there

is

no super-locus.

However,

the

existence

of super-locus

is

_perceived

by

ordinary experience.

Therefore,

the

locus

must

be

existing as real.

This

argumentation

is

the

commom and

indispensable

property

of

Dhatuvada(i4).

As

a conclusion of

this

_parer,

although

its

argumentation seems

to

have

been

insufficient,

I

would

like

to

state as

follows

:

It

is

doubtful

that

the

criticism of

itaretara-Safnptata

in

the

Laftkdvatara

agrees with

the

main

theme

of

the

sutra,

because

Lahkavatara

on

intaretaraglinyata

(Matsumoto)

to

advocate nothing other

than

itaretara-Stinyata.

(7)

(1)

It

seems

that,

to use

the

terminology

by

Tibetan

authors,

laksagea-Sunyata

and

itaretara-Stinyatd

can

be

called rait stofi

(voidness

in

respect to

its

own

essence) and

gshan

stoh

(voidness

in

respect

_to

_others) respectively.

On

the

relation

between

itaretara-Su-nyata

and

gshan

stoh, cf.

Ruegg

D.S.:

La

th6orie

du

tathdgatagarbha

et

du

gotra,

1969,

_pp.319-346.

Prof.

Zuiho

Yamaguchi

once

indicated

that

nonexistence

(abha-va)

means

A's

absense

in

B,

while voidness

(Stinyatd)

is

v'A's

absence

in

A.

This

idea

of

Prof.

Yamaguchi

concerning

the

difference

between

abhdva and

Stinyata-

was very stimulating

to

me,

becasuse

the

fundarnental

structure of

lalesa4a-Saayata-

and

itaretara-Su""nyatd

seems to

have

been

fully

made clear

by

this

idea.

To

use

Prof.

Yamaguchi

expression,

it

can

be

said

that

itaretara-Safnyatd

is

A's

absence

in

B,

while

laksapta-Safnyatd

is

v/A's

absence

in

A.

Cf.

Matsumoto:

Mddhyamilea

_philosophy

in

Tibet

on

the

mTha'

bral

dbza

"za'i

lugs,

The

Tbyo

Galewfutspt

Kenhpt#,

Vol.21,

No.2,

1982,

p.

178,

n.

16.

(2)

Cf.

Ruegg:

oP. cit.,

_p.

343,

n.2.

(3)

The

statement

is

found

in

the

Ptirvapaksa

of

MA

(sa

134b3).

(4)

The

_PtirvaPaksin

of

MA

is

the

Yogacaras.

Cf.

Matsumoto:The

ekaydna

theory

of

the

Maddyamakdteka,

foecrnal

of

SOtb

Skect

Research

"Flellows,

No.12,

1982,

pp.

299-295.

(s)

Cf.

Ruegg:

oP. cit.,

_pp.

320-322;

Takasaki

J.:

Sources

of

the

Lankdvatdra

and

its

position

in

Mahayana

Buddhism,

Indological

and

Buddhist

Stndies,

Canberra,

1982,

p.556.

(6)

Cf.

Bodhisattvabhu-mi

(Wogihara

ed.),

47,

11-13;Ruegg:

oP, cit., _p.

322.

(7)

Cf.

Matsumoto:

The

satyadvaya

_theory

of

Dharmapala,

.1:rBS,

Vol.27,

No.2,

1979,

p.

687.

'

(s)

Cf.

Matsumoto:

On

the ekayjna theory of

the

Srimata-stitra,

.lournal

of

the

Factilty

of

Bnddhism,

No.41,

Komazawa

Univ.,

1983,

p.

405.

(9)

This

commentary

has

many

_problems

concerning

its

_production.

However,

at

least

its

title must

be

corrected.

The

sanskrit title "Tathdgatahrdayalanzlea-ra",

as

is

indicated

in

the

modern catalogues according to the sanskrit title written at

the

beginning

of

the

work,

is

wrong,

judging

from

_the

Tibetan

_title

"De

bshin

gSegs

Pabi

s"ii

Pobi

rgyan".

According

to

its

colophon

(pi

363a8)

and

Bu

ston's report

(Chos

bbyuh,

_ya

160)

this

work was written

by

a

Chinese

master

-(8)

Lathkavatara

on

intaretaragtinyatti

(Matsumoto)

called

Jfitinavajra

(Ye

Ses

rdo rje) and was

translated

from

Chinese

into

tan.

However,

this

seems unlikely,

because

the

author of

this

commentary

knows

well about

the

texts

of

the

later

Indian

Buddhisrn

such as

the

Pramd"avarttika

(pi

21a5

etc.) and

Madityamaka-loka

(pi

137a8

etc.), which were never

translated

into

Chinese.

(10)

"na"

_in

_Nanjio's

edition.

My

reading relied on

Tibetan

version :

de

ni

_yod

de

med

paeafi

ma

_yin

noll

(fiu

143a8).

(11)

1'i

ltar

snaft

ba

tsam

_gyi

chos can ni

khas

len

dgos

la

(TGA

_pi

69a7).

(12)

Cf.

Matsumoto:

The

satyavaya

theory

of

Jfianagarbha,

Balele)iggaku,

Vol.5,

1978,

p.

137,

n.

so.

(13)

In

stateing

that

mind

is

purified

and

defiled,

mind

is

regarded _as

_the

locus

of

the

two

super-loci,

i.e.

purification

(Suddhi)

and

defilement

(samklesa).

I

have

argued

that

in

Dhatuvdda

there

is

a

keen

consciousness

that

mind

cannot

be

the

1ocus.

If

this

argument

is

true,

why

is

rnind stated

here

as

the

1ocus?

I

consider

this

statement

to

be

derived

from

the

special standpoint

of

the

Lahka-vatdra,

i.

e.

the

identification

of

dlayavil'fidna

with

tathagarbha.

Cf.

Matsumoto:

On

the

ekaya-na

theory

of

the

Srima-la-stitra,

_pp.

393-392.

(14)

Cf.

Matsumoto:

The

satyadvaya theory of

Dharrnap'a'la,

p.

687.