You are on page 1of 258

P O L S K A A K A D E M I A N A U K

KOMITET NAUK O R I E N TA L I S T Y C Z N Y C H

ROCZNIK

ORIENTALISTYCZNY

ukazuje si od 1915 r.

TOM LXV
ZESZYT 1

WARSZAWA 2012
Wydawca:

Komitet Nauk Orientalistycznych Polskiej Akademii Nauk

Komitet Redakcyjny:

Marek M. Dziekan (redaktor naczelny), Marta Woniak (sekretarz redakcji)


Jaakko Hmeen Antilla, Agata Bareja-Starzyska, Eduard Gombr, Lidia Kasareo,
Agnieszka Kozyra, Ewa Siwierska, Lidia Sudyka, Gbor Takacs

Rada Redakcyjna:
Janusz Danecki (Warszawa), Eva-Maria von Kemnitz (Lisboa)
Edward Lipiski (Bruksela), Alfred F. Majewicz (Pozna),
Yuri Tambovtsev (Novosibirsk), Piotr Taracha (Warszawa),
Przemysaw W. Turek (Krakw), Vladimir Uspensky (St. Petersburg),
Witold Witakowski (Uppsala)

Address:
Rocznik Orientalistyczny
Faculty of Oriental Studies
University of Warsaw
00-927 Warsaw 64
e-mail: rorient@pan.pl

Copyright by Komitet Nauk Orientalistycznych PAN and Dom Wydawniczy Elipsa

ISSN 0080-3545

Nakad 200 egz.

Realizacja wydawnicza:
Dom Wydawniczy ELIPSA,
ul. Inflancka 15/198, 00-189 Warszawa
tel./fax 22 635 03 01, 22 635 17 85
e-mail: elipsa@elipsa.pl, www.elipsa.pl
Oriental Studies Past and Present
Proceeding of the International Conference
of Oriental Studies
Warsaw 2010

edited by

Agata Bareja-Starzyska
and
Marek Mejor
Spis treci

Jerzy Baczerowski, T L nh, Phonetic and Morphological Coding of Minimal Syntactic


Units in Isolating Languages.................................................. 7
Agata Bareja-Starzyska, Byambaa Ragchaa, Notes on the Pre-existences of the First Khalkha
Jetsundampa Zanabazar according to His Biography Written in the Horizontal Square Script 24
Michael Farris, Lexical and Grammatical Properties of Personal Pronouns in Vietnamese:
Deicticity and Relationships .................................................. 41
Eli Franco, How to Distinguish between Non-existing Entities? Dharmakrti and Prajkaragupta
on Universals as Objects of Knowledge......................................... 51
Szymon Grzelak, Temporal Characteristics of Japanese Verbs from the Contrastive Perspective 63
Joanna Jurewicz, What do Ancient Indian Cosmogonies Tell us about Language?.......... 75
Karnina Kollmar-Paulenz, The Invention of Shamanism in 18th Century Mongolian Elite
Discourse.................................................................. 90
Norbert Kordek, Segmentotactics of Mandarin Chinese................................ 107
Jowita Kramer, Descriptions of Feeling (vedan), Ideation (saj), and the Unconditioned
(asaskta) in Vasubandhus Pacaskandhaka and Sthiramatis Pacaskandhakavibh.. 120
Marek Mejor, Tantric Exposition of the Dependent Origination according to the
Caamahroaatantra, Chapter XVI: prattyasamutpda-paala..................... 140
Mikoaj Melanowicz, Some Aspects of the New History of Japanese Literature............ 149
Nina Pawlak, Morphological Coding of Verb-Object Agreement in African Languages...... 158
Klaus Sagaster, Bible Terminology in Mongolian Translation........................... 171
Tadeusz Skorupski, The Buddhas Stpa and Image. The Icons of his Immanence and
Transcendence.............................................................. 180
Elliot Sperling, Pho-lha-nas, Khang-chen-nas, and the Last Era of Mongol Domination in Tibet 195
Piotr Taracha, Hittitology Up to Date: Issues and New Approaches...................... 212
Jerzy Tulisow, Mongolian and Manchu Studies in Poland ............................. 224
Vladimir Uspensky, Rivalry of the Descendants of Chinggis Khan and His Brother Khasar as
a Factor in Mongolian History................................................ 231
Monika Zin, When Stones are All that Survived: The Case of Buddhism in Andhra........ 236

Lista Autorw / List of authors................................................... 255


R O C Z N I K O R I E N T A L I S T Y C Z N Y, T. LXV, Z. 1, 2012, (s. 723)

JERZY BACZEROWSKI, T L NH

Phonetic and Morphological Coding of Minimal Syntactic Units


in Isolating Languages
(A tentative draft with insights from Chinese and Vietnamese)

Abstract

The problem of identifying minimal syntactically relevant intrasentential units in


isolating languages has not been treated sufficiently thus far within morphological typology.
Such units will be called here syntactic words or tagmons. This article inquires into
properties of tagmons and intends to provide some new insights into the structure of these
units in Chinese and Vietnamese. Two kinds of tagmons of these languages, namely, affixal
and adpositional, are subjected to examination in light of flection, paradigmatification, and
linguistic codematics. This, in turn, together with the partisentential categorizability of
tagmons justifies treating them as relatively coherent integral wholes capable of entering
flectional relations which are controlled by intrasentential syntax. The codematic approach
to tagmons reveals their morphological and phonological patterning.

Keywords: phonetics, morphology, Chinese, Vietnamese, linguistics

1. Introduction

Our article originates from the quandaries in which we have been caught for quite
some time while pondering over the morphological and syntactic status of words occurring
in isolating languages and the possible similarities and differences between units of this
kind and their corresponding counterparts in agglutinative and fusional languages. Our
perplexity tends to increase because of inconclusive linguistic statements unable to resolve
doubts about the adequate description of the units at issue. Thus, for example, among the
properties based on which a given language obtains the status of a member in isolating
8 JERZY BACZEROWSKI, T L NH

languages the lack of morphological complexity of its words figures prominently (cf. Li,
Thompson 1989: 10f). Or, to put it differently, the relative morphological simplicity of
the words has been made to stand out as one of the characteristic features of a language
belonging to the isolating type.
The weakly developed grammatical morphology (syntactic morphology, syntacto-
morphology) of the isolating languages is usually illustrated with the alleged absence of
morphological marking of grammatical relationships which are not indicated by affixes
or by morphological alternations in the word but rather by word order. However, at
the same time, it is admitted that the use of grammatical morphemes such as particles,
adpositions, auxiliary verbs, etc. plays an important role in the indication of grammatical
relationships (cf. Norman 1988: 10, 84). But such a statement is evidently in contradiction
with the supposed lack of morphological marking of the relationships in question, since
grammatical morphemes are not independent words, and consequently as synsemantic
units they enter into the composition of the corresponding fully lexical words in order
to form syntactically manageable objects.
In order to account for the syntactic diversity of intrasentential words, a theory of
minimal syntactic units or tagmons will be offered for consideration. This theory, which
provides for the integration of words grammaticalized to a sufficient degree with the
corresponding fully lexical words whereby syntactically operatable objects are arrived
at, seems to be applicable pan-lingually.
Before the analysis of Chinese and Vietnamese language material is begun, some of
the primitive terms are enumerated and briefly explained. Subsequently, the possibility
of treating affixal and adpositional tagmons of these languages in light of flection,
paradigmatification, and linguistic codematics is considered.
We are aware of the hazardous nature of our non-conventional approach to morphology,
but nothing will keep us from changing or rejecting it, if we become convinced that it
fails to project an adequate image of the morphological reality.

2. Primitive and Some Defined Terms

By way of leveling the ground for our theoretical approach the terminology used
will be clarified first, since it considerably diverges from that traditionally applied in
linguistics. The terms enumerated below have been distinguished as primitive, and they
are as follows:

(i) Lgu the set of all relevant lingual units,


(ii) Fon the set of all sounds (actual phones),
(iii) Mor the set of all actual morphs (morphons),
(iv) Vcb the set of all actual words (vocabulons),
(v) Tgm the set of all syntactic words (tagmons),
(vi) hfn the relation of homophony,
PHONETIC AND MORPHOLOGICAL CODING OF MINIMAL SYNTACTIC UNITS IN ISOLATING LANGUAGES 9

(vii) dsg the relation of designation,


(viii) sgf the relation of signification,
(ix) lkf the relation of lexification,
(x) smf the relation of semification,
(xi) DSM the family of semic dimensions (parameters),
(xii) smc the relation of a not greater degree of semicalization (grammaticalization).

Lingual utterances may be segmented into units of various kinds. The set of lingual
units which are relevant linguistically has been denoted as Lgu. Within this set, among
other units, the following can be distinguished: sounds (actual phones), morphons (actual
morphs), vocabulons (actual vocables), tagmons (actual tagms). The relation of homophony
binds those units which are auditorily indistinguishable.
Not all of the lingual units however are signs. Those functioning as signs are bound
with extralingual entities by the relations of designation and signification. By means of
designation objects are represented as certain wholes, whereas by signification what is
represented are the properties of these objects. These properties are in turn conceived
of as meanings signified by the signs designating these objects. Thus, the relation of
signification binds lingual signs with the corresponding meanings. A meaning being
signified is either lexified or semified. These two modes of meaning signification find
reflection in the relation of lexification and the relation of semification, respectively.
These relations are but two sub-relations within signification. A meaning being
semified is more specific and conspicuous than a meaning being lexified. The set of
the semified meanings or sems can be classified into the family of semic dimensions
(parameters). Among such dimensions the following can be distinguished: Person, Casus,
Numerus, Tempus, Aspectus, etc. Each of these dimensions is a set of homogenous
meanings.
Semification may manifest itself as: (i) affixation, (ii) adpositioning, (iii) particulation,
(iv) auxiliation, (v) sequentialisation. The respective semificators appear as: affixes,
adpositions, particles, auxiliaries, and sequentials (that is, definite positions of the lingual
units within sentences).
The relation of homolexy binds those lingual units which lexify identical meanings
and, analogously, the relation of homosemy binds those lingual units which semify
identical meanings. Semificators may differ or not in the degree of their semicalization
(grammaticalization). In order to account for this property of semificators the relation
of a not greater degree of semicalization can be used. In terms of this relation, the
relation of a smaller degree of semicalization can be defined. Thus, for example, affixes
are more semicalized (grammticalized) than are adpositions, and these latter are more
semicalized than auxiliaries.
The terms vocabulon and tagmon are closely related to the term word. However,
this latter is not sufficiently precise to satisfy our theoretical requirements, and also for
technical reasons it is hardly convenient. Therefore we decided to use these two terms
instead of the term word. These two new terms will denote two different lingual entities.
10 JERZY BACZEROWSKI, T L NH

The vocabulon is conceived of here as the maximal unit of sequential (linear) ordering
within the sentence. Consequently, a linear structure of the sentence can be represented
as a sequence of its constituent vocabulons. The vocabulon is always linearly continuous,
and always functions as a sign. It also seems to be sufficiently conspicuous to native
linguators consciousness. Intrasentential vocabulons are semantically and syntactically
fairly differentiated, that is, they differ with respect to their auto- and synsemantic as
well as to auto- and synsyntactic status.
In the framework of intrasentential syntax or, to put it differently, at the intrasentential
syntactic level, the vocabulonic structure of the sentence is converted into tagmonic
structure. Intrasentential tagmons are syntactically homogeneous units. That is, they are
units which enjoy the same syntactic status, and hence are comparable with each other. They
result thus from the intervention of syntax into the intrasentential vocabulonic structure,
whereby all synsemantic vocabulons, that is, those already grammaticalized to a certain
extent, are attached to the corresponding fully-lexical vocabulons. Consequently affixes,
adpositions, particles, auxiliaries, and the like, are treated syntactically as intratagmonic
constituents. The tagmons emerge thus as minimal syntactic units capable of entering
syntactic relations with other tagmons in order to form intrasentential syntagms.
The operation of tagmonification of intrasentential vocabulons into the corresponding
tagmons may be exemplified with the following Chinese and Vietnamese sentences, which
are, respectively, mutually translatable:

CHINESE:
(2.1) Xushng zi fngjin yng dinno xi xn gi mma.
student in room with computer write letter to mother.
The student is writing a letter to his mother on a computer in a room.
This sentence consists of 9 vocabulons but only of the following 6 tagmons: xushng,
zi fngjin, yng dinno, xi, xn, gi mma.

(2.2) Sh cng zhozi shng dio do dxia le.


book from table on fell to floor ASP.
The book fell off the table onto the floor.
This sentence consists of 8 vocabulons but only of the following 4 tagmons: sh, cng
zhozi shng, dio le, do dxia.

VIETNAMESE:
(2.3) Cu sinh vin vi t th cho m bng my vi tnh trong phng.
CL student write letter to mother with computer inside room.
The student is writing a letter to his mother on a computer in a room.
This sentence consists of 11 vocabulons but only of the following 7 tagmons: cu sinh
vin, vi t, th, cho m, bng my vi tnh, trong, phng.
PHONETIC AND MORPHOLOGICAL CODING OF MINIMAL SYNTACTIC UNITS IN ISOLATING LANGUAGES 11

(2.4) Quy n sch ri t trn bn xu ng sn nh.


CL book fell from on table down floor.
The book fell off the table onto the floor.
This sentence consists of 7 vocabulons but only of the following 4 tagmons: quy n sch,
ri, t trn bn, xu ng sn nh.

3. Tagmons as Lexificators and Semificators

The lexificational and semificational function of tagmons can be examined at the


level of morphs as well as at the level of words (vocables). In Chinese and Vietnamese
lexification is accomplished by means of lexical morphs and words. The morphs, if they
are not bound, coincide with single words. The compound words, in turn, are composed
of at least two lexical morphs. However, we are not primarily interested here in the
types of lexificators but rather in the types of those semificators which contribute to the
creation of tagmons. Nevertheless, for the sake of illustration let us adduce three types
of lexificators.

Table 3.1 CHINESE:


Compound Translation
Types of lexificator Derived from
word in English
noun word + noun bound lexical morph dinno computer electricity-brain
noun bound LM + noun bound LM mxio alma mater mother-school
noun bound LM + affix zhotou omen sign + DAF

Table 3.2 VIETNAMESE:


Compound Translation
Types of lexificator Derived from
word in English
noun word + noun bound lexical morph c voi whale fish-elephant
noun bound LM + noun bound LM giang sn country river-mountain
affix + noun bound LM nh binh army DAF + military

Proceeding now to semification we shall concentrate only on tagmon-forming affixation


and adpositioning.
12 JERZY BACZEROWSKI, T L NH

3.1. Affixation

An affix is a morph which has attained the highest degree of semicalization


(grammaticalization), whence it always is a bound morph. In the following tables,
commonly used affixes (suffixes and interfixes) with their corresponding approximate
English translation are given.

Table 3.3 CHINESE:


Fully-lexical word Translation
Semified meaning Affix
presently in use in English
PLURALITY/COLLECTIVITY - men
- ji ji house, home
PERFECTIVITY/COMPLETION - le lio to finish
EXPERIENTIAL - guo gu to pass, cross
DURATIVITY/CONTINUATION - zhe zho touch, catch
- zi zi exist, remain
POTENTIALITY positive - de - d get, gain, obtain
POTENTIALITY negative - bu - b not

Table 3.4 VIETNAMESE:


Fully-lexical word Translation
Semified meaning Affix
presently in use in English
PLURALITY/COLLECTIVITY nhng -
cc -
mi -
PERFECTIVITY/COMPLETION -, - to be sated
- ri ri last, past
EXPERIENTIAL - qua qua to pass, cross
DURATIVITY/CONTINUATION ang
(ng) -
POTENTIALITY positive - c c get, gain, obtain

However, the morphological status of the affixes, which still have corresponding
homophonous fully-lexical words currently in use, is not clear. This is because their
PHONETIC AND MORPHOLOGICAL CODING OF MINIMAL SYNTACTIC UNITS IN ISOLATING LANGUAGES 13

autosemanticity is further supported by these words. Consequently, they do not seem to


have yet attained the status of true affixes, and it is safer to qualify them as quasi-affixes.

3.2. Adpositioning

Adpositions are words (vocables) semicalized (gramaticalized) to some extent though


less so than affixes. All the Chinese prepositions developed from verbs as a result of the
process of grammaticalization and most of them still have corresponding homophonous
fully-lexical verbs. And, it seems that to some degree their status as verbs may still
be felt in Chinese prepositions, and some grammarians still recognize in them certain
verbal properties (cf. Norman 1988: 161; Packard 2000: 131). Therefore the terms
adpositional verb or deverbal adposition would perhaps be more appropriate, at least
when homophonous fully-lexical verbs coexist with them. For the sake of illustration
several adpositions with approximate English translations and corresponding verbs are
given in the following tables.

Table 3.5 CHINESE:


English English
Preposition
Case-meanings translation of translation of the
(Postposition)
the preposition fully-lexical verb
GENETIVITY/POSSESSIVITY - de
DATIVITY/BENEFACTIVITY gi to, for give
do to, for arrive, reach, go
to
PATIENTIVITY b grasp, take, hold
INSTRUMENTALITY yng with, by use
n with take
ABLATIVITY/ELATIVITY cng from follow
LOCATIVITY/INESSIVITY -li in inside
zi in, at remain, exist
DESTINATIVITY do to, towards arrive
COMMITATIVITY/ gn with, together follow
SOCIATIVITY with
AGENTIVITY (in passive) bi by receive
jio by call
rng by let, allow
gi by give
14 JERZY BACZEROWSKI, T L NH

Table 3.6. VIETNAMESE:


English English
Preposition
Case-meanings translation of translation of the
(Postposition)
the preposition fully-lexical verb
GENETIVITY/POSSESSIVITY ca of
DATIVITY/BENEFACTIVITY cho to, for give
PATIENTIVITY b- suffer, get
c- get, obtain, win
INSTRUMENTALITY bng by, with, in, by
means of
ABLATIVITY/ELATIVITY t from
LOCATIVITY/INESSIVITY at, on, in live, stay
trong in inside
trn on, over
DESTINATIVITY/ n to arrive, come
ALLATIVITY ti to, towards arrive, reach
COMMITATIVITY/ vi with, together
SOCIATIVITY cng together, with
AGENTIVITY (in passive) bi by

4. Flection

We shall proceed from the assumption that flection is present in every language,
regardless of whether this language belongs to fusional, agglutinative, or isolating
morphological types. The differences may concern:
(i) the grammatical dimensions for which lingual units inflect, or
(ii) the significators of grammatical meanings.
The kind of flection considered here will be conceived of as the relation of opposition
binding homolexical and homopartiorational tagmons which differ with regard to one or
more grammatical dimensions. Below we shall briefly concern ourselves with the flection
of Chinese and Vietnamese tagmons for two dimensions, that is to say, for Number and
Case.
Generally speaking, the following two sets of meanings form corresponding numeral
dimensions:
Dm 1 {MASS, COUNT}
Dm 2 {SINGULARITY, DUALITY, PLURALITY}
PHONETIC AND MORPHOLOGICAL CODING OF MINIMAL SYNTACTIC UNITS IN ISOLATING LANGUAGES 15

The meanings enumerated in Dm 2 are in fact sub-meanings of the meaning COUNT


in Dm 1.
The exact specification of the numeral meanings and dimensions for Chinese and
Vietnamese is far from easy. According to some linguists neither of these languages
distinguishes between mass and count nouns. What is more, nouns in these languages
seem to signify neither SINGULARITY nor PLURALITY but rather are unspecified for
Number (cf. Ross; Ma 2006: 22). Let us inspect the following pairs of tagmons:

CHINESE:
(4.1) pngyu friend ~ pngyumen
xushng student ~ xushngmen
tngxu fellow student ~ tngxumen

VIETNAMESE:
(4.2) bn friend ~ cc bn
sinh vin student ~ cc sinh vin
bn hc fellow student ~ cc bn hc
The meaning semified by the Chinese suffix -men is not completely clear. This suffix
often referred to as plural suffix, is being attached to pronouns and human nouns only.
However, tagmons suffixed with -men seem to semify the meaning of COLLECTIVITY,
since they tend to designate groups of people taken together.

(4.3) tngxumen (a certain group of) fellow students


loshmen the teachers (cf. Norman 1988:159)
Regardless whether nouns suffixed with -men semify COLLECTIVITY or PLURALITY,
they are in the relation of flection with the corresponding homolexical nouns without
this suffix, since these latter may be unspecified for Number. Hence, the pair of tagmons
like (xushng, xushngmen) as well as other similar pairs belong to the flection for
Number in Chinese.
The dimension of Case will be conceived of as the set of case-meanings. These
meanings semified by tagmons underlie certain relationships between nouns or between
nouns and verbs within sentences. Tagmons, which semify different case meanings, that
is, which are bound by the relation of opposition with regard to the dimension of Case,
and which are homolexical, enter the relation of flection for Case. What is more, this
dimension will specify the corresponding paradigms.
A paradigm for Case will be defined as the set of all homolexical tagmons which
assume the meanings from this dimension, and belong to the same part of speech. For
the sake of exemplification let us consider the following fragments of tentative paradigms.
16 JERZY BACZEROWSKI, T L NH

Table 4.1 CHINESE:


pngyude (Genitive) chde (Genitive)
b pngyu (Accusative) b ch (Accusative)
bi pngyu by friend bi ch by the car
di pngyu to friend cng ch from the car
gi pngyu to, for friend gn ch with the car
gn pngyu with, from friend do ch li to the car
t pngyu for friend yng ch with the car
xing pngyu from friend zi ch li in the car

Table 4.2 VIETNAMESE:


ca bn b of friends ca my bay of the airplane
bi bn b by friends bi my bay by the airplane
ti bn b to friends t my bay from the airplane
cho bn b to, for friends vi my bay with the airplane
vi bn b with friends ti my bay to the airplane
t bn b from friends bng my bay by the airplane
bn b at/in friends my bay in the airplane

The prepositions semifying the case-meanings do not combine with all nouns, and hence
the resulting paradigms are defective. Perhaps it would be appropriate to treat them as
quasi-paradigms.

5. Morphological Codematics

The subject matter of linguocodematics are codes operating in language and language
communication. The code will be understood here as an operation of association of certain
lingual objects with other objects, lingual or extralingual. Metaphorically it could be said
that by means of language code lingual objects of one kind are converted into objects
of another kind. Thus coding is simply the application of a code.
Morphocodematics as a subdiscipline of linguocodematics is concerned with those
codes which operate in the domain of morphology, that is, which apply to morphological
units such as: tagmons, morphons, vocabulons. Every lingual unit is encoded by means
of all the kinds of its lower level constituent units. Consequently, since tagmons are
decomposable on four levels, they are simultaneously coded in terms of sounds, syllables,
morphons, and vocabulons. In Chinese and Vietnamese the mutual coincidence in phonic
substance of tagmons, vocabulons, morphons, and syllables can often be found. More
precisely, a tagmon may be here very often homophonous with a vocabulon, a morphon,
or a syllable.
PHONETIC AND MORPHOLOGICAL CODING OF MINIMAL SYNTACTIC UNITS IN ISOLATING LANGUAGES 17

The operation of coding tagmons will be conceived of as tagmonification (tgf),


within which, regarding the subtagmonic units, vocablic, morphic, syllabic, and phonic
tagmonification can be distinguished. Each of these operations, if inspected under
a generalized and formal light, could be viewed as a mapping (transformation) of the
powerset of the set of subtagmonic constituents types of one kind onto the set of all
tagmons types, that is, tagms. The creation of tagmons occurs within sentences. If
considered as an intrasentential operation, tagmonification should also provide tagmons
with information on their membership in the respective partisentential categories.
A more abstract operation, namely, the morphotagmification (Mtgf) could be
represented as the following mapping:

Mtgf: pot (MRF) pot (TGM).

This symbolic notation expresses the idea that certain subsets of morphs are combinable
into tagms, that is, they are tagmifiable. Each subset of tagmifiable morphs could be
termed a tagmomorpheme. Consequently, the set pot (MRF) divides into the following
two disjoint subsets:
(i) the set of all tagmomorphemes (TMFM), and
(ii) the set of non-tagmomophemes.
(5.1) For the sake of exemplification, let us consider a subset of Chinese morphs
{Yu, Png, Gn}. This subset is an element of the family pot (MRF), and by
virtue of the operation Mtgf lends itself to being converted into the corresponding
tagmomorpheme {YU, PNG, GN} which already is an object of a higher
morphological status. This conversion is possible in Chinese because there is
at least one sequence created out of the elements of this tagmomorpheme, and
functioning as a tagmic morphotacteme in the shape of GN PNG YU. This
tagmic morphotacteme, in turn, represents each corresponding tagmon in the shape
of gn pngyu. Let us also adduce here a Vietnamese example.
(5.2) A subset of morphs {Bn, B, Vi} (Friend, Collectivity, Commitativity) can
be converted into the corresponding tagmomorpheme {BN, B, VI}, because
there are four sequences with similar meanings (together with friends) created
out of these three elements: (i) VI BN B, (ii) VI B BN, (iii) VI BN
VI B, (iv) VI B VI BN which differ from each other not only in their
morphotactemic structure but also in their stylistic values, and function as the
tagmic morphotactemes representing all the corresponding tagmons in the shape
of vi bn b, vi b bn, vi bn vi b, and vi b vi bn, repectively.
Obviously, in order to form a tagmomorpheme as well as a tagmic morphotacteme
morphs have to satisfy certain requirements which could be formulated in terms of the
corresponding postulates. However, for obvious reasons, we shall refrain from giving
these postulates here.
18 JERZY BACZEROWSKI, T L NH

6. Phonological Codematics

Let us still mention that an approach to the phonic coding of Chinese dictionary
entries (words) was presented in Baczerowski 2009. However, a dictionary whose entries
were solely words given in phonetic transcription and with the exclusion of syntagms
was not available to us. What was available was the MDBG Chinese-English dictionary
(CC-CEDICT) the entries of which were given as ideographs (characters) and transliterated
in pnyn, but which also included certain syntagms. The effective exclusion of these
syntagms from the calculations was not possible. Therefore the calculations worked out
by P. Wierzcho operated with dictionary entries rather than with word-entries. Since the
dictionary did not give its entries in phonetic transcription, the calculation data reflect
the graphotactic rather than the phonotactic structure of the dictionary entries. In other
words, these data refer to the graphic coding of words, performed in terms of the letters
used in pnyn. Nevertheless, some interesting although only approximate inferences
concerning the phonetic coding of words can be made. In the Chinese dictionary 55.699
graphotactemes are encoded by means of 49.997 tactographemes. The results obtained
by Wierzcho for Chinese are given in Table 6.1 presented below. This table consists of
three columns A, B, and C, giving the following information:
(i) in A the number of letter-types, that is, graphs,
(ii) in B the number of tactographemes created out of the corresponding number of
graphs, and
(iii) in C the corresponding number of graphotactemes.
The magnitudes listed in Table 6.1 are represented in Table 6.2 by plotting the
values given in column A on the x-axis, and those given in B and C on the y-axis, in
a system of coordinates. The curves being obtained show the dependency of the number
of tactographemes and graphotactemes upon the number of graphs (letter-graphs) from
which they are created. These curves assume for Chinese the shape of a Gaussian curve.
The number of tactographemes and graphotactemes gradually increases with the increase
of the number of graphs (letter-graphs) up to 7, and then gradually decreases.

7. Breaking the tgf-code

To conclude, let us answer the question of what it means to break the tgf-code.
Generally speaking the tgf-code is broken when the principles underlying this code, and the
corresponding rules of coding, the application of which results in correct target-tagmons,
become known. Consequently, based on this knowledge each tagmon is recognized and
is morphologically, semantically and syntactically decoded.
Thus, practically, the breaking of the tgf-code presupposes knowledge of the
tgfgrammar which would be only one component of a more comprehensive morphological
grammar containing morphotactics.
The language code which underlies the creation of the entirety of language objects
may turn out to be no less complex than the genetic code.
PHONETIC AND MORPHOLOGICAL CODING OF MINIMAL SYNTACTIC UNITS IN ISOLATING LANGUAGES 19

Table 6.1 CHINESE


A (graphicity) B (tactographemicity) C (graphotactemicity)
1 12 12
2 254 265
3 548 604
A 4
(GRAPHICITY) B 719
(TACTOGRAPHEMICITY) C 783
(GRAPHOTACTEMICITY)

5 1 2911 12 3401 12
2 254 265
6 7560 8965
3 548 604
7 4
10043 719
12013 783
8 5 7801 2911 9030 3401
9 6 5439 7560 5742 8965
10 7 4692 10043 4770 12013
11 8 3874 7801 3927 9030
9 5439
12 2995 3028 5742
10 4692 4770
13 11
1645 3874
1666 3927
14 12 665 2995 669 3028
15 13 335 1645 338 1666
16 14 215 665 215 669
17 15 102 335 103 338
16 215
18 60 60 215
17 102 103
19 48 48
18 60 60
20 19 31 48 31 48
21 20 15 31 15 31
22 21 5 15 6 15
23 22 5 5 5 6
24 23 3 5 3 5
24 3 3
49997 55699
49997 55699

Table 6.2CHINESE
Table 6.2 CHINESE

14000

12000

10000

8000

6000

4000

2000

0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

7. BREAKING THE tgf-CODE




20 JERZY BACZEROWSKI, T L NH

8. Concluding Remarks

If the concept of tagmon, as proposed above, turns out to be adequate for the
identification of the minimal intrasentential syntactic units in isolating languages, then
the assertion of the morphological simplicity of syntactic words in these languages should
be revised. More than this, the tagmon as an integral whole functioning as the atom of
syntax, susceptible to flection and paradigmatification, and being multiply coded by means
of all the kinds of its constituent units, allows us to look upon the morphology of the
isolating type in a somewhat different light. Chinese and Vietnamese prefer adpositional
tagmons.

9. Bibliography

Baczerowski, J. 1992. Is generative phonology consequent upon structural phonology. In: Fisiak,J.;
Puppel, S. (eds), 1992, pp. 247307.
Baczerowski, J. 1997. Aspects of general morphology. Studia Germanica Posnansiensia XXIII,
pp. 1346.
Baczerowski, J. 1999. Towards a grammar of flection. Investigationes Linguisticae VI, pp. 584.
Baczerowski, J. 2008. Wyraz w ujciu morphologiczno-typologicznym (A morphologico-typological
approach to words). In: Bednarczuk, L.; Smoczyski, W.; Wojtya-wierzowska, M. (eds). 2008.
Jzykoznawstwo historyczne i typologiczne. W 100-lecie urodzin Prof. Tadeusza Milewskiego. Krakw:
Polska Akademia Umiejtnoci, pp. 233260.
Baczerowski, J. 2009. Aspects of Chinese phonotactics against a comparative background of Polish.
Scripta Neophilologica Posnaniensia X, pp. 722.
Batg, T. 1967. The axiomatic method in phonology. London: Routledge & Kegan Paul.
Bogusawski, A. 1959. O zasadach analizy morfologicznej. Biuletyn Polskiego Towarzystwa
Jzykoznawczego XVIII, pp. 8795.
Bogusawski, A. 1994. Sprawy sowa. Word matters. Warszawa: Veda.
Bogusawski, A. 2010. Dwa studia z teorii fleksji (i inne przyczynki). Warszawa: belstudio.
Borsley, R.D.; Jaworska, E. 1988. A note on prepositions and case-marking in Polish. Linguistic
Inquiry 19, pp. 685691.
Cardire, L. 1958. Syntaxe de la langue vietnamienne. Paris: cole Franaise dExtrme Orient.
Cao, Xun Ho. 1998. Ting vit, my v n ng m, ng php, ng ngha. (The Vietnamese
language some phonetic, grammatical and semantic issues). H Hi: Nxb GD.
Chao, Yuen-ren. 1968. A grammar of spoken Chinese. Berkeley and Los Angeles: University of
California Press.
Chen, H.-C.; Tzeng, O.J.L. (eds). 1992. Language processing in Chinese. Amsterdam: North-Holland
and Elsvier.
Croft, W. 1990. Typology and universals. Cambridge: Cambridge University Press.
Dip, Quang Ban. 1985. Th bn v mt c ch chuyn di t loi trong ting Vit (qua t cho).
Ngn ng 4, pp. 57
Dip, Quang Ban Nguyn, Th Thun. 2000. Li bn v vn cu b ng trong ting Vit.
(Rethinking passive sentences in Vietnamese). Ngn ng 7, pp. 1421.
Di Sciullo, A.; Williams, E. 1987. On the difinition of word. Cambridge: MA:MIT Press.
Doan, T.T.; Nguyen, K.H.; Pham, N.Q. 2001. A Concise Vietnamese Grammar. Hanoi: Hanoi National
University.
PHONETIC AND MORPHOLOGICAL CODING OF MINIMAL SYNTACTIC UNITS IN ISOLATING LANGUAGES 21

Du, N.H. 2005. Ng Php Ti ng Vit (Ph n t loi). Ho Chi Minh City: University of Education Press.
Duanmu, San. 2007. The phonology of standard Chinese. Oxford: Oxford University Press.
inh, Vn c. 2001. Ng php ti ng vit (Vietnamese Grammar). H Ni: Nxb HQG HN.
Fillmore, Ch.J. 1968. The case for case. In: Bach, E.; Harms, R.T. (eds). 1968. Universals in
linguistic theory. New York: Holt, Rinehart & Winston, pp. 188.
Fisiak, J.; Puppel, S. (eds). 1992. Phonological investigations. Amsterdam/Philadelphia: John Benjamins.
Goddard, C. 2005. The languages of East and Southeast Asia. Oxford: Oxford University Press.
Greenberg, J.H. 1960. A quantitative approach to the morphological typology of languages. International
Journal of American Linguistics 26, pp. 178194.
Grzegorczykowa, R.; Laskowski, R.; Wrbel, H. (eds). 1998. Gramatyka wspczesnego jzyka
polskiego. Morfologia, t. 12, Warszawa: Wydawnictwo Naukowe PWN.
Haspelmath, M.; Knig, E.; Oesterreicher W.; Raible, W. (eds). 2001. Language typology and language
universals. An international handbook, vol. 1, Berlin-New York: Walter de Gruyter.
Havas, F. 1974. A magyar, a finn s az szt nyelv tipolgiai sszehasonltsa. Budapest: Akadmiai
Kiad.
Heinz, A. 1988. Pozycja kategorii morfologicznych w jzyku. In: Heinz, A. 1988. Jzyk i jzyko-
znawstwo. Warszawa: Pastwowe Wydawnictwo Naukowe, pp. 312320.
Hong, Vn Hnh; H, Quang Nng; Nguyn, Vn Khang. 1998. T ti ng Vit: Hnh thi cu
trc t ly t ghp chuy n loi. H Ni: Nxb. KHXH.
Hoosain, R. 1992. Psychological reality of the word in Chinese. In: Chen, H.-C.; Tzeng (eds).
1992, pp. 111130.
H, L. 1976. Vn c u to t ca ti ng Vit hin i. H Ni: Nxb KHXH.
Jones, D. 1950. The phoneme: Its nature and use. Cambridge: Heffer.
Jones, Robert B., Jr.; Hunh, Sanh Thng. 1960. Introduction to Spoken Vietnamese. Washington,
D.C.: American Council of Learned Societies.
Karlsson, F. 1985. Paradigms and word forms, In: Laskowski, R. (ed.). Studia gramatyczne VII.
Wrocaw etc.: Zakad Narodowy im. Ossoliskich, pp. 135154.
Karolak, S. 1965. Przypadek i przyimek. Biuletyn Polskiego Towarzystwa Jzykoznawczego XXIII,
pp. 143158.
Kastovsky, D. 1985. Typological changes in the nominal inflectional system of English and German. In:
Laskowski, R. (ed.), Studia gramatyczne VII. Wrocaw etc: Zakad Narodowy im. Ossoliskich, pp. 97117.
Kastovsky, D. 1992. Typological reorientation as a result of level interaction: the case of English
morphology. In: Kellermann, G.M.; Morrissey, M.D. (eds). Diachrony within synchrony: Language history
and cognition, (Duisburger Arbeiten zur Sprach-und Kulturwissenschaft 14). Franfurt/M.: Lang, pp. 411428.
Kastovsky, D. 1994. Typological differencies between English and German morphology and their
causes. In: Swan, T. et al. (eds). 1994, pp. 135157.
Kempf, Z. 1978. Prba teorii przypadkw. Opole: Zakad Narodowy im. Ossoliskich.
Kramsky, J. 1969. The word as a linguistic unit. The Hague, Mouton.
Kuryowicz, J. 1948. Struktura morfemu. Biuletyn Polskiego Towarzystwa Jzykoznawczego VII,
pp. 1028.
Laskowski, R. 1975. Uwagi o morfemie. Biuletyn Polskiego Towarzystwa Jzykoznawczego XXXIII,
pp. 107114.
Laskowski, R. 1987. On the concept of the lexeme. Scando-Slavica 33, pp. 169178.
L, Bin, 1998. T loi ti ng Vit hin i. H Ni: Nxb Gio dc.
L, nh T. 2008. Ng php ti ng Vit (Vietnamese Grammar). H Ni: H H Ni.
L, nh T. 2010. W poszukiwaniu adekwatnego podejcia do tzw. kategorii czasu (Quest towards
an adequate approach to the category of tense). Investigationes Linguisticae XX, pp. 4059.
L, Vn L. 1968. S tho ng php Vit Nam. Si Gn: Trung tm hc liu xut bn.
Li, Charles N.; Thompson, Sandra A. 1989. Mandarin Chinese. A functional reference grammar.
BerkeleyLos AngelesLondon: University of California Press.
22 JERZY BACZEROWSKI, T L NH

L, Ton Thng. 1997. Loi t v cc ti u loi danh t trong ti ng vit. (Classifiers and classification
of parts of speech in Vietnamese). Ngn ng 2, pp. 113.
Matthews, P.H. 1974. Morphology. An introduction to the theory of word structure. Cambridge:
Cambridge Univ. Press.
Milewski, T. 1965. Jzykoznawstwo. Warszawa: Pastwowe Wydawnictwo Naukowe.
Nguyn, Cao m. 1994. H hnh c php trong mt s kiu cu n hn nh trong ting Vit,
In: Lu, Vn Lng (ed.). 1994. Nhng v n ng php ti ng Vit hin i. H Ni: Nxb. KHXH,
pp. 109120.
Nguyn, c Dn; L, ng. 1985. Phng thc lin kt ca t ni. Ngn ng 1, 3240.
Nguyn, Kim Thn. 1977. ng t trong ti ng vit. (The verb in Vietnamese). H Ni: Nxb KHXH.
Nguyn, Th Ly Kha. 1998. S h ha ca danh t hay l s chuyn loi t danh t thnh gii t.
Ng hc tr 98. Hi Ngn ng hc Vit Nam, pp. 6065.
Norman, J. 1988. Chinese. Cambridge etc.: Cambridge University Press.
Pan, Junlin. 1993. The nature of Chinese adpositions and their constituent order. EL Two Talk,
vol. 1, No 1, Spring, pp. 1627.
Parkard, Jerome L. 2000. The morphology of Chinese: A linguistic and cognitive approach. Cambridge:
Cambridge University Press.
Parkard, Jerome L. (ed.). 1997. New approaches to Chinese word formation: Morphology, phonology
and lexicon in modern and ancient Chinese. Trends in Linguistics Studies and Monographs 105. Berlin
and New York: Mouton de Gruyter.
Pawlak, N. 1990. Expressing spatial relations in the Hausa language. Orientalia Varsoviensia 3,
pp. 59110.
Ross, C.; Ma, Jing-heng Sheng. 2006. Mordern Mandarin Chinese grammar. London and New York:
Routledge.
Selkirk, E. 1982. The syntax of words. Cambridge, MA: MIT Press.
Smyth, D. 2002. Thai: An essential grammar. London: Routledge.
Spencer, A. 1991. Morphological theory: An introduction to word structure in generative grammar.
Oxford, UK; Cambridge, Mass., USA: Basil Blackwell.
Sproat, Richard; Shih, Chilin. 1996. A corpus-based analysis of Mandarin nominal root compound.
Journal of East Asian Linguistics 5, pp. 4971.
Swan, T.; Mrck, E.; Westvik, O.J. (eds). 1994. Language change and language structure: older
Germanic languages in a comparative perspective. Berlin: Mouton de Gruyter.
Tokarski, J. 1971. Sownictwo. Teoria wyrazu. Warszawa: PZWS.
Trng, Vn Chnh; Nguyn, Hin L. 1963. Kho lun v ng php Vit Nam. (A treatise on
Vietnamese Grammar). Hu: i hc Hu.
Weinsberg, A. 1973. Przyimki przestrzenne w jzyku polskim, niemieckim i rumuskim (Spatial
prepositions in Polish, German and Romanian). Wrocaw: Zakad Narodowy im. Ossoliskich.
Wierzcho, Piotr. 2004. Gramatyka diakrytologiczna. Studium ortograficzno-kwantytatywne. Pozna:
Wydawnictwo Naukowe UAM.
Wierzchowski, J. 1968. Jednostka struktury jzyka wyraz. Biuletyn Polskiego Towarzystwa Jzy-
koznawczego XXVI, 197206.
Williams, E. 1981. On the notions lexically related and head of word. Linguistic Inquiry 12,
pp. 145174.
irmunskij, V.M. 1966. The word and its boundaries. Linguistics 27, pp. 6591.
irmunskij, V.M.; Arutjunova, N.D. (eds). 1970. Morfologieskaja struktura slova v indoevropejskich
jazykach. Moskva: Nauka.
irmunskij, V.M.; Sunik, O.P. (eds). 1963. Morfologieskaja struktura slova v jazykach razlinych
tipov. MoskvaLeningrad: Izdatelstvo AN SSSR.
Zabrocki, L. 1961. Sprachkode. Zeitschrift fr Phonetik, Sprachwissenschaft und Kommunikations-
forschung 14/1, pp. 6473.
PHONETIC AND MORPHOLOGICAL CODING OF MINIMAL SYNTACTIC UNITS IN ISOLATING LANGUAGES 23

Zabrocki, L. 1962. Phon, Phonem und distinktives Morphem. Biuletyn Fonograficzny V, pp. 5987.
Zhou, X.; Murslen-Wilson, W. 1994. Words, morphemes and syllables in the Chinese mental lexicon.
Language and Cognitive Processes 9, pp. 393422.
Zhou, X.; Murslen-Wilson, W. 1995. Morphological structure in the Chinese mental lexicon. Language
and Cognitive Processes 10, pp. 545600.
Zwicky, A. 1985. Clitics and particles. Language 61, pp. 283305.
R O C Z N I K O R I E N T A L I S T Y C Z N Y, T. LXV, Z. 1, 2012, (s. 2440)

AGATA BAREJA-STARZYSKA, BYAMBAA RAGCHAA

Notes on the Pre-existences of the First Khalkha Jetsundampa


Zanabazar according to His Biography Written in the Horizontal
Square Script

Abstract

The article deals with previous incarnations (pre-existences) of the first Khalkha
Jetsundampa Zanabazar (16351723) called ndr Gegeen according to the biography
written in the Horizontal Square Script invented by him around 1686. Main information
about the manuscript and its peculiarities is provided as well as comparison of several
lists of previous incarnations of the Jetsundampas.

Keywords: Mongolian Buddhist incarnations, Jetsundampa, Zanabazar, Horizontal Square


Script

Life of the first Khalkha Jetsundampa Zanabazar (16351723) or Blo bzang bstan pai
rgyal mtshan, known as ndr Gegeen Zanabazar, the eminent Buddhist master, artist
and Khalkha leader1 was described in various biographies. They were written in Tibetan
language either during his lifetime, such as a biography written in 1702 by his disciple,
Khalkha Zaya Pandita Luvsanprinlei (Blo bzang phrin las, 16421715)2 or many years

1 There are numerous publications about this important figure, both in Mongolian and other languages.
2 It is included in the (Sha kyai btsun pa blo bzang phrin las kyi) zab pa dang rgya che bai dam pai chos
kyi thob yig gsal bai me long, i.e. Clear Mirror, Vol. IV (nga), standard Peking xylograph, folios 62v6-77v2
(reproduced by Lokesh Chandra 1982, pp. 124154. There exists a separate bilingual, Tibeto-Mongolian version
entitled: Blo bzang bstan pai rgyal mtshan dpal bzang poi khrungs rabs bco lngai rnam thar. It is included in
Lokesh Chandra 1982, pp. 411549.
NOTES ON THE PRE-EXISTENCES OF THE FIRST KHALKHA JETSUNDAMPA ZANABAZAR... 25

later.3 Besides biographies written in Tibetan there were also texts composed in Mongolian.
The best known is the biography ndr gegen- namtar, examined and published with
English translation by Charles Bawden (Bawden 1961).4 All these biographies have
been studied by scholars, however, they still pose many questions and require in-debth
comparative analysis.
Moreover, one of Zanabazars biographies deserves special attention. It is a short text
entitled: K-tg-t rje-btsin dam-pa bla-mai a-dig ro-i-ba5 written in the Horizontal
Square Script, which was invented by Zanabazar circa 1686 (Byambaa 2005, p. 9). The
biography was discovered by Byambaa Ragchaa while collecting texts written in the
Horizontal Square Script among materials presented to him by S. Agvanlodoi (19091992).
No details about the texts author, composition or transmission of this particular manuscript
are known. It seems to be the only extant copy of the text (Byambaa 2005, p. 65).
This biography belongs to a special genre of Tibetan Buddhist writings, namely to
accounts of previous incarnations (pre-existences) of a given master. Usually they are based
on a prayer to the master called gsol debs. It is a kind of a reverential petition or prayer
which includes a list of previous incarnations and may be combined with a biography
of the master, as it is for example the case with the text about Padmasambhava: gSol
debs rnam thar dri med ma by Nyang ral nyi mai od zer.6
In the case of the first Jetsundampa we can find gsol debs prayer of his previous
incarnation, i.e. Trantha Kun dga snying po (15751634) written in Tibetan language
and entitled: rJe btsun Tranthai khrungs rabs gsol debs smon lam dang bcas pa
included in the Collected Works (gsung bum) of ndr Gegeen Zanabazar.7 Comparison
of this gsol debs prayer with the text of the biography written in the Horizontal Square
Script revealed that the two texts were quite similar, but not identical. More information
about the comparison will follow.
The text under examination written in the Square Script includes in its title the word
adi or adig, which is a Mongolian word based on the Sanskrit term jtaka, which
denotes story of a former incarnation of the Buddha or a Buddhist saint (Lessing 156a).
It was used in Mongolian language to represent life-stories of the Buddha or eminent
Buddhists, regarded as enlightened beings. With passing of the time it also started to
denote biography or life-story of any being.

3 Byambaa 2006, Bareja-Starzyska 2009.


4 Although some of the manuscripts described by Bawden do not bear the title. See also editions by Dashbadrakh
1995, Khrelbaatur 2009, papers by Futaki 2011, Bareja-Starzyska 2010, 2011.
5 Facsimile of the manuscript together with transliteration, study and Modern Mongolian Khalkha translation

are published in Byambaa 2005, pp. 6583. However, the book is not easily available and since information about
this biography in English is not widespread the authors have decided to write the present paper focusing on the
subject of Zanabazars pre-existences and to include new details.
6 Zaya Pandita in his Clear Mirror also included Jetsundampas gsol debs (f. 30a) before presenting his

biography (f. 63a), see Lokesh Chandra 1982.


7 Byambaa 2006, p. 7, No 00006. In the Urga edition entitled: rJe nyid kyi gsang bai rnam thar tshigs bcad

sngon tshe bde gshegs ma pachen chos rgyan gyis mdzad pai rnam thar gsol debs dang bcas.
26 AGATA BAREJA-STARZYSKA, BYAMBAA RAGCHAA

Hence, it is not the same etymologically as the Tibetan term rnam thar, which means
exemplary liberation or perfect life-story serving as a model for liberation from the
cycling existence. However, both Mongolian adig and Tibetan rnam thar served as
terms to denote biography or better yet hagiography written from the point of view of
Buddhists. Therefore, while translating this word into English, terms such as biography,
hagiography or life-story can be used interchangeably to denote the literary genre
which it represents.
The title K-tg-t rje-btsin dam-pa bla-mai a-dig ro-i-ba consists of Mongolian
and Tibetan words: k-tg-t for Classical Mongolian qututu, rje-btsin dam-pa bla-mai
should be rendered as rje btsun dam pa bla mai in Classical Tibetan. The word ro-i-ba
written in the Square Script stands for Class. Mong. oroiba and can be translated as is
presented. Therefore the whole title, which is given in the hybrid Mongolian-Tibetan,
can be translated into English as: The Biography of the Holy Jetsundampa(s). Such
hybrid nature of writing is characteristic for the whole manuscript which is written in
the Classical Mongolian language, however, includes many Tibetan words, mainly names
and titles.

K-tg-t rje-btsin dam-pa bla-mai a-dig ro-i-ba


as a monument of the Horizontal Square Script

The Horizontal Square Script was invented in order to facilitate writing in at least
three languages: Mongolian, Tibetan and Sanskrit.8 It is necessary to explain here that
the Horizontal Square Script is a syllabical script which is more fitting to write down
the Tibetan and Sanskrit languages since they too employ syllabical scripts, than to
write in Mongolian. The 17th century Mongolian language was written in the Classical
Mongolian script based on the Arameic script and emphasizing front and back vowels
opposition. The Classical Mongolian language did not reflect the actual pronunciation of
its users but only conventional orthography. Therefore, while writing Mongolian words
in a different script, such as the Horizontal Square Script, a scribe could choose between
conventional orthography of Classical Mongolian and the actual pronunciation reflecting
a particular Mongolian dialect.
Also, the Horizontal Square Script uses at times a different set of signs for the
three languages: Tibetan, Sanskrit and Mongolian. The same sign may be employed
to denote different sounds depending on the language it represents, for example: ga in
Mongolian, but ka in Tibetan and in Sanskrit; da in Mongolian, ta in Tibetan and in
Sanskrit, while a in Mongolian and in Tibetan (sha), but a in Sanskrit. It was noted
that in that particular manuscript a scribe used a dot under the sign to mark that the
consonant should not be vocalized.

8 Detailed study of the Horizontal Square Script was presented in Byambaa 1997 and in the improved and

enlarged second edition: 2005. See also Shagdarsren 1981.


NOTES ON THE PRE-EXISTENCES OF THE FIRST KHALKHA JETSUNDAMPA ZANABAZAR... 27

While examining this particular manuscript of Jetsundampas biography (K-tg-t


rje-btsin dam-pa-bla-mai a-dig ro-i-ba) one is struck by the flexibility of using letter
signs by the scribe. Mongolian words are written with the use of signs designed for
both Mongolian and Tibetan. This leads to ambiguous readings of words, which hardly
represent the standard Classical Mongolian orthography. On the other hand the use of the
syllabical script for copying Mongolian text provides many interesting phonological data,
especially when it reflects the authentic pronunciation. For example, quite many words
are written with long vowels, such as a-ri-n (clear, clean), Class. Mong. ariun,
Modern Khalkha ariun or ko-g-son (empty), Class. Mong. qoosun, Modern Khalkha
khooson. The forms with long vowels indicate probably the actual pronunciation and
though /g was not always dropped, there are changes in pronunciation which may lead
to omission of /g There are many other examples very interesting from the point of
view of linguistics. However, in the case of the examined manuscript it is very hard to
draw definite conclusions regarding Mongolian phonology which it represents. The same
word is written in two or even more different ways, for example: Tibet or Tibetan are
written as: to-bed, t-ped, t-bed, te-bd, te-bed, t-bod etc.9 Therefore, it is also difficult
to determine the date of the composition of the text and this particular manuscript on the
basis of grammatical forms and phonological peculiarities which it presents.
It is possible to conclude at this phase of the research that the scribe was probably
not very well acquainted with Mongolian orthography and probably did not spend much
time reflecting on the way of writing and choosing the adequate letter signs. He knew
Tibetan language well since the Tibetan words are usually written with more care and
with less mistakes. It seems quite probable that he was a Mongolian monk who was well
versed in Tibetan and who did not know Classical Mongolian script well, or perhaps at
all. The Horizontal Square Script does not require distinguishing between front and back
vowels and consonants connected with them as in the case in the Classical Mongolian
in which back vowels require certain consonants (such as q and ) and should not be
written with others (g etc.). Therefore, we may assume that the scribe chose freely
between several signs of the Horizontal Square Script, even while writing the same
words. Nevertheless, we may observe the frequent use of long vowels which may point
to colloquial pronunciation of that time, for example: j-k (f. 2b1), Class. Mong. jauqu,
Modern Khalkha zuukh. The form used in the text is closer to the Khalkha form.10 Itis
also possible to propose here an improvement in exposition of the Horizontal Square
Script for writing in Mongolian, such as the transcription of the first sign of the syllabical
table (alphabet) not only as k and g, but also q and , see qututu.

9 Byambaa 2005 pp. 6869 discusses peculiarities of the use of the script in the manuscript and provides
examples. See also Shagdarsren 1981.
10 The use of the final long vowel here is not clear, but it may serve to indicate the case ending. Byambaa

(2005, p. 74) translates it in the Instrumental case as zuuhaar in order to bite.


28 AGATA BAREJA-STARZYSKA, BYAMBAA RAGCHAA

Pre-existences of the First Jetsundampa ndr Gegen Zanabazar

According to the biography in the Horizontal Square Script the list of pre-existences
is as follows:
1. Ma-i tein a-ri-gn o-yo-tu = Shin tu blo gros rgyal mtshan (Suviuddhamati)
2. ma-lai Ba-da-ra-gi = Bar bai gtso bo
3. ye-ke i-di-t Ka-la-ca-ra-ya = grub chen Nag po spyod pa (Mahsiddha Kcrya)
4. ye-ke Rat-na = Ratna chen po
5. Dar-ma-ba-da-ra = Rong zom chos bzang (Dharmabhadra)
6. Dar-ma-dbang-phyugs = Dar ma dbang chug
7. Ge-rel-cog-tu = Od zer dpal
8. L da-go-to do-ja11 = Brug sgra rgyal mtshan
9. Bor-ga ye-ke bu-t = Sang rgyas ras chen
10. Sa-g-ba-dar = Sanghabhadra
11. Jam-ya or-je = Jam dbyangs chos rje
12. Coi-gyi in-byed = Chos kyi nyin byed
13. G-ge-sgrol-og = Kun dga grol mchog
14. Gaa-byed-sa-sgyong = Dga byed sa skyong
15. Gn-gas i-bo-gra-es-rgyal-an = Kun dga snying po bkra shis rgyal mtshan
(Trantha)
Here fifteen names of incarnations from the time of the Buddha until Trantha are
provided. Therefore, Zanabazar appears as the sixteenth incarnation.

According to the gsol debs of Trantha which is included in the Collected Works
(gsung bum) of Zanabazar under the title: rJe btsun Tranthai khrungs rabs gsol debs
smon lam dang bcas pa the list of previous incarnations is as follows:
1. Blo gros shin tu rnam dag
2. Bar bai gtso bo
3. sPyod pai rdo rje
4. Ra tna chen po
5. Rong zom chos bzang
6. Dar ma dbang phyug
7. Od zer dpal
8. Brug sgra rgyal mtshan
9. Sangs rgyas ras chen
10. Sang gha bha dra
11. Jam dbyangs chos rje
12. Chos rje nyin byed
13. Jam mgon bla ma

11 Byambaa 2005, p. 75 renders this name in Modern Khalkha as Luu duut duaz and identifies him with the

Tibetan name Brug sgra rgyal mtshan.


NOTES ON THE PRE-EXISTENCES OF THE FIRST KHALKHA JETSUNDAMPA ZANABAZAR... 29

14. dGa byed sa skyong


15. rJe btsun rdo rje dzin (i.e. Trantha)
The list consists similarly of fifteen incarnations. And similarly Zanabazar would be
the sixteenth embodiment. The main difference is in position 13, which is Jam mgon
bla ma according to the gsol debs and Kun dga grol mchog according to the biography
in the Horizontal Square Script. More about this will be explained further in this work.
If we assume that the gsol debs was written by Zanabazar himself and on that basis
included into his Collected Works (gsung bum),12 we may regard this version of the
list as the most genuine.

Another list of the Khalkha Jetsundampas previous incarnations was composed by


the eminent Tibetan scholar Londol Lama (Klong rdol bla ma, 17191794):
1. Gyad stobs chen (dang po = 1) not mentioned in other cited sources
2. Blo gros shin tu rnam dag (gnyis pa = 2)
3. Bar bai gco bo (gsum pa = 3)
4. Mi bskyod rdo rje (bzhi pa = 4) not mentioned in other cited sources
5. Ratna chen po (lnga pa = 5)
6. Rong zom chos bzang (drug pa = 6)
7. Sangs rgyas ras chen (bdun pa = 7)
8. Od zer dpal (brgyad pa = 8)
9. Brug sgra rgyan (dgu pa = 9)
10. (Not mentioned. According to other sources should be Dar ma dbang phyug)
11. Sangga bhadra (bcu gcig pa = 11)
12. Jam dbyangs chen po rje bkra shis (bcu gnyis pa = 12)
13. Nyin byed bzang po (bcu gsum pa = 13)
14. Kun dga grol mchog (bcu bzhi pa = 14)
15. dGa byed sa skyong (bco lnga pa = 15)
16. Jo nang ta ra na thaam Kun dga snying po (bcu drug pa = 16)
17. Khal kha rje btsun dam pa Blo bzang bstan pai rgyal mtshan (bcu bdun pa = 17)
Zanabazar was the seventeenth incarnation according to this list. There are evident
omissions and repetitions on it. The nature of mistakes was discussed by G. Somlai (1988,
pp. 449451). Since names are given together with ordinal numerals it is easy to observe
that number ten was omitted. Somlai concluded also that the lineage given by Klong rdol
contains two mistakes. First he left the name of Bab/Bha rom pa Dar ma dbang phyug.
Second that he (or his copyist) wanted to remedy this mistake and having forgotten the
name of Bab/Bha rom pa Dar ma dbang phyug, took that of Sangs rgyas ras chen in
its place. However, even G. Somlais explanations require further corrections. Number
one, Gyad stobs chen was not mentioned in other sources. His inclusion made Blo gros
shin tu rnam dag appear as the second incarnation instead of being the first on the list

12 Byambaa 2004, p. 7 quotes the colophon, in which his name is given: Blo bzang bstan bai rgyal mtshan

rang byung gi ming Janana bajra.


30 AGATA BAREJA-STARZYSKA, BYAMBAA RAGCHAA

in other sources. Number four, Mi (b)skyod rdo rje, may possibly (?) refer to the eighth
Karmapa (15071554) and was not listed in other sources. The name might have been
included in Londol Lamas list to confirm connection of the Jetsundampa lineage with
the Karma Kagyu tradition. Certainly the Londol Lamas list requires a separate study.
We may propose a working hypothesis that with passing of the time the list of previous
incarnations of ndr Gegen was changed due to different reasons: scribal mistakes
and/or problems connected with Zanabazars affiliation. Since he was recognized as the
reincarnation of a Jonangpa master, Trantha, he might have served as an important link
in the Jonangpa transmission. However, as it was explained elsewhere,13 there are no
traces of his activity as a Jonangpa lineage holder. All information that is at our disposal
points to his Gelugpa education and affiliation.
Lists of the First Jetsundampas previous incarnations are included also in other
writings, such as for example a composition entitled Boda ibundamba-yin angqan
trl-n ner-e orusibai. A manuscript of this text is preserved in the Central State (Now
National) Library of Ulan Bator.14
Here (pp. 13), the list of Jetsundampas previous incarnations is as follows:
1. Lodoisintunamda = Blo gros shin tu rnam dag
2. Bandida Barbajobo = Pandita Bar bai gtso bo
3. Naabuujodba = Nag po spyod pa
4. Tbcen Ratnaebegle = Ratna chen po
5. Rongsongcoisang = Rong zom chos bzang
6. Daramvangcu = Dar ma dbang phyug
7. Naljorba Uunserbil = Od zer dpal
8. Brudajamjin = Brug sgra rgyal mtshan
9. Sangnairanijin = Sangs rgyas ras chen
10. Sanghadari = Sanghabhadra
11. Tasibaldan = Bkra shis dpal ldan
12. Coyijininjid = Chos kyi nyin byed
13. Gungadolchog = Kun dga grol mchog
14. Gajidsajung = Dga byed sa skyong
15. Jibcuntaranatan = rJe btsun Trantha
This list is almost identical with the list written in the Horizontal Square Script.
In the biography of Zanabazar written by his disciple Zaya Pandita during the formers
lifetime in 1702 Jetsundampas gsol debs entitled as rJe btsun blo bzang bstan pai
rgyal mtshan dpal bzang poi khrungs rabs gsol debs precedes the actual biography
of Zanabazar.15 Zaya Pandita starts his list of pre-existences with Bar bai gtso bo and

13 Bareja-Starzyska 2010b.
14 Under call no. 5308/96 (on the cover no 6324). The manuscript occupies 80 pages. There are additional
pages emended to the main text. Zanabazars biography is described on pages 426, but occasionally there is also
information about him elsewhere in the text (for example on p. 30).
15 See Byambaa 2005, p. 67. It starts with folio 30r6 of the IV vol. of Zaya Panditas Clear Mirror. It can

be edited separately, see for example its reprint in Byambaa 2004b.


NOTES ON THE PRE-EXISTENCES OF THE FIRST KHALKHA JETSUNDAMPA ZANABAZAR... 31

therefore concludes with Zanabazar as the fifteenth reincarnation. This agrees with the
title of the biography by Zaya Pandita: bco lnga pa rje btsun [+ blo bzang] [414-1] bstan
pai rgyal mtshan dpal bzang poi thung mong ma yin pai rnam thar pa.16
Also a list of the Jetsundampas previous incarnations appears in the biography written
in Tibetan in 1847 by Ngag dbang blo bzang don grub.17
This one is as follows (f. 10a3-12b1):
1. Blo gros shin tu rnam dag
2. Bar bai gtso bo
3. Grub chen Nag po spyod pa
4. Ratna chen po
5. Rong zom chos bzang
6. Dar ma dbang chug
7. Od zer dpal
8. Brug sgra rgyal mtshan
9. Sang rgyas ras chen
10. Sangha bha dra
11. Jam dbyangs chos rje
12. Chos kyi nyin byed
13. Kun dga grol mchog
14. Dga byed sa skyong
15. Trantha
According to Byambaa Ragchaas study the relevant fragment concerning previous
incarnations is closest to the biography in the Horizontal Square Script (Byambaa 2005
p. 67).
Having compared the list presented in gsol debs (known from Zanabazars Collected
Works) with lists of previous incarnations of Zanabazar in the above mentioned texts,
excluding Londol Lamas list which differs at several points, it is clear that the difference
concerns mainly one name in the position 13. All sources except gsol debs list there Kun
dga grol mchog (15071566), an important master in the transmission of the Klacakra
initiation according to the Jonangpa tradition.18 His name was mentioned in Tranthas
own writings, namely: dPal ldan shangs pai chos skor gyi byung khungs yid kyi mun sel.19
As it was stated above, in the biography written in the Horizontal Square Script
the name is given as Kun dga grol mchog, but in Tranthas gsol debs by the first
Jetsundampa it is written as Jam mgon bla ma, which may be a polite address form which
can be used when referring to lamas belonging to different traditions.20 It is important to

16 Lokesh Chandra 1982, pp. 413414.


17 Byambaa 2004b, f. 10a3-12b1.
18 See Sheehy 2009.
19 Vol. 34, pp. 194285 of his Collected Works, dpal brtsegs bod yig dpe rnying zhib jug khang edition, and

other sources. [TSD] http:/rywiki.tsadra.org/index.php/Jetsun-Kunga-Drolchog.


20 It is most often used while referring to the Sa skya hierarchs or to Tsongkhapa, the most important master

of the Gelugpa tradition, see Nitartha. However, it was also used as an address to Trantha, see TBRC, Jo nang
32 AGATA BAREJA-STARZYSKA, BYAMBAA RAGCHAA

recall that the recognition of the Jonangpa master Tranthas incarnation in the person
of a Khalkha Mongolian khans son seems to be a very intriguing solution. More detailed
explanations were presented in Stearns 1999, pp. 7172 and Bareja-Starzyska 2010b,
pp. 246247. Here we may speculate that perhaps it was not proper for Zanabazar to
mention Kun dga grol mchogs name in gsol debs composed by him at the time of
Jonangpas persecutions by Gelugpa, even if he was recognized as Tranthas incarnation.
It was stressed before (Bareja-Starzyska 2010b) that ndr Gegeen Zanabazar followed
closely Gelugpas teachings. And therefore, instead of the name of the Jonangpa incarnation,
name of another master was given.21
Yet another difference is that Zaya Pandita started his list from the actual first
incarnation (and he clearly stated that this was the first embodiment: dang po) while
other authors included the name of Blo gros shin tu rnam dag, in Sanskrit Suviuddhamati,
the Buddhas contemporary, in their lists.

Conclusion

The biography of Zanabazar in the Horizontal Square Script is an important source


material since it is a rare example of the use of the Horizontal Square Script for Mongolian
language. The nature of the script provides interesting information about Mongolian
language and the manuscript may still serve as a basis for linguistic study.
Regarding the list of the previous incarnations of ndr Gegeen Zanabazar included in
this text, though it does not contain new information, it confirms the information known
from other sources. Since it is close to the reverential prayer gsol debs referring
to the Jonangpa master Trantha, Zanabazars immediate predecessor, we may suspect
that it was based on it or even written concurrently with it. Perhaps it served as a basis
for the list included in Zanabazars biography written in 1847 by Ngag dbang blo bzang
don grub to which it is very similar. Further study of previous incarnations of ndr
Gegen Zanabazar from a wider range of sources may bring more interesting information.

Bibliography

Bareja-Starzyska, A. 2009. Biographies of the First Jetsundampa ndr Gegeen Zanabazar Blo
bzang bstan poi rgyal mtshan (16351723) Brief Survey of Sources, Zentralasiatische Studien 38,
(2009), International Institute for Tibetan and Buddhist Studies, Halle, pp. 5363.

chos byung zla bai sgron me by Blo gros grags pa, TBRC-W4JW5404, p. 0115: Jam mgon bla ma rje kun dga
snying po.
21 In the commentary to the first Jetsundampas gsol debs composed by Zaya Pandita: rJe btsun blo bzang

bstan pai rgyal mtshan dpal bzang poi khrungs rabs gsol debs (ff. 1-50b3, reprinted in Byambaa R. 2004b) it
is explained (ff. 23b3-24a25b6) that he was a disciple of Panchen Jampa Lingpa (pan chen Byams pa gling pa).
About Jampa Lingpa see Bareja-Starzyska 2010b, pp. 250252.
NOTES ON THE PRE-EXISTENCES OF THE FIRST KHALKHA JETSUNDAMPA ZANABAZAR... 33

Bareja-Starzyska, A. 2010a. Mongolian Biography of Zanabazar (16351723), the First Mongolian


Jetsundampa A Preliminary Study of the Manuscript from the Kotwicz Collection. In: Altaica et
Tibetica. Anniversary Volume dedicated to Stanisaw Godziski on His Seventieth Birthday, Rocznik
Orientalistyczny, vol. 63 (1), pp. 3143.
Bareja-Starzyska, A. 2010b. The Mongolian Incarnation of Jo nang pa Trantha Kun dga snying
po : ndr Gegeen Zanabazar Blo bzang bstan pai rgyal mtshan (16351723). A case study of the Tibeto-
Mongolian Relationship. In: Roberto Vitali (ed.), Proceedings of the International Seminar on Tibetan and
Himalayan Studies, Held at the Library of Tibetan Works and Archives, September 2009 on the Occasion
of the Thank you India Year, The Tibet Journal, Autumn 2009Summer 2010, vol. XXXIVXXXV,
pp. 243261.
Bareja-Starzyska, A. 2011. Some Remarks on the Mongolian manuscript labeled History of Erdeni
Ju kept in the Kotwicz Collection at the Archives of Sciences in Cracow, Poland. In: Takashi Matsukawa;
Ayudai Ochir (eds), The International Conference on Erdene-Zuu: Past, Present and Future, Ulaanbaatar,
pp. 176183.
Bareja-Starzyska, A. (in print), History of Erdeni Ju or Biography of the First Jetsundampa
Zanabazar (j. 69a): information about a Mongolian manuscript kept in the Kotwicz Collection at the
Archive of Sciences in Cracow, Poland. In: Osamu Inoue; Jerzy Tulisow; Agata Bareja-Starzyska; Ewa
Dziurzynska (eds), In the Heart of Mongolia. 100th Anniversary of W. Kotwiczs Expedition to Mongolia
in 1912 (Selected Source Materials and Study) planned for publication in 2012.
Bawden, Ch. 1961. The Jebtsundamba Khutukhtus of Urga. Text, Translation and Notes, Asiatische
Forschungen, Wiesbaden.
Bira, Sh. 1995. ndr gegeeniin namtruud orshvoi, Ulaanbaatar.
Byambaa, R. 1997. Khevtee drvljin seg tnii dursgaluud, Ulaanbaatar.
Byambaa, R. 2004a. The Bibliographical Guide of Mongolian Writers in the Tibetan Language and
the Mongolian Translators, 1, Mongol Bilig Series, Ulaanbaatar.
Byambaa, R. 2004b (ed.). The Collected Works (gsung bum) of Blo bzang phrin las, re-edition,
Mongol Bilig Series, Ulaanbaatar.
Byambaa, R. 2005. Zanabazaryn drvljin seg, Mongol Bilig, Ulaanbaatar.
Byambaa, R. (ed.). 2006. The Collected Biographies of Jebzundampa Khutugtus of Khalkha, re-edition,
Mongol Bilig Series, Ulaanbaatar.
Lokesh, Chandra. 1982. Life and Works of Jebtsundampa I, ata-Piaka Series, vol. 294, New Delhi.
Dashbadrakh, D. 1995. ndr gegeenii namtar. In: Bira 1995, pp. 147.
Futaki, H. 2011. On the Mongolian Biography of the Jebtsundamba Khutugtus. In: Takashi
Matsukawa; Ayudai Ochir (eds), The International Conference on Erdene-Zuu: Past, Present and Future,
Ulaanbaatar, pp. 3549.
Khrelbaatur, L. 2009. ndr gegeen- namtar, br mongol-un arad-un keblel-n quriy-a, Kkeqota.
Lessing, F. 1982. Mongolian-English Dictionary, The Mongolia Society, Bloomington, Indiana.
Shagdarsren, Ts. 1981. Mongol seg zi, Ulaanbaatar.
Sheehy, M.R. 2009. A Lineage History of Vajrayoga and Tantric Zhentong from the Jonang Kalacakra
Practice Tradition. In: As Long as Space Endure. Essays on the Kalacakra Tantra in Honor of H.H. the
Dalai Lama, Edward A. Arnold Editor on behalf of Namgyal Monastery Institute of Buddhist Studies,
Snow Lion Publications, 2009, Ithaca, New York, pp. 219236.
Somlai, G. 1988. The Lineage of Trantha According to Klo-rdol-bla-ma. In: Helga, Uebach;
Jampa, L. Panglung (eds). Tibetan Studies. Proceedings of the 4th International Association for Tibetan
Studies. Schloss Hohenkamer Munich 1985, Mnchen, pp. 449451.
Stearns, C. 1999. The Buddha from Dolpo. A Study of the Life and Thought of the Tibetan Master
Dolpopa Sherab Gyaltsen, Suny Series in Buddhist Studies, State University of New York Press.
34 AGATA BAREJA-STARZYSKA, BYAMBAA RAGCHAA

1a 1b

1a 1a 1b 1b
The Biography of Jetsundampa Lama Qututu
(K-tg-t rje-btsin dam-pa
TheBiographyofJetsundampaQutugtu( bla-mai a-dig
TheBiographyofJetsundampaQutugtu( ro-i-ba)

)
)
NOTES ON THE PRE-EXISTENCES OF THE FIRST KHALKHA JETSUNDAMPA ZANABAZAR... 35

2a 2a 2b
2a 2b 2b




36 AGATA BAREJA-STARZYSKA, BYAMBAA RAGCHAA

3a 3b
3a 3b
3a 3b



NOTES ON THE PRE-EXISTENCES OF THE FIRST KHALKHA JETSUNDAMPA ZANABAZAR... 37

4a 4b
4a 4a 4b 4b



38 AGATA BAREJA-STARZYSKA, BYAMBAA RAGCHAA

5a 5b
5a 5a 5b5b



NOTES ON THE PRE-EXISTENCES OF THE FIRST KHALKHA JETSUNDAMPA ZANABAZAR... 39

6a 6a 6b
6a 6b 6b




40 AGATA BAREJA-STARZYSKA, BYAMBAA RAGCHAA

7a 7a



R O C Z N I K O R I E N T A L I S T Y C Z N Y, T. LXV, Z. 1, 2012, (s. 4150)

MICHAEL FARRIS

Lexical and Grammatical Properties of Personal Pronouns


in Vietnamese: Deicticity and Relationships

Abstract

Since all languages have ways of referring to the person speaking and the person
being addressed, some have been misled into thinking that the category of person works
in the same manner in all languages. Recognition that pronouns work differently in some
East Asian languages is usually limited to observations about the number of different
words used to refer to the speaker and addressee. This paper looks at the ways that
Vietnamese speakers use words to refer to themselves and their addressees. The system
that emerges is complex and dynamic. It makes use of both transreferential deictic words
(that refer to participants in the speech act) as well as non-deictic words that only refer
to individuals in terms of characteristics of the conversational participants such as age,
sex and social role and the relationship between interlocutors.

Keywords: linguistics, Vietnamese, personal pronouns, grammar, deicticity

Introduction

All human languages have ways of referring to the participants in conversational


interactions. The system used for this is traditionally referred to as person. The primary
indicators of person are personal pronouns, which in many languages are reinforced by
agreement affixes on verbs, nouns or other parts of speech. Since all languages have
ways of referring to the person speaking and the person being addressed some have
been misled into thinking that the category of person works in the same manner in all
languages. Close examination of a number of languages, however, indicates that this is
not the case.
42 MICHAEL FARRIS

The goal of this paper is to look at some of the ways that Vietnamese speakers use
words to refer to themselves and their addressees. The system that emerges from close
examination is complex, in that it is made up of a number of subsystems, and dynamic
in its application. In short, the system of self-reference and address in Vietnamese makes
use of both deictic words (that refer to conversational interlocutors) as well as non-deictic
words that refer to individuals not in terms of their participation in conversation but on
the basis of personal characteristics such as age, sex, social role and the relationship
between the interlocutors.
The traditional model of person posits three basic distinctions (Siewierska 2004: 1)
which are found in Figure 1.

first person : speaker


second person : addressee
third person : specific entity which is neither speaker nor addressee

Figure 1.

Person is traditionally part of a larger system of deictic reference whose point


of reference is linguistic interaction. Crucially, both first and second person identify
participants in a conversation. First person refers to the speaker (or sender) of a linguistic
message and second person refers to the intended recipient of the message. Typically
within a conversation both interlocutors use both and the identity of a given first or
second person referent depends upon who is speaking. This property (of referring equally
to any interlocutor depending on who is speaking at a given moment) is referred to here
as transreferentiality. Third person, on the other hand, is not transreferential in that it
refers to a specific entity (human or not) but only refers to the speech situation negatively
(as not being involved).

Person in East Asian languages

It has long been noticed, for example, that personal reference in a number of the
major languages of East Asia works very differently in a number of respects from that
found in languages from Europe i.e. belonging to Indo-European language family.
In describing Japanese, a language whose system of self-reference and address shares
some features with Vietnamese, Suzuki 1978 (p. 116) notes that in European languages,
the words for speaker and address are a stable and mostly closed set, and cognates
which date to prehistory appear across languages. Therefore the first person singular
pronouns in different Indo-European languages such as Polish ja, Portuguese eu, Danish
jeg, Lithuanian a, English I, Greek eg and others are all modern reflexes of a single
proto-Indo-European root, reconstructed by Sihler (1995: 369) as *eoH. This kind of
stability reaching across languages for thousands of years is unimaginable in Japanese
LEXICAL AND GRAMMATICAL PROPERTIES OF PERSONAL PRONOUNS IN VIETNAMESE... 43

where the set of words used to refer to the speaker is much larger and unstable in that
words enter and leave everyday usage much more rapidly than in European languages.
The situation in Vietnamese is similar. The set of words a speaker might use to
refer to themselves at present includes but is not limited to ti, t, tao, mnh, anh, em,
c, chu and others. Further, if a similar set for words used to refer to addressees is
assembled it might include my, bn, cu, anh, em, c, bc and others. Interestingly, it
will immediately be noticed that the two lists overlap, which is not likely or possible
in European languages.
In the case of Vietnamese, Nguyn 2004 describes Vietnamese pronouns as substitutes
and distinguishes between Personal substitutes (p. 124) which indicate the age and
social status of the interlocutor and Status substitutes (p. 126) which use kinship terms
fictively to indicate the relative age and status of referents.
Thompson 1991 distinguishes between personal pronouns which can be pluralized
with chng and absolute pronouns which cannot. He distinguishes four courtesy levels
for personal pronouns (p. 248). These are respectful, superior, familiar and abrupt.
He also claims that for the first two of these levels only first person forms exist and that
general nouns are used for second and third person and that these may also be used as
first person references to indicate humility (in the case of a younger person speaking
with an older person) or formality or abruptness if an older person is speaking with
a younger person (p. 299).

Grammatical Person and Lexical Person

Before further analyzing the words used by speakers of Vietnamese to refer to speakers
and addressees it is necessary to distinguish between two different aspects of person
reference. These are here called grammatical person and lexical person respectively.
The author first explored this issue in Farris 2003 where lexical person was referred
to as logical person. Grammatical person refers to a formal category as determined by
agreement markers or etymological factors. Lexical person on the other hand refers to
the pragmatic meaning. Most of the time these two categories are identical. But over
time, for a variety of reasons, mismatches may occur as semantic content changes while
the grammatical form remains unchanged.
One important example of how grammatical and lexical person may diverge, is found
in a number of European languages, including Polish, Spanish and Hungarian (with
belongs to the Fino-Hungarian language family). It involves the use of forms which are
grammatically third person for deferential or polite references to the addressee (lexical
second person). The system as used in Polish will be briefly examined here. The lexical
second person paradigm (most common forms) for the verb mwi (to speak) in Polish
is found in Figure 2.
The familiar second person pronouns ty and wy are not required and are usually
omitted. The formal second person pronouns (Pan et al.) are required and may not be
44 MICHAEL FARRIS

omitted. The forms given here are only the most common, least marked forms used with
Pan, et al. Rarely second person verb forms can occur with the singular forms but these
are highly marked and will not be discussed further here.

second person singular second person plural


(ty) mwisz (wy) mwicie
Pan mwi (masculine) Panowie mwi (masculine)
Pani mwi (feminine) Panie mwi (feminine)
Pastwo mwi (mixed)

Figure 2.

It should also be noted that Pan/Pani retain their meanings as common nouns.
Depending on the context Pan may be translated as gentleman, man, or the title Mr. in
English while Pani might be translated as lady, woman or the titles Mrs., Miss and Ms.,
again depending on context. A common noun used pronominally this way which does not
lose its original meaning may be known as a pronominalized noun. In Polish, the pattern
of a pronominalized noun and a third person verb form as a polite form of address is
not limited to these examples. It is fairly productive, in that in specific situations other
nouns can act in similar ways, the most common of these include ksidz (priest), siostra
(nurse, lit. sister) and sometimes within the family, kinship terms.
It should also be noted that the use of Pan et al. as a pronominalized noun differs in
some ways from both unambiguously second person pronouns and from the homophonous
common nouns. As noted previously, Polish is a null-subject or PRO-drop language in
which first and second person subject pronouns may be omitted. But Pan and similar
forms may not be omitted as illustrated in Figure 3.

Czy ty znasz j? Do you (Ty) know her?


Czy znasz j? Do you (Ty) know her?
Czy Pan zna j? Do you (Pan) know her?
* Czy zna j? Do you (Pan) know her?

Figure 3.

Pan used as a pronominal noun also is different from pan as a common noun in
that the latter may be subtituted by third person pronouns anaphorically while the former
cannot. This is illustrated in Figure 4.

Widz pana, ale nie znam go. I see the man but I dont know him.
Widz Pana, ale nie znam Pana. I see you but I dont know you.

Figure 4.
LEXICAL AND GRAMMATICAL PROPERTIES OF PERSONAL PRONOUNS IN VIETNAMESE... 45

Pronoun Systems in Vietnamese

Mismatches between grammatical and lexical person are easy to spot in languages with
person agreement forms. They can be harder to spot in languages with no agreement forms.
Vietnamese lacks agreement forms but has an extensive and rich system of pronominal
reference. An examination of their usage reveals three categories of pronoun forms, two
rely to a greater or lesser degree on the traditional three person system, while the third
is neutral as to person.

Transreferential pronouns in Vietnamese

Vietnamese does appear to have a category of transreferential personal pronouns.


These have stable meanings which refer to one of the three person categories. Although
some of them have origins as common nouns, those meanings are now distinct from
their use as pronouns. Some of the more common are found in Figure 5.
Of these, the most commonly used is ti, which is always acceptable as a first
person singular reference. The second person forms however are problematic. They are
normally abrupt or arrogant. While some speakers accept my as an intimate second
person pronoun, others do not, finding it rude. Also the third person forms n, and hn
are not always acceptable for human referents and often indicate disdain.

first person: ti, ta, tao, t


second person: my, mi, bay
third person: n, hn, h
Figure 5.

Pronominalized nouns

This is a large, productive category in Vietnamese and there is no way in this limited
space to indicate the full richness of the system. The crucial point to remember in this
system is that the choice of term does not depend on conversational roles. That is, the
pronominalized noun does not refer to a persons momentary role in the conversation,
but acts as a label for a specific person, irregardless of whether they are speaking, being
addressed or or being referred to. Within the traditional framework of person these
function not as first or second person but as third person.
The most important pronominalized nouns are kinship terms. Within the family, these
are commonly used instead of pronouns, but they are also widely used with non-kin. In
this case, the choice of term is determined by relative age of the interlocutors as well as
the kind of relationship between them. Not all forms are equally likely to be used. The
most common kin terms which are extended to non-kin are found in Figure 6. Here the
Vietnamese form is followed by the literal kin definition and general indication of what
kind of person they could be used of.
46 MICHAEL FARRIS

ng (grandfather) adult male of greater age or status than interlocutor


b (grandmother) adult female of greater age or status than interlocutor
anh (older brother) adult male of similar age and status as interlocutor
ch (older sister) adult female of similar age and status as interlocutor
c (aunt) young adult woman
cu (uncle) young adult man
em (younger sibling) child or adult younger than interlocutor
chu (grandchild, neice, nephew) adult younger than interlocutor

Figure 6.

Other pronominalized nouns include but are not limited to occupational titles and
personal names.

Pronominalized and other nouns used transreferentially

There is also a small class of pronominalized nouns that can be used as true
pronouns, identifying conversational roles, instead of a specific person independent of
their conversational role. The most common are shown in Figure 7.

mnh (body, self) first person, second person


bn (friend) second person
cu (uncle) second person

Figure 7.

When used as a personal pronoun, mnh is normally used for first person, but also
finds limited use, as an intimate you between spouses. As a pronominalized noun it is
a reflexive which can refer to all three persons.
In modern Vietnamese, bn is the closest equivalent to a neutral second person
pronoun (used, for example, in impersonal written documents). It may also be used as
a pronominalized noun. Modern speech finds cu used as a familiar second person sex-
neutral pronoun, although as a pronominalized noun it is specifically male.

Use of different pronoun systems in Vietnamese

Not only are there different classes of pronomial words in Vietnamese but they
interact in multiple ways. There are three patterns of pronoun usage in Vietnamese. The
first of these uses transreferential pronouns exclusively so that the words used refer to
conversational roles. This is probably the rarest pattern, perhaps partly because of its
inherent limitations in describing the relationships between interlocutors and in expressing
honorific. A short example can be seen in Figure 8. The words used as personal pronouns
LEXICAL AND GRAMMATICAL PROPERTIES OF PERSONAL PRONOUNS IN VIETNAMESE... 47

are in bold. Each line of the original dialogue is followed by a word by word literal
translation and finally by a free translation. The conversation is taken from Halik, Hong
1997 (p. 8).

A: Cho bn, i u vi th ?
greeting friend(you), go where hurry so
Hello, where are you in such a hurry to?
B: A, xin cho, mnh i hc. Bn lm g vy?
Ah, ask greeting, self(I) go learn friend(you) do what then
Ah, hello. Im going to school. What are you doing?
A: Mnh ch m. M mnh mua bn trong ch.
self(I) wait mother. mother self(I) buy sell in market
Im waiting on my mother. Shes gone shopping in the market.

Figure 8.

In this short conversation, both interlocutors, young males of close age and similar
status, use mnh as a transferential first person pronoun and bn as a transreferential
second person pronoun.
Unlike the first pattern which is not especially frequent, the second pattern is very
common and uses pronominalized nouns exclusively. A short fragment can be seen in
Figure 9. This is an excerpt from the 1934 novel Na chng xun (Halfway through
spring) by Khi-Hng, cited in Nguyn 1997 (p. 260.) The glosses and translations are
from Nguyn.

B n: Chc con c i u g phi n mun m con giu m.


surely child(you) have matter whatever worry that child(you) hide mother(I)
You must have something troubling youre trying to hide from me.
Lc: Vng, c th . B m m, con kh lm.
yes emphatic so. respectfully report mother(you) child(I) unhappy very
Yes, I have. Mother, Im very unhappy.
B n: Chuyn g th con?
story what so child(you)
What is the matter son?
Lc: M c tha ti cho con th con mi dm tha.
mother(you) emphatic forgive sin to child(I) then child(I) only-then dare report
Mother, I dare tell you only if you forgive me.

Figure 9.

Here, both the mother and her adult son refer to the mother as m mother and both
the adult son and his mother refer to the adult son as con child. In other words neither
48 MICHAEL FARRIS

is using transreferential pronouns but rather non-transreferential pronouns. Taken out of


context, the conversation could just as easily be about other (third) persons as about the
interlocutors themselves.
Here two points should be made. Motherese (language interactions between mothers
and very young children) in a number of languages (including Polish and English) at
times uses this same kind of third person reference instead of normal first and second
person. The difference between those languages and Vietnamese is that the same pattern
can continue even when the child is an adult. The conversation quoted in Figure 9 is
ultimately about the sons desire to marry a young woman. It would be extremely strange
for a young man and his mother to continue such a form of address into adulthood in
English while it is not strange in Vietnamese. It should also be noted that the English
translation does include some vocative expressions, but such vocative expressions cannot
be used as arguments of verbs in English while they can in Vietnamese.
The third pattern depends on mixed usage of words belong to different pronomial
classes. Most commonly in this type of usage, the speaker uses a true pronoun for first
person and pronominalized nouns for second and third person references. A short example
can be seen in Figure 10.

Tally: Anh tho ng ph c cha?


older-brother(you) already skilled road at city old yet
Do you know your way around the old town yet?
Hiro: Chu! khng hi u sao ti khng th nh n i ng khu ph c.
suffer! not understand why I not can remember road at area city old
No way! I dont understand why I cant remember the streets in the old town.
Ti b lc lun, khng lm sao nh hng n i.
I suffer lost often, not do why plan direction plans.
I often get lost. I cant find my orientation there.
Ch chc tho ri.
older-sister(you) certain skill already
Surely you know your way around already.
Tally: Ti cng vy thi.
I also so just.
Im just the same way.

Figure 10.

Functionally, this might be referred to as a two person system as the only consistent
distinction made is between first and non-first person. It is difficult to consistently
distinguish between second and third person. While some claim the difference between
pronominalized nouns in the second and third person can be made with a demonstrative
pronoun, this is not consistently used and the categories tend to merge. Thus, while first
person remains a separate category from other person distinctions in this usage, second
LEXICAL AND GRAMMATICAL PROPERTIES OF PERSONAL PRONOUNS IN VIETNAMESE... 49

and third person appear to merge as a non-first person category since shorn of context
it is impossible to tell if any particular instance to anh or ch is referring to one of the
interlocutors or some other person not involved in the conversation.
It should also be noted that these different arrangements of person categories are
not segregated with each conversation using one or the other. Quite the opposite, the
system is fluid and in a single conversation, usage may drift from one pattern to the
other and back again. This can be seen in Figure 11. This conversation is taken from
Halik, Hong 1997 (p. 41).

Tun: Lan i u y?
Lan(you) go where emphatic
Lan, where are you going?
Lan: Cho Tun, mnh ra ph c cht vic.
greeting Tuan(you), self(I) out street have little work
Hello Tun, Im going out to take care of something.
Tun: Mnh cng ra ph, th chng ta cng i c c khng?
self(I) also out street, so we together go have can not
Im also going out, so maybe we can go together?
Lan: Ti sao li khng? Tun nh mua g? cho ai?
why again not? Tuan(you) decide buy what? for who?
Why not? What do you want to buy? Whos it for?
Tun: Cho ai th mnh cha th ni c. (....) Lan gip Tun nh. (...)
for who then self(I) not-yet can say can. Lan(you) help Tun(I) right(...)
I cant say who its for yet. Youll help me okay?
Lan: Mnh s c gng gip Tun.
self(I) will do-best help Tun(you)
Ill do my best to help you.

Figure 11.

In this conversation both interlocutors use the transreferential pronoun mnh to refer
to themselves; they also use their personal names not only vocatively as terms of direct
address but as verbal arguments to refer to each other and themselves. There is also
a plural form here, the first person inclusive (including addresee) form chng ta. Plural
forms in Vietnamese carry their own complications however and for the sake of simplicity
are not discussed here.

Conclusion

Often in the description of non-Western languages, the uncritical usage of grammatical


categories or terminology may obscure rather than enlighten. Much more research is
50 MICHAEL FARRIS

needed to adequately describe usage and to discover the factors that trigger one or the
other kind of usage.The published grammars of Vietnamese do not entirely adequately
describe either the categories of Vietnamese pronouns nor their usage. It is probable that
the concepts of grammatical person and lexical person as well as that of transreferentiality
may shed new light on the structure of Vietnamese as well as those of other languages
such as Japanese and Thai with ambiguous or still poorly understood pronoun systems.

References

Farris, M. 2003. Notes on the categories of logical person and grammatical person. 4th International
Conference of PhD Students, University of Miskolc, 1117 August 2003.
Halik, T.; Hong, H.T. 1997. Podrcznik jzyka wietnamskiego: Cze II. [Vietnamese Handbook,
Part 2], Dialog: Warsaw.
Nguyn, D.H. 1997. Vietnamese: Ti ng Vit Khng Son Ph n, Amsterdam: John Benjamins.
Nguyn, T.N. 1998. Ti ng vit nng cao (Cho ngi ngoi Quy n 1) (Intermediate Vietnamese
(For non-native speakers). Hanoi: Nh xut bn gio dc.
Siewierska, A. 1991. Person. Cambridge: Cambridge University Press.
Sihler, A.L. 1995. New Comparative Grammar of Greek and Latin. Oxford: Oxford University Press.
Suzuki, T. 1978. Words in Context. Tokyo: Kodansha (translated by A. Miura).
Thompson, L.C. 1991. A Vietnamese Reference Grammar. Honolulu: University of Hawaii Press.
R O C Z N I K O R I E N T A L I S T Y C Z N Y, T. LXV, Z. 1, 2012, (s. 5162)

ELI FRANCO

How to Distinguish between Non-existing Entities?


Dharmakrti and Prajkaragupta
on Universals as Objects of Knowledge1

Abstract

The paper concerns the very beginning of the chapter on perception in Dharmakrtis
Pramavrttika. It focuses on an apparent contradiction in Digngas and Dhamrakrtis
epistemology, namely, that universals do not exist and yet aare considered to be objects
of valid inferences. Thus, universals have to be distinguished from other non-existing
entities, such as objects of illusions. The paper elucidates Dharmakrtis theory with the
help of his great commentator Prajkaragupta.

Keywords: Buddhism, logic, philosophy, Dharmakrti, Prajkaragupta, Buddhist


epistemology

This paper concerns the very beginning of the chapter on perception in Dharmakrtis
Pramavrttika, which was conceived as a commentary on the second verse of the
1 The following paper is one of the results of research project FR 2531/2 funded by the German Research

Foundation (DFG), which has been conducted at the Institute of Indology and Central Asian Studies, University of
Leipzig since February 2008: A Study on the Buddhist Theory of Perception Based on Dharmakrtis Pramavrttika,
Chapter III: A Critical Edition, Translation and Study of Pramavrttika III 1122. I would like to take this
opportunity to thank the DFG for its generous grant. The project is part of an international effort to translate the
Pramavrttika, Dharmakrtis magnum opus, into English; more specifically, three teams are working on the chapter
dealing with perception. The Leipzig team aims at a translation of the first 122 verses, John Taber together with
graduate students of the Philosophy Department, University of New Mexico, Albuquerque, will translate verses
123240, and Birgit Kellner, Heidelberg University, will translate selected verses of the remaining part.
I would like to thank very warmly Miyako Notake, my collaborator in the above-mentioned project. Very special
thanks are also due to my wife Karin Preisendanz for her most helpful comments.
52 ELI FRANCO

Pramasamuccaya. In this verse, Dignga states that there are only two means of
knowledge or instruments of valid cognition (pramas), namely perception and inference,
because the objects of these means of knowledge have two characteristics: Perception
has the particular as its object, and inference the universal.2 However, with the same
breath Dignga also states that the universal is only a conceptual construction and that
as such it does not exist in reality. Conceptual construction is the association of a name,
genus, and so forth [with the thing perceived which results in a verbal designation of the
thing].3 Therefore, the question immediately arises: How could something non-existing
be an object of a means of knowledge?
Dignga neither raises nor answers this question. This is indeed one of the greatest
mysteries of the Pramasamuccaya, that is, not only what Digngas answer to this
question would have been, but also why he saw no need even to raise the question. If
this is such an obvious problem, as it seems to us, how could it be that Dignga had
nothing to say about it?
I am not going to propose an answer to this question. As far as I can see, it cannot
be answered with our present knowledge of Digngas work. What I propose to do
instead is to try to explain, very briefly, Dharmakrtis answer to this question. Further,
as indicated in the title of the paper, I will dwell, again very briefly, on a specific aspect
of this answer, namely, the issue of what distinguishes non-existing universals from other
non-existing entities, such as the object of illusion, for instance, that there are two moons
in the sky or floaters in the clear air.
Dharmakrti begins his comments on Digngas verse by proposing four criteria for
the distinction between particulars and universals. The first and most important of them is
that particulars are capable of efficient action, and universals are not. And already in the
first verse of the chapter Dharmakrti states that floaters are not objects, that is, not real
objects because they are not determined as real objects (artha),4 presumably by people in
the context of their everyday practice.5 This statement, albeit not an argument, is already
a clarification and an indication towards where things are going: illusory objects are not
real objects, which implies that universals are.
A word about the word artha. The Sanskrit word artha literally means aim, goal,
purpose, telos, and a most commonly derived meaning is referent. For instance,

2 Cf. PS I, p. 1.15-20 (I quote the verse together with the Vtti): pratyakam anumna ca prame. te dve
eva, yasml lakaadvaya prameya. na hi svasmnyalakabhym anyat prameyam asti. svalakaaviaya
ca pratyaka smnyalakaaviayam anumnam iti pratipdayiyma.
3 PS I, p. 2.8: nmajtydiyojan. The translation quotes Hattori 1968: 25.
4 PV 3.1:

prama dvividha meyadvaividhyc chaktyaaktita |


arthakriyy kedir nrtho narthdhimokata ||
The PV verses quoted here follow our forthcoming edition. For the reading prama dvividha meyadvaividhyt
we follow Inami et al. 2002. See also Dreyfus 1997: 297298 and n. 53, Dunne 2004: 391, Hugon 2011: 369.
5 This is how Manorathanandin and Jinendrabuddhi interpret the statement. It could also be understood, of

course, as determined by means of knowledge or philosophical investigation. However, the exact implication is of
no consequence for us here.
HOW TO DISTINGUISH BETWEEN NON-EXISTING ENTITIES? 53

in a linguistic context, it is often used to refer to the referent of a word, in which case
it is habitually translated as meaning. In an epistemological context, the word is used
to refer to the referent of a cognition or an awareness (jna, vijna, etc.), and unless
one presupposes an idealistic or illusionistic metaphysics, it refers to the cognitions
counterpart in reality, to a real thing that corresponds to the cognitions content or to
the image that appears in the cognition. The common German translation of artha as
Gegenstand conveys this meaning of artha well: something that stands against or towards
something else.
To come back to our question, how can the universal, which does not exist, be the
referent, the objective counterpart, of a valid cognition? And inversely, if the universal
is a real object, how could it not be capable of efficient action? After all, the criterion
of true existence (paramrthasattva) is the capacity to act efficiently.6
Dharmakrti is a very laconic, at times obscure writer, and to understand his arguments
we often have to rely on his commentators. The most important of his commentators is
Prajkaragupta who lived in the second half of the 8th century (750810 according to Ono)
and was one of the most brilliant philosophers in the Buddhist epistemological tradition.
In the following, I will rely on his great (both in size and importance) commentary the
Pramavrttiklakra.
In his comment on the above-mentioned verse (PV 3.1), Prajkaragupta explains the
negation in the statement about the non-capacity of the universal to produce efficient action
as an implicative negation, referring thereby to the well known division of negations as
implicative and non-implicative (paryudsapratiedha and prasajyapratiedha). A non-
implicative negation is merely a negation, whereas an implicative negation actually implies
a positive statement by means of exclusion. For instance, when we say that someone is
not at home, we usually imply that she is outside; when we say that someone does not
eat during the day, we imply that she eats at night. So what does the implicative negation
mean in our case? It means that the universal does not produce efficient action directly,
but that it produces it indirectly. An illusory object like a floater produces no efficient
action, neither directly nor indirectly, and is therefore not a real object neither directly
nor indirectly (PVA 169.25-26):

6 PV 3.3:
arthakriysamartha yat tad atra paramrthasat |
anyat savtisat prokte te svasmnyalakae ||
This verse has been often translated and discussed many times, for instance in Dunne 2004: 292. Yoshimizu
(2007: 1050) deals with the Sautrntika background of this distinction, and Moriyama (1991: 202, 206) points out
Jnagarbhas and Haribhadras response and criticism of it. (Dharmakrtis tenet is, of course, an anathema to the
Madhyamakas who consider causally arisen entities unreal.) For Go ram pas comment on PV 3.2a see Yoshimizu
2007: 10671068 and for Gyel tsaps Dreyfus 1997: 117. See also Ono 1999: 313 on the relation of this verse to
the sattvnumna; on its relation to Abhidharma see Katsura 2011: 275276.
The first half of the verse is repeated in PVin II 55ab; cf. also PV I 166ab. See also Lindner 1984: 152153.
54 ELI FRANCO

praparyea aktir evakti paryudsavtty, na aktyabhva. kedes tu


taimirikopalabdhasya bddyupalabdhasya ca nrthatattva skt praparyea v
aktatvenpratte.

Non-capacity is nothing but capacity in an indirect way by the [negations] function


of exclusion, [it is] not absence of capacity [as such]. The hairs, etc., however, which
are perceived by a person who suffers from timira, and something that is perceived by
a verbal [cognition] and so on, do not possess reality as objects because they are not
perceived as capable, neither directly nor indirectly.
However, with this we do not solve the problem, but only take it elsewhere. If the
universal does not exist, how can it produce an efficient action, even indirectly? In PV
3.7, Dharmakrti returns to the distinction between universals and illusory objects. Floaters
and so forth are not universals because they are not determined as real objects (artha).7
And Prajkaragupta explains that the universal is certainly (eva) a real object (artha)
inasmuch as a real object which is a particular by nature is in this case being perceived
in the form of a universal, that is, in a general form.8
The somewhat puzzled opponent then asks: If the real object is a particular, then
since in this case it causes an efficient action how is it a universal? And if it does not
cause an efficient action, how could it be a real object (artha)? For something is a real
object (artha) by the fact that it causes an efficient action. And one and the same thing
cannot cause and not cause an efficient action.9
Prajkaragupta answers that this is not impossible. The thing under discussion is
auniversal as well as a real object inasmuch as it is incapable of efficient action directly,
but is capable indirectly; thus is capable and incapable at the same time. The universal
is a real object because it is determined as a real object. From the point of view of
absolute reality, the universal is not a real object; the status of a real object is ascribed
to it metaphorically because of a mental transposition of the characteristics of a particular
to it, because it indirectly has the effect of a real object.10
7 PV 3.7:
kedayo na smnyam anarthbhiniveata |
jeyatvena grahd doo nbhveu prasajyate ||
This verse is translated in Dunne 2004: 394, and briefly discussed in Meindersma 1991: 171 and Hugon 2011:373.
For the second half of the verse see also below.
8 PVA 189.23: artha eva hi smnyarpatay svalakatm pratyamna smnyam. Here, Prajkaragupta

alludes, of course, to PV 3.54cd: tasya svapararpbhy gater meyadvaya matam ||


9 PVA 189.23f.: yady asau svalakaam arthakriykritvt katha smnyam anarthakriykritve ca katham

artha. arthakriykritvenrthatvt. na ca dvayam ekatrrthakriykritvam itarac ca.


10 PVA 189.25-27: naitad durghaam.

skd arthakriyakta pramparyea aktita |


smnya ca tad artha ca akta caktam eva ca ||173||
smnyam arthdhyavasyenrtha. paramrthato nrtha eva. upacaritam arthatvam asya paramparayrthakryatvt
(anyrthatvt).*
*Sktyyanas note 3 thereon: T na bhntare. I am not quite sure what he means; perhaps that anyrthatvt
has no equivalent in the Tibetan.
HOW TO DISTINGUISH BETWEEN NON-EXISTING ENTITIES? 55

Another problem arises in this connection: If universals are real objects, what about
referents of negative assertions? An example for a negative assertion would be: Here
on the table there is no bottle. Now the referent of this statement, namely, the absence
of a bottle, is considered by Dharmakrti to be a universal, but surely, an absence cannot
be a real object (artha). On this point there seems to be a difference in terminology
between Dharmakrtis commentators. For Prajkaragupta absences are arthas,11 whereas
for Manorathanandin they are not, yet they are conventionally determined to be objects
of awareness, which in this context probably means objects of valid cognition (jeya).12
The difference in terminology among the commentators may be explained as being due
to a shift in Dharmakrtis terminology. Hugon has suggested recently (2011: 375) that
Dharmakrti considered the universal to be artha in the Pramavrttika, but abandoned
this position in the Pramavinicaya.
Thus, we are dealing with three kinds of phenomena: illusory objects, universals such
as being a bottle, and special universals such as absences or non-existences of things
which are the results and referents of inferences that something does not exist. The latter
are also distinct from illusory objects because they are correctly cognized.13 Therefore, at
least according to Prajkaragupta, absences too are real objects/things (artha), and that
for two reasons: First, inasmuch as they are in the final analysis other things, which have
efficient action; for instance, the object of the apprehension of the absence of a bottle on
the table is in the final analysis nothing but the table. Second, and here Prajkaragupta
relies on the literal meaning of the word artha, absences serve a purpose, for instance,
dispel fear or doubt about something by proving that it does not exist.
PVA 190.10-14: abhvo pi hi kenacid krea jeya eva. anyabhva evbhva
iti siddhnta, sa crthakriykryea. atha vsv api pramena paricchidyata eva.
prayojanavattvc csv artha. abhvenpi hi prayojanam eva niakagamanrthatvt.
tatas tatrpy arthdhimoka iti, artha evbhva svena rpea, na puna kedir,
bhrntavijeyatay paricchedt.
Absence too is indeed nothing but an object of [valid] cognition with a certain form.
The established tenet is that absence/non-being is nothing but the presence of something
else/another being. And it is [correctly known (jeya)] through its result, which is efficient
action. Or rather, this [absence] too is indeed determined by a means of knowledge.
And because it fulfils a purpose, this [absence] is a real object. For a purpose is also
[fulfilled] by an absence because its purpose is fearless going [and so forth]. Therefore,
it too is determined as a real object. Consequently, absence is indeed a real object sui
generis (svena rpea), but floaters and so forth are not [real objects] because they are
determined as erroneous objects of cognition.
Dharmakrtis universal is of course not the eternal entity maintained by the realists.
It is not an entity distinct (pthak) from the particular; it also does have the particular
11 Cf. PVA 190.10-14 quoted below.
12 PVV 115.13-14: yady apy abhve rtharpaty (read tay?) adhyavasyo nsti jeyatay tv asti lokasya.
13 PV 3.7cd (quoted above): jeyatvena grahd doo nbhveu prasajyate ||. jeyatva clearly has to be understood

here as prameyatva.
56 ELI FRANCO

as its support (raya, cf. PV 3.25cd). It has no svabhva (cf. PV 3.27b, 28) because it
cannot be established as different from the particulars.14 According to Prajkaragupta, if
the universal would be different from the particular, it is not connected to the particular
and thus not perceived, and therefore lacks own nature.15 For him, this also distinguishes
the universal from non-existing things: Non-existing things do not have dependence
relationship with individual things, but universals do. Prajkaragupta explains the
difference between universals and non-existent things in their relationship to individual
things as follows.
PVA 200.5-7: vastusmnyavdino pi gotva karkdikasya kasmn na bhavati.
apekbhvt. na hi karkdivyakty tad vyajyate. eva tarhi baleydiv apekbhvd
eva kharavidikasya na tatsmnyarpat. gotvasya tu tadrayaena bhvt tatra
ca pravartand bhavaty apeklakaa sambandha iti na doa.
Why is it also for the [opponent] who holds that universals are real that [individual]
white horses and the like do not possess cow-ness?
[Opponent:] Because there is no relationship of dependence [between them]; for it
(i.e., cow-ness) is not manifested by individuals such as white horses.
[Proponent:] If so, because [non-existent things] such as the horn of a donkey do
not at all depend on [individuals] such as spotted cow (baleya), they (i.e., non-existent
things) do not have the form of their universals. Cow-ness, on the other hand, has
a relationship characterized by dependence [on such individuals as the spotted cows]
because it exists as being supported by them, and also because one is prompted to act
towards them. Therefore, there is no fault [in our theory].
To go back to Dharmakrti, the universal only appears to have an own nature due to
an error that is caused by repeated experience of individuals and that has no beginning:

tasy rpvabhso yas tattvenrthasya v graha |


bhrnti sndiklnadaranbhysanirmit || PV 3.29

The appearance of a form/own nature in respect to a universal, or the apprehension


of an object as that (i.e., as having the nature of a universal),16 is an error that is caused
by repeated experience that has no beginning.17
Although the cognition of a universal is valid inasmuch as it leads to an efficient action,
it is also false: In the cognition of a universal one perceives things that are distinct from
14 There is an extensive literature on Dharmakrtis theory of the universal which relies mainly on the more detailed

exposition in the first chapter of the PV. See, for instance, Dreyfus 1997, chapter 7 and Dunne 2004: 113144.
15 According to the PVV on 27ab, the universal cannot be expressed as identical with or different from the

individual things. In fact, it cannot even be expressed as identical with or different from other universals (PV3.33cd).
The impossibility of being expressed as identical with or different from something real (tattvnyatva) is occasionally
used by Dharmakrti as a criterion for unreality.
16 This interpretation follows Manorathanandin (PVV 133.16: tattvena jtisvabhvena). Prajkaraguptas

interpretation of tattvena is more complex and possibly offers two readings of the text: tattvena and tattve na.
For lack of space, it cannot be discussed here.
17 See also Dunne 2004: 141.
HOW TO DISTINGUISH BETWEEN NON-EXISTING ENTITIES? 57

each other in a general form, and this form is false, for one perceives an external thing
in a form that it does not have. The general form of the universal spils (abhiplava), so
to speak, onto the individuals.18 Consequently, the universal has a somewhat ambiguous
status: It is said to be a real object (artha), but is not a real thing (vastu)19 (PV 3.33a)
and has no own nature (nisvabhva, arpa) (PV 3.32c); its cognition is false (bhrnta)
and yet it is valid (pram).
Dharmakrti takes up this topic again in 3.50 and the following verses: The universal
as such, i.e., as distinct from the individuals, is never perceived, and is thus incapable of
producing even its own cognition. For this reason it is said to lack own nature because
this incapacity is the characteristic sign of unreal things (avastu).20
Although unreal, the universal is divided into three kinds, depending on whether it has
an existing thing as its support, non-existing thing, or both.21 Here too the commentators
differ in their explanations. According to Manorathanandin, who follows Devendrabuddhi,
the division refers to the extension of the universals. For instance, the universal being
produced (ktakatva), which is an inferential sign, has existing things, such as the visible
(rpa), as its support; non perception (anupalabdhi) of something that is in principle
visible and non having been produced (anutpattimattva), etc., extends to non-existing
things; these too are inferential signs. Non-perception in general (anupalabdhimtra)
extends to both non-existing perceptible objects and existing imperceptible objects;
similarly the universal being knowable (jeyatva) extends to both since both existing
and non-existing things can be known. However, Manorathanandin does not mention that
these universals serve as universal signs,22 and I am not aware of inferences in which
they do. Prajkaragupta suggests two alternative explanations of the above division of
universals. The text of the second is somewhat corrupt, but the explanation seems to
be the same as the one proposed by Devendrabuddhi and Manorathanandin. In the first
interpretation, however, the three kinds of universals correspond to the three types of
valid reasons (svabhvahetu, anupalabdhihetu and kryahetu). A conceptual cognition
that has something existing as its support is one that has arisen from a reason that is its
own nature (svabhvahetu). A conceptual construction that has non-existing thing as its
support has arisen from a reason consisting in non-perception (anupalabdhihetu), and

18 Cf. PV 3.31:
smnyabuddhau smnyenrpym apkat |
arthabhrntir apyeta smnya spy abhiplavt ||
19 The terminology is indeed confusing; in Sanskrit philosophical texts artha (in the sense of object) is usually

considered to be vastu.
20 PV 3.50d: tad (i.e., jnamtrrthakaraam) dhy avastuu lakaam. See further PVSV 84.10: tasmt sarva

smnyam anarthakriyyogyatvd avastu. See also Dunne 2004: 86.


21 PV 3.51cd: smnya trividha tac ca bhvbhvobhayrayt ||. On Go ram pas response to this verse,

see Dreyfus 1997: 157. See also the parallel passages in PVSV discussed in Steinkellner 1971: 195197 and Hugon
2011: 373374.
22 PVV 131.2-5: kicid bhvopdna smnya yath rpdn bhvn ritya ktakdiabdavcya ligam.

abhvopdna yathopalabdhilakaaprptasysato nupalabdhir anutpattimattvdi ca. tad dhy abhvraya


smnya ligam. ubhayrayam anupalabdhimtra jeyatvdi ca bhvbhvasdhraatvt.
58 ELI FRANCO

the one that is supported by both existing and non-existing things from a reason that is
an effect (kryahetu).23 Prajkaraguptas two interpretations are not mutually exclusive,
but their difference is considerable: according to the first interpretation, the scope of
the statement about the triple division is limited to universals insofar as they occur in
inferences, whereas according to the second interpretation, the division of universals is
a general one (which also includes, of course, universals used as reasons of inference).24
The fact that the universal has an existing thing as its support or basis does not mean
that its cognition has an existing thing as its object (PV 3.52). If this were the case, the
sense faculties would be superfluous. If the universal, which is the referent of a word,
were real, the cognition of a word such as peach would have the same effect as seeing
or tasting a peach. Thus, one would not need the sense faculties in order to experience
aparticular peach. Furthermore, it is clear that the cognition of a universal does not refer
to existing things because one can speak about past and future things, e.g., when one
says, there was a pot, I will make a pot, and so on (PV 3.52d).25
In the final analysis, the object to be cognized is only the particular (PV 3.53d: meya
tv eka svalakaam). If one says that there are two kinds of valid cognition, this is
only because the particular is cognized in two different forms, with its own form and
with a general form (PV 3.54cd: tasya svapararpbhy gater meyadvaya matam).26
Dharmakrti has made up a nice example to illustrate the difference between universals
and illusory objects (or in fact among different universals). Even though all universals
are unreal, and even though all cognitions of universals are false, certain cognitions of
universals lead to efficient action and other do not. Imagine two persons, the one mistaking
the light of a lamp falling through a keyhole27 for a jewel, the other mistaking the glow
of the jewel for the jewel itself. They both run to fetch the desired object, but only one
of them will get the jewel.28

23 PVA 212.8-9: tath hi bhvopdno vikalpa svabhvahetor uday. abhvopdno nupalabdhe. ubhayopdno

vikalpa kryahetor iti.


24 The threefold division of universals appears also in PV 1.205-206:

andivsanodbhtavikalpaparinihita |
abdrthas trividho dharmo bhvbhvobhayraya || 205||
tasmin bhvnupdne sdhye synupalambhanam |
tath hetur na tasyaivbhva abdaprayogata || 206||
See also PVin III 53. Verse 205 is translated in Steinkellner 1971: 189, n. 43, see also Tillemans 1999: 176. In
neither of these contexts I can discern a threefold division of universals according to the three kinds of reason.
25 One has to note, however, that universals and particulars are not co-extensive with things that are the objects

of a word or not respectively. As Prajkaragupta clearly puts it, the universal is an object only of a word, but it
is not the case that whatever is the object of a word is a universal. And what is not the object of a word is not
thereby a particular. Rather, the particular is not, i.e., never, the object of a word. See PVA 170.20-21: abdasyaiva
viaya smnyam. na tu abdasya yo viayas tat smnyam eva. aviayo pi na svalakaam evpi tv aviaya eva
svalakaam ity artho vivakita. I do not know to which further objects of words Prajkaragupta refers here.
26 See Katsura 1991: 136 and 2011: 276, Hugon 2011: 370.
27 Cf. NBh 197.22-23: kucikvivaranisty prabhym
28 See PV 3.57. This verse has been translated many times, most recently in McCrea 2011: 321. It is repeated

by Dharmakrti in PVin 2.5.


HOW TO DISTINGUISH BETWEEN NON-EXISTING ENTITIES? 59

The example is illuminating, but not unproblematic from a systemic point of view.
kyabuddhi raises an objection that the cognition in the second case, the one of the
glow of the jewel that leads to the obtainment of the jewel, would amount to a third type
of means of knowledge, since it is neither perception nor inference. It is not perception
because it is erroneous and conceptual. Nor is it an inference because it does not arise
from an inferential sign (liga). Nor is it a cognition that is not prama because it
agrees with reality (vastusavda) inasmuch as it leads to efficient action. kyabuddhis
reply is that the cognition in question is in fact an inference. The general characteristic
of inference, he says, is this: Inference is the cognition of an object beyond the range
of the senses (paroka) from its connection to another object. And in the present case,
the erroneous cognition of the jewel arises because of the connection of the sense to the
light of jewel and because the latter (i.e. the light of the jewel) is the result of the jewel.
Therefore, because the cognition of the jewel arises from an inferential sign that is a result
of that jewel (kryaligajatva), it is nothing but inference. First a non-erroneous visual
cognition of the light of the jewel arises. Through this it is apprehended that it has the
nature of an inferential sign that consists in a result (kryaligasvarpa) because only
the particular is an inferential sign29 However, the reasoning is not very convincing.
Although one can establish a concomitance in the form wherever there is light of ajewel,
there is a jewel, the putative inference happens unconsciously. One would similarly
have to assume that the erroneous cognition of fire with regard to smoke is actually an
inference of fire from smoke.
Whatever the case may be, a further distinction between universals seems necessary.
In the examples mentioned above, the universals that rely on both existing and non-
existing things such as being known (jeyatva), and even those that rely exclusively on
non-existing things (e.g., the non-perception of non-existing, but in principle perceptible
things), agree somehow with reality.30 They could be said to cognize particulars in an
indirect way and in a form that is not the particulars own (pararpa, cf. PV 3.54cd quoted
above). There are, however, universals, (or could these be concepts without universals?31)
29 Sanskrit fragment in Steinkellner 1981: 291: tad pratyaknumnavyatirikta ttyam ida pramam
patitam. tath hi maiprabhy maibuddher (read buddhir) na pratyakam, bhrntatvt savikalpakatvc ca.
pratyaka tv etadvipartam. npy anumnam, aligajatvt. na cprama vastusavdt. atrocyate. anumnam
evaitat. tath hy anumnasya smnyalakaam anantara sthpayiyate: parokrthasynyasabandht pratipattir
anumnam iti. iha ca maau maiprabhsambandht tatkryatvt tasym <mai>prabhy maibhrntir
utpadyate. tata kryaligajatvd anumnam eva. tath hi maiprabhym dv abhrntam eva cakurvijnam
upajyate, tena ca kryaligasvarpam adhigatam. yata svalakaam eva ligam
30 A different interpretation is found in Steinkellner 1971: 195, n. 59. According to Steinkellner the third kind

of universals refers to metaphysical errors such as primordial matter (prakti); see also Vetter 1968. Note that
both Steinkellner and Vetter refer to the PV 1.205 (not PV 3.51) and that the example of pradhna is suggested
by Karakagomin. However, the formulation in the two verses is too similar to warrant a different interpretation.
Interestingly, Manorathanandin on PV 1.206b suggests that pradhna and vara rely both on existing and non-existing
things: ubhayopdna pradhnevardi. This does not seem to make sense and perhaps the text is corrupt here.
31 Steinkellner (1971: 195ff.) speaks of Vorstellung ohne Begriff. However, this distinction does not amount

to the postulation of a conceptual construction without a universal; rather, if I understand him correctly, only
universals that rely on existing things are termed Begriff.
60 ELI FRANCO

that have no connection at all to any particulars, for instance, the concept of primordial
matter (prakti, pradhna) mistakenly maintained by the Skhyas, the concept of
God (vara) postulated by the Nyya-Vaieikas, or simply the famous son of a barren
woman.32 Such fictitious entities, inasmuch as they have no counterpart in reality,
cannot exclude anything from any existing particular. Dharmakrti (PVSV 92.21, see
also Steinkellner 1971: 197) calls these conceptual constructions the chief or principal
distinctions (mukhyaviveka). I assume that viveka here is interchangeable with apoha33:
the chief distinction must be the exclusion of everything that exists from something
that does not exist.34 Basically, however, these conceptual constructions operate just as
those that rely on existing things, and in fact they are the clearest cases of exclusion.
For even though there could be a doubt whether conceptual constructions that refer
to existing things do not have a referent in reality after all, in the case of conceptual
constructions referring to non-existing things such doubt does not arise in the first
place.35

32 This example (vandhysuta) is used by ntarakita in the parallel passage TS 1201.


33 Steinkellner (1971: 197) provides two interpretations of viveka; the one, which I follow here, as Sonderung,
the other, in his translation of PVSV 92.21, as Fehlen (absence). In the latter case, he seems to follow Frauwallner
1933: 94 (= Kl. Schr. 449).
34 The question may arise as to whether two different concepts of illusory entities, which equally exclude

all things, would have the same meaning (apoha). For instance, the notions of prakti and purua, both being
completely non-existent and thus excluding all existing things (at least according to the Buddhists), would have to
be synonymous. This, however, is not the case because the exclusion applies not only to individuals/particulars but
also to all other concepts. Thus, the exclusion has a double function; the distinction between these two functions
roughly corresponds to the one between sense and reference, connotation and denotation, intention and extension,
etc. This double function is also seen, of course, in concepts/exclusions that rely on existing things, e.g., being
a product and being impermanent; see the discussion in Tillemans 1986. I owe this reference to the kindness
of Pascale Hugon. The twofold function of apoha can be contrasted with the inability to distinguish universals
from their particulars and from other universals. According to Dharmakrti, the unversals as conceived by the
Nyya-Vaieikas cannot be said to be either different from or identical to (their) particulars or other universals.
It is this inexpressibility which testifies to their unreality; cf. PV 3.32cd-33:
nisvabhvatayvcya kutacid vacann matam ||
yadi vastu na vastnm avcyatva kathacana |
naiva vcyam updnabhedd bhedopacrata ||
Because [the universal] lacks its own nature, it cannot be expressed [as being identical to or different from
the particulars]. If it is thought to be a real thing because it is expressed [as being different] from some [other
universal], [it should also be possible to express it as being identical to or different from the particulars, because]
real things are in no way inexpressible. [A universal] can certainly not be expressed [even as being identical to or
different from other universals] because one speaks about a difference [between universals] only figuratively on
the basis of the difference between [their] substrata.
The inexpressibility of identity and difference is used by Dharmakrti several times as a criterion for unreality,
e.g., PVSV 144.11, 148.5, PV 3.25-27, VN 14.11, 15.6.
35 PV I 185:

rpbhvd abhvasya abd rpbhidhyina |


naky eva siddhs te to vyavacchedavcak ||
Because a non-existing thing has no form (i.e., own nature), words [that denote non-existing things] cannot at all
be suspected to designate [an existing] form. Therefore, it is established for you [that words] express exclusion.
HOW TO DISTINGUISH BETWEEN NON-EXISTING ENTITIES? 61

To conclude, our world is populated by different kinds of non-existing things. There


are illusory objects (conceptual36 and non-conceptual), which include universals, which
again include absences. The universals are subdivided into two groups, depending on
whether they lead to efficient action or not. Although different examples are mentioned and
discussed, Dharmakrti does not seem to suggest different mechanisms for misconceptions
that appertain to everyday life such as the light of a lamp mistook for a jewel, metaphysical
errors such as the concept of primordial matter and God, or misconceptions involving
logical impossibilities such as sons of barren women. The former group of universals,
i.e., those that lead to efficient action, is further divided into three groups depending
on whether they rest on existing things, non existing things, or both. These universals,
which are objects of knowledge, are, to be sure, unreal, and yet they are not completely
unreal like other objects of illusion, for they are somehow and indirectly connected to
real objects. They can be said to be the erroneous aspects through which one cognizes
real things.

Abbreviations and Bibliography

Dreyfus 1997: George Dreyfus, Recognizing Reality. Albany 1997.


Dunne 2004: John Dunne, Foundations of Dharmakrtis Philosophy. Boston 2004.
Frauwallner 1933: Erich Frauwallner, Beitrge zur Apohalehre. I. Dharmakrti. bersetzung
(Fortsetzung). WZKM 40, 1933: 5194.
Hattori 1968: Masaaki Hattori, Dignga, On Perception. Cambridge, Mass. 1968.
Hugon 2011: Pascale Hugon, Is Dharmakrti Grabbing the Rabbit by the Horns? A Reassesment of
the Scope of prameya in Dharmakrtian Epistemology. Journal of Indian Philosophy 39, 2011: 367389.
Inami et al. 2002: Masahiro Inami, Miyako Notake, Keijin Hayashi, Shinya Moriyama, Prajkaraguptas
View on the Two Kinds of Valid Cognitions and the Two Kinds of Objects. Nanto Bukky 81, 2002:
153 [in Japanese].
Katsura 1991: Shoryu Katsura, Dignga and Dharmakirti on apoha. In: Ernst Steinkellner (ed.),
Studies in the Buddhist Epistemological Tradition. Wien 1991: 129146.
Katsura 2011: Shoryu Katsura, From Abhidharma to Dharmakrti With a special reference to the
concept of svabhva. In: Helmut Krasser et al. (eds), Religion and Logic in Buddhist Philosophical
Analysis. Wien 2011: 271280.
Lindner 1984: Christian Lindner, Marginalia to Dharmakrtis Pramavinicaya. WZKS 1984:
149175.
McCrea 2011: Laurence McCrea, Prajkaragupta on the pramas and their objects. In: H. Krasser
et al. (eds), Religion and Logic in Buddhist Philosophical Analysis. Wien 2011: 319328.
Meindersma 1991: Tabe E. Meindersma, A Brief Reference to apoha Theory in the Section on
paralokasiddhi in Pramaavrttika II. In: Ernst Steinkellner (ed.), Studies in the Buddhist Epistemological
Tradition. Wien 1991: 169174.

Cf. also TS 1202:


rpbhvd abhvn abd jtydivcak |
naky eva siddhs te nirbhsasyaiva scak ||
36 Every universal is a conceptual construction, but if, as Prajkaragupta stated, there are some objects of

words that are not universals, does that mean that not every conceptual construction is a universal? See note 25
above.
62 ELI FRANCO

Moriyama 1991: Shinya Moriyama, The Later Madhyamaka and Dharmakrti. In: Ernst Steinkellner
(ed.), Studies in the Buddhist Epistemological Tradition. Wien 1991: 199210.
NBh: Nyyabhaa of Bhsarvaja. Ed. Svm Yogndrnanda. Vrs 1968.
Ono 1999: Motoi Ono, Dharmakrti on asdhranaikntika. In. Shoryu Katsura (ed.), Dharmakrtis
Thought and its Impact on Indian and Tibetan Philosophy. Wien 1999: 301318.
PS I: Digngas Pramasamuccaya, Chapter 1. A hypothetical reconstruction of the Sanskrit text
with the help of the two Tibetan translations on the basis of the hitherto known Sanskrit fragments and
the linguistic materials gained from Jinendrabuddhis k. By Ernst Steinkellner: http://ikga.oeaw.ac.at/
Mat/dignaga_PS_1.pdf
PV 3: Text in: Hiromasa Tosaki, Bukky-ninshikiron no kenky. 2 vols. Tokyo: Dait shuppansha,
19791985.
PVA: Rhula Sktyyana, Pramavrttikabhshyam or Vrttiklakra of Prajkaragupta: Being
a Commentary on Dharmakrtis Pramavrttikam. Patna: K.P. Jayaswal Research Institute, 1953.
PVin: Pramavinicaya Chapter 1 and 2 in: Ernst Steinkellner ed. Dharmakrtis Pramavinicaya,
Chapter 1 and 2. Beijing/Vienna 2007.
PVV: Rhula Sktyyana, ed. Dharmakrtis Pramavrttika with a commentary by Manorathanandin.
Appendix to JBORS 24-26, 19381940.
PVSV: Raignero Gnoli, The Pramavarttikam of Dharmakrti, the first chapter with the autocom-
mentary. Roma 1960.
Steinkellner 1971: Ernst Steinkellner, Wirklichkeit und Begriff bei Dharmakrti. WZKS 1971: 179211.
Steinkellner 1981: Ernst Steinkellner, Philological Remarks on kyamatis Pramavrttikak. In:
Studien zum Jinismus und Buddhismus, ed. K. Bruhn and A. Wezler, Wiesbaden 1981: 283295.
Tillemans 1986: Tom J.F. Tillemans, Identity and Referential Opacity in Tibetan Buddhist Apoha
Theory. In: B.K. Matilal and R.D. Evans (eds.), Buddhist Logic and Epistemology. Dordrecht/Boston/
Lancaster/Tokyo 1986: 207227.
Tillemans 1999: Tom J.F. Tillemans, Scripture, Logic, Language. Essays on Dharmakirti and his
Tibetan Successors. Boston 1999.
TS: Tattvasagraha of Shntarakita. Ed. Dwarikadas Shastri. Varanasi 1981.
Vetter 1968: Tilmann Vetter, Das Problem des metaphysischen Beweises in der logisch-
erkenntnistheoretischen Periode der indischen Philosophie. ZDMG 118, 1968: 351356.
Yoshimizu 2007: Chizuko Yoshimizu, Causal efficacy and spatiotemporal restriction: An analytical
study of the Sautrntika philosophy. In: Pramakrti. Ed. Birgit Kellner et al. Wien 2007. Vol. 2:
10491078.
R O C Z N I K O R I E N T A L I S T Y C Z N Y, T. LXV, Z. 1, 2012, (s. 6374)

SZYMON GRZELAK

Temporal Characteristics of Japanese Verbs


from the Contrastive Perspective

Abstract

In this study we investigate an alternative way to look into the aspectual properties
of Polish and Japanese verbs from the contrastive perspective. The analysis is based on
aformationally motivated classification of Polish aspect forms put forward in Mynarczyk
(2004). Polish verbs are divided into five classes exhibiting a high degree of correspondence
with the Vendlerian verb categories, being also compatible with the main stream accounts
of Japanese aspectual meanings. In the core part of the paper, verbal forms of Polish
and Japanese are tentatively contrasted within the grid of 15 Aktionsarten yielded by
the above verb classes.

Keywords: linguistics, Japanese, Japanese-Polish contrastive analysis

Introduction

This paper outlines a new approach to the contrastive analysis of verbal aspect and
Aktionsart in Polish and Japanese. The main theoretical problem in the comparative studies
juxtaposing the temporal properties of Polish and Japanese verbs is the incompatibility
of the conceptual frameworks applied to the two languages. This study utilizes a recent
analysis proposed by Mynarczyk (2004) to establish as much correspondence as possible
between aspectual forms of Polish verbs and their Japanese counterparts. The paper is
divided into three parts. First, a few remarks on the relation between aspect and Aktionsart
will be presented along with a brief discussion of the most prominent studies on aspect in
both languages. In the second part, some cross-linguistic studies on Polish verbal aspect
will be discussed to demonstrate the main theoretical issues typical for such analyses.
64 SZYMON GRZELAK

Finally, in section three, a possible framework will be presented for a comprehensive


contrastive analysis of Polish and Japanese verbal aspect.

1. Aspect and Aktionsart

The distinction between aspect and Aktionsart initially arose in the field of Slavic
and German linguistics, being rather foreign to the Japanese linguistic tradition. Agrell
(1908) defined Aktionsart as the semantic functions of verb forms that express the way in
which an action is accomplished in a more precise manner than the perfective-imperfective
aspectual opposition, based solely on the criterion of completion of a given action or
process. In this study, aimed at bringing together the two languages traditionally analyzed
in different frameworks, aspect and Aktionsart will be seen, along with tense, as parts
of the temporal characteristics of verbs. The following subsection outlines the relation
between aspect and Aktionsart observed in Polish verbs.

1.1. Aspect and Aktionsart in Polish

Apart from some exceptions, such as modal verbs, Polish verbs obligatorily form
aspectual pairs consisting of a perfective and imperfective member (Polish tryb dokonany
tryb niedokonany), as can be seen in the following examples:

(i) Janek uczy si, ale si nie nauczy.


lit. Johnny studied, but has not studied enough (to the extent of being prepared
for the test)

(ii) Pracuj ju cay dzie, popracuj jeszcze troch i pjd do domu.


I have worked all day already, let me work a bit more and go home.

Imperfective verbs can occur in present, past, and future forms, while the perfective
ones can have past or future tense. Prefixation, suffixation and morphonological changes
are the morphological mechanisms that lead to aspectual forms. The nature of these
formational processes has been subject of some controversy, voiced by Czochralski (1975)
among others, that was extensively discussed in Mynarczyk (2004).

Imperfective Perfective
zielenie to become green zzielenie to have become green
czyta to read przeczyta, poczyta to have read
klika to click zaklika, klikn, poklika to have clicked
TEMPORAL CHARACTERISTICS OF JAPANESE VERBS FROM THE CONTRASTIVE PERSPECTIVE 65

As can be seen from the above table, there is not always a one-to-one correspondence
between the imperfective and perfective forms, which makes it more appropriate to
speak about aspectual tuples or series rather than pairs in the case of majority of Polish
verbs. In the current analysis, Aktionsart will be seen as a more fine-grained semantic
distinction within the aspectual opposition, such as between the three forms zaklika,
klikn and poklika, all being perfective but expressing different distributions of the
event of clicking on the time axis.
Due to the introductory character of this paper, some important theoretical problems
are left out here for the sake of brevity, such as the difference between affixation and
suffixation and their role in the formation of aspectual forms. For example, the verb je
to eat can yield a variety of forms, such as zje to have eaten, podje to eat abit,
przeje to fritter away, nadje to take a bite out of sth, wyje to eat sth out of sth/eat
up everything etc. Distinguishing between the aspectual forms and separate lexical entries
can be carried out by means of diagnostic tools such as secondary imperfectivisation.
The list of Aktionsarten occurring in the accounts of Polish most typically encompasses
the following classes: inchoative, ingressive, terminative, effective, resultative, iterative,
momentanous, durative, intensive semelfactive, among others, but there is no list of
semantic properties that would be unanimously accepted. Further examples of the
Aktionsarten of Polish will be presented in section 2.

1.2. Aspect in Japanese

In his classical account of the Japanese aspectual system Kindaichi (1950, 1976)
classified Japanese verbs into the following categories: stative verbs (zyootai doosi),
continuative verbs (keizoku doosi), instantaneous verbs (syunkan doosi), and type four
(dai yonsyu no doosi) verbs. The first three classes correspond to the Vendlerian (1957)
categories of States, Activities and Achievements, whilst the fourth one is based on the
criterion of co-occurrence of a verbal gerund with the te-iru construction. Below are
some examples of verbs belonging to the above categories (cited after Tsujimura 1996).

Stative aru be, dekiru can do, hanaseru can speak, mieru be visible,
yoo-suru require
Continuative yomu read, kaku write, warau laugh, utau sing, aruku walk,
miru look, nomu drink, osu push, hataraku work
Instantaneous sinu die, kieru turn off, sawaru touch, kimaru decide, sameru
wake, hazimaru begin, tootyaku-suru arrive
Type 4 sobieru tower, sugureru be outstanding, zubanukeru be outstanding,
arihureru be common

The analysis based on the criterion of co-occurrence with the te-iru form has been
further developed by a number of researchers, including Ogihara (1998). More recently,
66 SZYMON GRZELAK

an analysis proposed by Kiyota (2008) in his comparative account with Salish brought
a more detailed division within the class of states.

1. Homogeneous states: iru be/exist, aru be/exist, dekiru be able to


2. Inchoative states: tukareru get tired, okoru get mad
3. Activities: aruku walk, hataraku work, benkyoosuru study
4. Accomplishments: naosu repair s.t., tateru build s.t.
5. Achievements: tuku arrive, todoku reach, mitukaru get found

Other important accounts dealing with aspectual properties of Japanese predicates


include Fujii (1978), Jacobsen (1984), Kudo (1995), and Nishiyama (2006).

2. Cross-linguistic studies on Polish aspect

The existing contrastive accounts on Polish aspect can be divided into three categories:
(i) in the comparative studies with other Slavic languages (e.g. Schuyt 1990), Polish
exhibits a very high degree of morphological and semantic parallelism with differences
due to different paths of historical development of Slavic languages; (ii) comparative
studies analyzing Polish against the background of other Indo-European languages, e.g.
Czochralski (1975) on German; (iii) least frequently, the Polish aspectual system is also
compared with non-Indo-European languages, e.g. Majewicz (1985) on Japanese. In the
following part of the section, the two latter studies will be briefly presented.

2.1. Polish-German contrastive study by Czochralski (1975)

This study, being a contrastive analysis treating Polish and German, is one of the
most robust accounts of Polish aspect that have been proposed to date. The empirical
material used for supporting the analysis consists of a corpus of about 9000 Polish verbs.
The Aktionsarten of Polish are divided into two groups: perfective and imperfective ones:

1. Perfective Aktionsarten:
(a) resultative czyta lesen przeczyta durchlesen; robi machen zrobi
fertigmachen
Reperowae rower, ale go nie zreperowae.
Du warst eine Zeitlang mit dem Reparieren des Fahrrads beschftigt, bist aber
damit nicht fertig geworden.
(b) ingressive: lecie fliegen polecie losfliegen; strajkowa streiken
zastrajkowa in den Ausstand treten
(c) delimitative: czyta lesen poczyta eine Zeitlang lesen; grzeszy sndigen
pogrzeszy ein paar Snden begehen
(d) distributive: gubi verlieren pogubi (alles) nacheinander verlieren
TEMPORAL CHARACTERISTICS OF JAPANESE VERBS FROM THE CONTRASTIVE PERSPECTIVE 67

(e) evolutive: taczy tanzen roztaczy si sich eintanzen


(f) partially resultative: leczy kurieren podleczy teilweise auskurieren

2. Imperfective Aktionsarten:
(a) iterative: by sein bywa zu sein pflegen, pisa schreiben pisywa ab
und zu mal schreiben
(b) comitative: gra spielen przygrywa zu etwas spielen
(c) extended iterative: stka chzen postkiwa fters ein bisschen chzen

Czochralski presents the following functional characteristics of aspect in Polish based


on three semantic distinctions.

Perfective Imperfective
instantaneous, non-durative durative, non-instantaneous
semelfactive, non-iterative iterative, non-semelfactive
resultative, non-preresultative preresultative, non-resultative
In this analysis, Aktionsarten of Polish differ from aspects with respect to the
distribution of the above properties. In the case of an aspectual opposition, the verbs
are ascribed different sets of properties, while in the case of Aktionsarten the differences
are based on single features.

2.2. Polish-Japanese analysis by Majewicz (1985)

This comparative study summarizes the views on Japanese and Polish aspect, discussing
the temporal properties of verbal forms occurring in the two languages. The aspectual
categories of Japanese and their morphological markers as well as the aspectual semantic
classes of Polish are summarized below.

Japanese
Aspectual category Markers
Progressive -te iru, -te oru, -te iru tokoro-da, -te iku, -te kuru, tsuzukeru,
-tsuzuku, -tsutsu aru
Resultative -te aru, (-te iru), -te oku, -ta{tokoro/bakari}da, -ta koto-ga aru,
-tate (da)
Intermittent -tari ...-tari suru
Perfective -te shimau, -kiru, -seru, -tsukusu, -ageru, -tsu, -hateru
Prospective - to suru, -u{tokoro/bakari}-da, -kakaru
Habitual -tsukeru, -nareru, -narawasu, koto-ni shite iru, koto-ga aru
Iterative -kaesu
Ingressive -hajimeru, -dasu, -kakeru, -someru, -te kuru
Terminative -owaru, -yamu, -yameru, -sasu
68 SZYMON GRZELAK

Polish: aspectually unmarked forms, imperfective, durative, iterative, perfective, ingressive,


future perfective and future ingressive.

As can be seen from the two accounts presented above, there are several theoretical
problems that need to be addressed in a comprehensive and adequate contrastive analysis
of Polish and Japanese verbal aspect. Firstly, the question arises of how many Aktionsarten
actually exist in Polish, since the analyses proposed by different authors vary with respect
to their theoretical assumptions. Also, what would be particularly desired is a method
of capturing the correspondence between the morphological form of verbs (affixation,
morphological changes) and their aspectual and Aktionsart-related properties that does
not require relying on semantic intuitions. In addition, in the case of Polish, the issue of
polysemy of some aspectual morphs (eg. po-) needs to be addressed. In the following
section, we demonstrate the view that the analysis proposed by Mynarczyk (2004)
can indeed serve as a theoretical framework for a contrastive analysis of the temporal
properties of Polish and Japanese verbs in a way that quite reasonably addresses the
issues mentioned above.

3. Contrastive analysis based on Mynarczyk (2004)

In this section, we first outline the account that will serve a theoretical base and
subsequently attempt a preliminary contrastive analysis that would consist in actual
confronting of the corresponding verbal forms of the two analyzed languages.

3.1. The classification of Polish verbs

In her 2004 thesis, Mynarczyk provides a very thorough overview of the existing
accounts of verbal aspect, looking in particular into the studies that deal with Polish. The
biggest contribution of the analysis by Mynarczyk is that it systematizes the seemingly
unpredictable and complex system of Polish verbal aspect in a way that is based on the
morphological (formational) criteria, but at the same time induces a semantic classification
corresponding to the theories proposed by Vendler and Dowty (1972). This fact makes
the classification compatible with the studies on the Japanese aspect, which paves the
way to a more adequate contrastive analysis of the two languages.
As the table below shows, Polish verbs are divided into five classes, based on the way
the members of the aspectual pairs are formed. The morphological processes involved in
the formation are prefixation by empty prefixes (ep), prefixation by the delimitative prefix
po-, stem alternation -n-, and morphonological change (mpc), such as vowel insertion,
vowel change, suffix insertion and suffix change. Empty prefixes are morphs that convey
solely the aspectual meaning without contributing any other semantic components to
the overall meaning of the verb they are attached to. This group typically contains such
morphs as prze-, u-, na-, and z-, among others.
TEMPORAL CHARACTERISTICS OF JAPANESE VERBS FROM THE CONTRASTIVE PERSPECTIVE 69

Example ep po- -n- mpc


Class 1 wierzy believe uwierzy
Class 2 gra play pogra
Class 3 gotowa cook ugotowa pogotowa
Class 4 stuka knock zastuka postuka stukn
Class 5 zamkn close zamyka

As we stated above, this formationally-driven classification actually induces semantic


distinctions that are not imposed a priori, but can be related in a straightforward manner to
the morphological properties of the verbal forms. The five semantic classes are summarized
in the following table.

ep po- -n- mpc


Class 1. States/gradual transitions yes
Class 2. Processes yes
Class 3. Culminating processes yes yes
Class 4. Unitisable processes yes yes yes
Class 5. Culminations yes

The final preparatory step in creating the framework for Polish-Japanese analysis
consists in deriving all the possible Aktionsarten occurring in the above five classes. As
can be seen, the total number of all such aspectual subclasses amounts to 15. This grid
of Aktionsarten will serve a basis for the subsequent contrastive analysis.

3.2. Outline of Polish-Japanese contrastive analysis

This section contains a tentative attempt of a contrastive analysis carried out within the
grid of the 15 Aktionsarten obtained from the analysis by Mynarczyk. The Aktionsarten
of Polish yielded by the above classification will be marked as A1 A15.

Class1 States

A1: stative, Imperfective


nienawidzi
(1) Taro nienawidzi Hanako.
Taroo wa Hanako o nikumu / nikundeiru.
Taro hates Hanako.
70 SZYMON GRZELAK

A2: Perfective
ep znienawidzi
(1) Taro znienawidzi Hanako.
Taroo wa Hanako o nikunda / nikumihazimeta / ?nikumidasita /?nikundekita.
Taro started to hate Hanako.

Class1 Gradual transitions


A3: Imperfective
mdrze
(1a) Taro mdrzeje (kadego dnia).
Taroo wa (mainiti) kasikoku natteiru/naru/naritutu aru.
Taro gets smarter (every day).

A4: Perfective
ep zmdrze
(1a) Taro zmdrza.
Taro owa kasikoku natta/nattekita.
Taro has become smart.

Class2 Processes
A5: Imperfective
pracowa
(2) Taro pracuje.
Taroo wa hataraiteiru/hataraku.
Taro works/is working.

A6: Perfective
po- popracowa
(2) Taro popracowa (godzin) (i poszed na spacer).
Taroo wa (itizikan) hataraita (ato sanpo ni itta).
Taro worked (for an hour) (and went for a walk).

Class3 Culminating processes


A7: Imperfective
czyta
(3) Hanako czyta gazet. / Hanako czyta gazety.
Hanako wa sinbun o yondeiru / Hanako wa sinbun o yomu/yondeiru.
Hanako reads a newspaper / Hanako reads newspapers.
TEMPORAL CHARACTERISTICS OF JAPANESE VERBS FROM THE CONTRASTIVE PERSPECTIVE 71

A8: Perfective
ep przeczyta
(3) Hanako przeczytaa gazet.
Hanako wa sinbun o yonda/yondesimatta/yomiowatta/yomikitta/yomitukusita/
yomiageta/yomitoosita/yomihateta.
Hanako has read a newspaper.

A9: perfective2
po- poczyta
(3) Hanako poczytaa gazet (i zabraa si do pracy).
Hanako wa sinbun o (sibaraku/sukosi) yonda (ato sigoto ni torikatakkta).
Hanako read a newspaper (for a while) (and started working).

Class4 Unitisable processes

A10: Imperfective
kwaka
(4) Kaczka kwacze.
Ahiru ga gaa-gaa naku/naiteiru.
The duck quacks / is quacking.

A11: Perfective
ep zakwaka
(4) Kaczka zakwakaa.
Ahiru ga gaa-gaa naita/nakidasita/nakihazimeta/#naitekita/#nakikaketa.
The duck quacked.

A12: po- pokwaka


(4) Kaczka pokwakaa.
Ahiru ga (sukosi/sibaraku) gaa-gaa naita/?naiteita.
The duck quacked a bit / for a while.

A13: mpc kwakn


(4) Kaczka kwakna.
?Ahiru wa ikkai gaa-gaa naita.

Ahiru wa gaa(t)to naita. (cf. koro-koro korogatta vs. korori-to / koron-to korogatta)
The duck quacked once.
72 SZYMON GRZELAK

Class5 Culminations

A14: Perfective
dosta
(5a) Hanako dostaa ksik.
Hanako wa hon o moratta.
Hanako got a book.

wrci
(5b) Hanako wrcia do domu.
Hanako wa ie ni kaetta.
Hanako came back home / has come back home.

A15: Imperfective
mpc dostawa
(5a) *Hanako dostawaa ksik.
*Hanako wa (issatu no) hon o moratteita/moraitutuatta/moraituduketa.
*Hanako was getting a book.

wraca
(5b) Hanako wracaa do domu.
Hanako wa ie ni #kaetteita/kaeritutuatta. (cf. kaeru totyuu datta)
Hanako was going back home.

In view of the above examples, the following correspondence between Class1 Class5
and the the aspectual classes of Japanese verbs can be established.

Polish Japanese
States/gradual transitions States
Processes Activities
Culminations Achievements
Unitisable processes Semelfactive verb phrases (onomat./ikkai)
Culminating processes Accomplishments

The only class that is normally not differentiated in the accounts that deal with
Japanese are the semelfactive verbs referred to as unitisable processes by Mynarczyk,
who looks at them from the opposite side than is usually the case, i.e. departing from
the imperfective verbs that can potentially yield semelfactive forms. These meanings
cannot be encoded in Japanese by means of verbal morphology and need to be obtained
from adverbial expressions as in the examples above.
TEMPORAL CHARACTERISTICS OF JAPANESE VERBS FROM THE CONTRASTIVE PERSPECTIVE 73

Despite its many advantages, the analysis has also some weak points. Firstly, it is
unidirectional, and therefore does not show the whole picture. Some aspectual categories
are not marked in the Polish verbal morphology (e.g. progressive), therefore, for a class
of cases, no Japanese-Polish formational correspondence can be established. Secondly,
some verbs are not covered by class1 class5 (e.g. iterative/habitual, suppletive pairs
such as bra wzi to take to have taken, modal verbs mc, musie), the analysis
needs thus to be supplemented. Finally, an important theoretical issue is the mismatch
between the word- and phrase/sentence-level aspect. Much more can be said about the
temporal characteristics of bare verb forms in Polish than in Japanese where analysis of
the whole phrase/sentence is necessary.

Summary

In spite of the tradeoff between the simplicity and regularity on one hand, and
range of coverage on the other, the formationally-driven classification proposed by
Mynarczyk (2004) seems to be successfully applicable in a contrastive analysis of
Polish and Japanese verbal aspect. In order to account for all possible Aktionsarten
of Polish, the 15 eventualities yielded by the classification should be analyzed
and the analysis should be complemented by the verb classes that are not covered.
A classification based solely on the morphological properties of the verbal forms is
fragmentary by its nature, and any comprehensive account covering the event structure
of verbs will have to rely on semantic intuitions as well. Finally, it goes without saying
that the general idea presented in this paper requires further development, based on
substantial empirical material from both languages.

References

Agrell, S. 1908. Aspektnderung und Aktionsartbildung beim polnischen Zeitworte: ein Beitrag zum
Studium der indogermanischen Prverbia und ihrer Bedeutungsfunktionen. Lund: Ohlsson.
Czochralski, J. 1975. Verbalaspekt und Tempussystem im Deutschen und Polnischen. Eine konfrontative
Darstellung. Warszawa: Pastwowe Wydawnictwo Naukowe.
Dowty, D. 1972. Studies in the logic of verb aspect and time reference in English. Doctoral Dissertation,
Department of Linguistics, Austin, Texas.
Fujii, T. 1976. Doosi + te- iru no imi (The meaning of verb + te- iru). In: Kindaichi, H. (ed.).
Nihongo doosi no asupekuto. Tokyo: Mugi Shobo, pp. 97116.
Jacobsen, W.M. 1984. Lexical Aspect in Japanese. In: Testen, D.; Mishra, V.; Drogo J. (eds.),
Papers from the Parasession on Lexical Semantics. Chicago: Chicago Linguistic Society, pp. 150161.
Karolak, S. 1996. O semantyce aspektu (w dwudziest rocznic publikacji rozprawy F. Antinucciego
i L. Gerbert Semantyka aspektu czasownikowego). Bulletin de la Socit Polonaise de Linguistique,
pp. 956.
Kindaichi, H. 1976. Kokugo doosi no itibunrui. In: Kindaichi, H. (ed.). Nihongo Doosi no Asupekuto.
Tokyo: Mugi Syobo.
74 SZYMON GRZELAK

Kiyota, M. 2008. Situation Aspect and Viewpoint Aspect: from Salish to Japanese. Doctoral Dissertation.
The University of British Columbia.
Kudo, M. 1995. Asupekuto-tensu taikei to tekusuto. Tokyo: Hituzi Syobou.
Majewicz, A.F. 1985. The grammatical category of aspect in Japanese and Polish in a comparative
perspective. Pozna: Wydawnictwo Naukowe UAM.
Mynarczyk, A. 2004. Aspectual Pairing in Polish. Doctoral dissertation, UiLOTS Utrecht University.
Nishiyama, A. 2006. The Meaning and Interpretations of the Japanese Aspect Marker -te-i-. Journal
of Semantics 23, pp. 185216.
Ogihara, T. 1999. Tense and Aspect. In: Tsujimura, N. (ed.). The Handbook of Japanese Linguistics.
Oxford: Blackwell Publishers.
Schuyt, R. 1990. The Morphology of Slavic Verbal Aspect. A Descriptive and Historical Study.
Amsterdam & Atlanta, GA: Rodopi.
Tsujimura, N. 1996. An Introduction to Japanese Linguistics. Oxford: Blackwell Publishers.
Vendler, Z. 1957. Verbs and times. The Philosophical Review 66, pp. 143160.
R O C Z N I K O R I E N T A L I S T Y C Z N Y, T. LXV, Z. 1, 2012, (s. 7589)

JOANNA JUREWICZ

What do Ancient Indian Cosmogonies Tell us about Language?

Abstract

The aim of the paper is to reconstruct the Vedic concept of language as presented
in the Brhmaas and to show that, according to the Vedic thinkers, language not only
expresses thoughts but also it is a social phenomenon which cannot develop without a
speaking community. It is argued that cosmos, gods and society divided into four social
states (vara) can be seen as Gods language and that the creation of the cosmos and
society conveys the concept of God, who not only wants to express itself in its creation
but also to talk via various speakers. This would reveal the Vedic concept of language
seen as the means of self-expression in discourse.

Keywords: India, ancient cosmogonies, Veda, language

1. Introduction

Contemporary studies on culture see thinking and language as social phenomena. In


his book The interpretations of culture Clifford Geertz wrote:

Human thought is basically both social and public () its natural habitat
is the house yard, the market place, and the town square. Thinking consists
not of happenings in the head (though happenings there and elsewhere
are necessary for it to occur) but of a traffic in () significant symbols
words for the most part but also gestures, drawings, musical devices like
clocks, or natural objects like jewels, anything, in fact, that is disengaged
from its mere actuality and used to impose meaning upon experience.1

1 Geertz 1973: 45.


76 JOANNA JUREWICZ

Enrique Bernrdez summarises the basic assumptions of the study of language as a


social activity in the following way:

() Human languages exist only in the form of social activity. ()


Linguistic activity is essentially collective, cross-individual, i.e. is not
simply carried out inside a (social) group, but the reasons for its realisation,
the form of its realisation and the results of the activity itself are collective,
social in nature: in other words, the process of linguistic activity cannot
be understood solely in terms of the individual.2

Norman Fairclough goes even further when he sees language as

a part of society; linguistic phenomena are social phenomena of a special


sort, and social phenomena are (in part) linguistic phenomena.3

In this paper, I would like to reconstruct the Vedic concept of language as presented
in the Brhmaas and to show that, according to the Vedic thinkers, language not only
expresses thoughts but also it is a social phenomenon which cannot develop without
a speaking community. This concept of language is close to our modern understanding.
In order to show that, I will use the basic assumptions of cognitive linguistics. This
branch of linguistics assumes close relationship between human experience, language
and thought, and it also sees language as inextricably connected with society.4 Human
thought and language are motivated by physical, cognitive and social experience. As
Evans and Green put it: we can only talk about what we can perceive and conceive,
and the things we can perceive and conceive derive from embodied experience (Evans,
Green 2006:46). Universal to all human beings and closely connected with their physical
functioning and perception, this experience is the ground for universal concepts. However,
as it is largely mixed with culture-specific experience, many concepts vary according to
a culture or a set of cultures (see e.g. Kvecses 2005).
We organise our knowledge in mental structures that are activated in discourse via
significant symbols (in Geertzs terminology): they can be linguistic, but also visual,
auditory, etc. In his cognitive analysis of language, Taylor (2002: 589) calls these structures
domains5 and defines them as any knowledge configuration which provides a context
for conceptualisation. For example, in order to understand what thumb means we have
to activate the concepts of the palm and the hand, which in turn lead us towards the
concept of the human body.

2 Bernrdez 2005: 211.


3 Fairclough 2001: 19.
4 Cf. egg. Geeraerts 2005, Sinha 2009.
5 These conceptual wholes are also called frames, idealised cognitive models, mental spaces. Each of

these terms highlights a different aspect of the conceptual whole within its specific theoretical framework (Radden
1992: 527, Cienki 2007: 183184).
WHAT DO ANCIENT INDIAN COSMOGONIES TELL US ABOUT LANGUAGE? 77

As cognitive research shows, there are two basic conceptual strategies that human
beings use to think and to speak about various elements of their experience. These are
called metaphor and metonymy. Metaphor is a mapping that projects one concept onto
another and thus makes it possible to think about the latter in terms of the former. The
concept that lends its categories is called the source domain, the concept conceived by
means of these categories is the target domain. For example, in the Indo-European culture
cognition is conceived in terms of seeing;6 seeing is the source domain, cognition
is the target domain. Conceptual metaphor reflects itself in language, which is why we
can meaningfully say: I see what you mean. It is obvious that in this sentence we do
not speak about seeing, but we have to evoke this concept in order to understand aspects
of the concept of cognition expressed by it. As we can see, the target domain is more
abstract, but, conceived in terms of a more concrete source domain, closely connected
with everyday life experience, it becomes easier to understand and speak about.
Metonymy operates within one conceptual domain: its one element (the vehicle)
provides access to the concept or to its element (the target).7 For example, there is a
metonymy that operates between the concepts of part and whole. This metonymy is of
two kinds. In the first case, the target concept is part, the vehicle is whole (Whole For
Part). Thanks to this metonymy, we can meaningfully say during Christmas: Lets light
the Christmas tree instead of Lets light the candles on the Christmas tree.8 In the second
case, the target concept is whole, the vehicle is part (Part For Whole) and thanks to this
metonymy we can be sure that we will be understood showing a photo and saying This
is my daughter instead of This is the face of my daughter.
One of the most important concepts that serves as the source domain in the Vedic
philosophy is human being: in these terms the Vedic thinkers thought about God, cosmos
and society.9 As it is shown by cognitive linguistics, not all the aspects of the source
domain are mapped onto the target domain.10 The most important features of the human
being that were used in metaphoric conceptualisation in the Vedic philosophy include:
firstly, cognitive and emotional abilities and, secondly, language in which thinking is
expressed. In the frames of this conceptualisation, creation of the world was seen as a
cognitive process during which God wants to, and consequently, does express himself
in language. This view of creation reveals the general assumption, underlying the whole
Vedic philosophy, according to which cognition precedes, creates, and conditions being.
In this paper, I will argue that cosmos, gods and society divided into four social states
(vara) can be seen as Gods language and its speakers. And I hope to show that
understood in these terms the creation of the cosmos and society conveys the concept

6 Sweetser 1990.
7 Radden, Kvecses 1999: 21.
8 Radden, Kvecses 1999: 31.
9 In the Upaniads, conceptualisation of the soul in terms of the human being begins (the soul is called man

in size of a thumb, agutamtra purua: Kaha Upaniad 4.12-13, 6.17, vetvatara Upaniad 3.13, 5.8).
Cf. DesCamp, Sweetser 2005 for a cognitive analysis of metaphorical thinking about God in the Bible.
10 Lakoff, Johnson 1980.
78 JOANNA JUREWICZ

of God, who not only wants to express itself in its creation but also to talk via various
speakers. This would reveal the Vedic concept of language seen as the means of self-
expression in discourse.

2. Creation of cosmos is creation of language

The concept of the cosmos created by God as he creates language is ubiquitous in the
Brhmaas and was analysed by many scholars.11 I will analyse the cosmogony presented
in atapatha Brhmaa 11.1.6 in order to show not only the cognitive dimension of
creation but also how it is expressed via conceptual metonymies and metaphors:
[1] atapatha Brhmaa 11.1.6 (Eggelings translation with my minor changes):12

Verily, in the beginning this [universe] was water, nothing but a sea of
water. The waters desired, How can we be reproduced? They toiled
and became heated13, when they were becoming heated, a golden egg
was produced. This year, indeed, was not then in existence: this golden
egg floated about for as long as the space of a year. (1)
In a years time a man, this Prajpati was produced there from; and hence
a woman, a cow, or a mare brings forth within the space of a year; for
Prajpati was born in a year. He broke open this golden egg. There was
then, indeed, no resting-place: only this golden egg, bearing him, floated
about for as long as the space of a year. (2)
At the end of a year he tried to speak. He said bh: this [word] became
this earth; bhuva: this became this air; sva: this became yonder sky.
Therefore a child tries to speak at the end of a year, for at the end of
a year Prajpati tried to speak. (3)
When he was first speaking Prajpati spoke [words] of one syllable and
two syllables; whence a child, when first speaking, speaks [words] of
one syllable and of two syllables. (4)
These [three words consist of ] five syllables: he made them to be the
five seasons, and thus there are these five seasons. At the end of the
(first) year Prajpati rose to stand on these worlds thus produced; whence
a child tries to stand up at the end of a year, for at the end of a year
Prajpati stood up. (5)

11 E.g. Lvi 1898, Gonda 1983, Holdrege 1994, Holdrege 1996: 47 ff.
12 Since English is not my first language, I quote Sanskrit texts in already published translations. Sometimes
I introduce changes when the translation goes, in my opinion, too far from the original.
13 Eggeling 1994, V: 12: performed fervid devotions but he adds in a note: Or, they toiled and became heated

(with fervid devotion).


WHAT DO ANCIENT INDIAN COSMOGONIES TELL US ABOUT LANGUAGE? 79

The source domain of the creative process is a complex scenario of a sexual intercourse,
insemination, pregnancy, birth-giving and growing up of a child. The description begins
with the conventional formula X idam agra st/sa but it replaces Prajpati, the usual
subject of the formula, with the concept of water (pas). However, in my opinion, the
cosmogony does not present a different vision of the beginnings of creation. It rather
begins its description with the next phase of creation, which is conceived in terms of water.
It is possible to reconstruct the experience that lies behind this description. In order
to do it, we have to refer to the cosmogony presented in atapatha Brhmaa 6.1.3. It
begins with the most conventional form of the cosmogonic formula (prajpatir v idam
agra st, Verily, Prajpati alone was here in the beginning14), then it presents Prajpati
who heats himself, water coming out of him. As the cosmogony says, this is the reason
why people, when they are heated, sweat. The cosmogony sees everyday experience as
the result of what happened during creation but the real motivation runs in the opposite
direction: it is everyday experience that influences thinking about creation. Since in
everyday life people sweat from heat, Prajpati, metaphorically conceived in terms of
a human being, sweats too. In the Vedic thought, sweating is treated as transformation
under the influence of heat15 and in cosmogonic descriptions the concept of sweat becomes
the source domain for the first manifested form of Prajpati.16 According to atapatha
Brhmaa 6.1.3, this form heats again to become the next form of Prajpati, that of foam,
then, successively, Prajpati heats his manifested forms to become clay, sand, pebble,
stone, metal ore and, finally, gold. The logic of the source domain of a sweating human
being allows the recipient to understand that only a part of Prajpati, conceived in terms
of sweat, undergoes creative transformations. This is his external part which becomes
shining and thus visible while the internal part remains unmanifested and cannot be seen.
The composer of atapatha Brhmaa 11.1.6 (example [1]) metonymically refers to
the same concept of a sweating human being. In this way, he begins cosmogony from
the stage when Prajpati has already manifested himself. At the same time, thanks to
this metonymy, the description of the beginnings of creation is more abstract because
water is a more abstract concept than a sweating human being. What is more, the use
of the general notion of water, without explicit reference to the concept of sweating,
allows the recipient to activate various metaphoric and metonymic mappings that give
him a profound insight into creation.
The concept of water metonymically prompts the recipient to evoke the concept of
a pregnant woman (Water For Womb, Womb For Woman).17 If the recipient understands
the first manifested form of Prajpati in terms of a woman, he will understand that during

14 Eggelings translation (Eggeling 1994, III: 157).


15 Cf. Jurewicz 2006; 2010: 268.
16 One of the clearest example is Gopatha Brhmaa 1.1.
17 Metonymy Water For Womb is an example of metonymy Contents For Container. Metonymy Womb For

A Woman is an example of metonymy Part For Whole.


80 JOANNA JUREWICZ

the creative act Prajpati, who in his unmanifested form is conceived as a man, becomes
his own opposition, which is conceived in terms of a woman.18
Secondly, the concept of water activates the metaphor Language Is Water.19 The
coherence of this activation is confirmed by another metaphor of language, well grounded
in the Veda, i.e.: Language Is A Woman.20 In this case, the recipient conceives the
beginnings of creation as the appearance of language in general.21
Later, water is shown to be heating itself, and then the golden egg appears. This agrees
with conventional way of presenting cosmogonic transformation in terms of toiling and
heating (expressed by the formula: so rmyat sa tapo tapyata, he toiled, he became
heated), which confirms that atapatha Brhmaa 11.1.6 follows the general cosmogonic
pattern of the Brhmaas.22
The golden egg (hiramayam am) is a complex conceptual whole. Firstly, it
evokes the concept of an egg in terms of which the beginning of life is conceived. In
this way, the composer activates the concept of a sexual act and successful insemination.
In these terms, the creative unity of the opposing aspects of the Creator is conceived.
If the recipient evokes conceptualisation of language in terms of water/woman, he
will understand the cognitive dimension of this phase of creation during which Prajpati
wants to express himself in words; as we will see below, expression of thoughts in
language is conceived in the Veda in terms of a sexual intercourse between a man and a
woman. This cognitive dimension is also expressed via the concept of boiling water which
produces the golden egg. The recipient can metonymically activate the concept of gold
seen as the product of heating. In terms of such processes thinking is conceived in the
Veda: in the frames of this conceptualisation gold is the source domain for thought.23
Finally, the concept of the golden egg evokes the concept of the sun24 as the first
form of the cosmos which will be later divided and organised. Thus the recipient will
see the ontic dimension of the cognitive act.
We can see then that the careful use of concepts that function as the source domains
for the conceptualisation of creation allows the composer to evoke the scenario of sexual

18 Prajpati becomes a woman in creation: in atapatha Brhmaa 6.1.2 (see, below, example [2]) he gets

pregnant, in atapatha Brhmaa 2.5.1.3 he feeds his offspring with milk from his breast.
19 Cf. e.g. Tya Mahbrhmaa 20.14.2, Jaiminya Brhmaa 2.244. For other references cf. Holdrege 1996:

445 (note 111).


20 Interestingly enough, the conceptualisation of language in terms of a woman reveals aspects of the Vedic

(maybe not only Vedic?) stereotype of a woman: her main features are frivolousness and fickleness. In this way,
the difficulty with gaining and preserving the language was expressed (cf. atapatha Brhmaa 3.2.1.18ff, 4.2.4.2,
3.5.1.21ff.).
21 Cf. Holdrege 1994: 45, 1996: 4950.
22 In atapatha Brhmaa 6.1.3 gold is also the product of heating.
23 Cf. Jurewicz 2010: 181. The concept of gold-transformation was very much elaborated by the Buddha, cf.

Covill 2009: 184214.


24 The concept of golden egg evokes the concept of the semmetonymically (via metonymy Colour For The

Substance) and on the basis of similarity of shape.


WHAT DO ANCIENT INDIAN COSMOGONIES TELL US ABOUT LANGUAGE? 81

human activity and transformation under the influence of heat. In terms of these activities,
cognitive and creative processes are conceived.
The following sections of atapatha Brhmaa 11.1.6 develop the scenario of human
sexual activity with its prototypical results: pregnancy, giving birth and development of
a child. The golden egg remains in water for one year and then a boy is born. In these
terms the next manifested form of Prajpati is conceived, immanent to the world. Since
the Vedic thinkers assumed that father was reborn in his own son, the recipient knows
that Prajpati immanent to the world is the same as Prajpati in his unmanifested form.
Prajpati conceived in terms of a new-born child cannot walk for one year (he has
no resting-place) and he is very close to his mother (inside the golden egg which is
only half-open). After a year he begins to stand up and speak. And, we could say, he
expresses his experience: he successively pronounces words that correspond to the stages
of his movement up: first, the word for the earth, next, words for the space between the
earth and the sky, and the sky itself. The words have creative power: to say the word
earth is to create the earth, to say the word space is to create the space, to say the
word sky is to create the sky. The creation of spatial division is simultaneous with
the creation of time division realised while Prajpati stands up and begins to move. We
can see then that Prajpati, who creates the cosmos, is conceived in terms of a child
who, learning the language, as if creates the world for himself. Contrary to everyday
experience, the cosmos is ontologically identical to Prajpati: he becomes what he says.
If we interpret water as language in its primeval, undifferentiated form, we can see
that now it is divided into words, so that Prajpati can precisely express himself and his
creative activity. Thus Prajpati can express the way he manifests himself in the cosmos:
he divides himself into three parts and moves. Words bh, bhuva, sva become parts
of the cosmos.25

2. Creation of gods is creation of language

In many places in the Brhmaas, the cognitive core of creation is reduced to


interaction between mind and language. In some cosmogonies, Prajpati is conceived
in terms of the mind and he gives birth to language. Language is conceived in terms
of a woman and it copulates with Prajpati, who is a man in this case.26 Metaphoric
conceptualisation of Prajpati in terms of the mind is motivated by metonymy Mind
For Human Being. We can see again how metonymical thinking leads to abstraction:
the concept of the mind is more abstract than the concept of the human being and the
metaphorical concept of God-as-mind is more abstract than the metaphorical concept of
God-as-human being.
25 It is important to note that the concept of language with its ontological dimension is evoked in the final
part of the cosmogonic description of atapatha Brhmaa 6.1.3 discussed above: the gold flows (akarat) and
becomes eight-syllable gyatr which in turn becomes the earth (6.1.3.6).
26 Cf. Lvi 1898: 2223, Gonda 1983: 29ff., Holdrege 1994: 41ff., 1996: 47ff.
82 JOANNA JUREWICZ

In other cosmogonies Prajpati gives birth to both mind and language, which copulate
with each other,27 as in the following cosmogony that presents the creation of gods who
inhabit three spheres of the cosmos:

[2] atapatha Brhmaa 6.1.2 (Eggelings translation with my minor changes):


Having created these worlds, he desired, May I create such creatures as
shall be mine in these worlds! (5)
By his Mind (manas) he entered into union with Language (vc)28:
he became pregnant with eight drops. They were created as those eight
Vasus, he placed them on this earth. (6)
By his Mind (manas) he entered into union with Language (vc): he
became pregnant with eleven drops. They were created as those eleven
Rudras, he placed them in the air. (7)
By his Mind (manas) he entered into union with Language (vc): he
became pregnant with twelve drops. They were created as those twelve
dityas he placed them in the sky. (8)
By his Mind (manas) he entered into union with Language (vc):
he became pregnant. He created the All-gods: he placed them in the
quarters. (9)

The gods are created in a cognitive act where the relation between mind and language
is conceived in terms of a sexual union between a man and a woman. The Brhmaic
philosophers knew that thoughts are expressed in language:
[3] atapatha Brhmaa 12.9.1.13 (Eggelings translation with my minor changes):

Whatever one thinks in his mind (manas) of that he speaks with his
language (vc),29 and what he speaks with his language that one hears
with ones ears.30

We can then presume that the concept of sexual act was used to conceive the
externalisation of thoughts in language: when spoken thoughts and words become one as
a man and a woman become one during sexual act. We can also infer that the happiness
gained during this act was the source domain used to conceive satisfaction of the speaker
gained thanks to adequate verbalisation of thinking. At the same time, the logic of the
scenario of giving birth allows the composer to express the ontological results of speaking.
Gods are divided into groups. The number of group members is conventionalised in
the Brhmaas: there are eight Vasus, eleven Rudras and twelve dityas. However, it is

27 Cf. Lvi 1898: 2223, Gonda 1983: 29ff., Holdrege 1994: 41ff., 1996: 47ff.
28 Eggeling 1994, III: 149150 translates vc as Speech.
29 Eggeling 1994, V: 263 translates vc as Speech.
30 According to the atapatha Brhmaa 12.9.1.11 the sacrificer is the language which is the manifested

(pratyakd) mind, in atapatha Brhmaa 10.5.1 language is the first manifestation of the mind.
WHAT DO ANCIENT INDIAN COSMOGONIES TELL US ABOUT LANGUAGE? 83

important to note that the number of members of each group agrees with the number of
syllables of the meters that correspond to the regions of the world: eight-syllabled gyatr
corresponds to the earth, eleven-syllabled triubh corresponds to the space between
the earth and the sky, twelve-syllabled jagat corresponds to the sky.31 The atapatha
Brhmaa testifies correspondence between Vasus, Rudras and dityas on the one hand
and the earth, the space and the sky, respectively.32 So the recipient can see the creation
of the gods as the creation of metres. In this case, atapatha Brhmaa 6.1.2 (example
[2]) would present a more complex diversification of Prajpatis language than atapatha
Brhmaa 11.1.6 (example [1]): there are 31 syllables that can be ordered in metres. At
the same time, since the expression in language has its ontic results (the words for parts
of the cosmos become parts of the cosmos), the gods are not only syllables but also the
speaking subjects who will use Prajpatis language.
We can conclude then that in the creative act Prajpati thinks about himself and
expresses himself in a more and more complicated language. In this cognitive process,
he becomes the world and its inhabitants. Such a concept of creation appears already in
the gveda 1.164, where the Creator is conceived in terms of she-buffalo who multiplies
her legs in water.33 The legs can be interpreted as words and the water as the first
undifferentiated form of language expressed in words during creation. Thus the whole
image presents creation conceived in terms of a cognitive process: the Creator describes
her worldly appearance more and more adequately. In the same way, Prajpati multiplicates
himself in words that express his cognitive activity more and more precisely.
But expressing oneself is only one role of language. Another one is interaction with
others, communication with them. To realise it fully, God creates manifold users of
language. Some of them are gods. Others are human beings.

3. Creation of society is creation of language

As it is well known, Vedic cosmogonies do not describe the creation of the first
human being. The mankind is created as a whole and is identified with the Indian society
divided into four states (vara). As I have already said, the society is metaphorically
conceived in terms of a human being too: parts of Prajpati become parts of the society,
in the same way as his parts are the parts of the cosmos. Creation of four social states
is described in the following cosmogony:

31 Smith 1994b: 127 ff.


32 atapatha Brhmaa 6.5.2.3-5, 13.2.6.4-6, cf. Smith 1994b: 299. The All-gods are usually treated as gods of
the Vaiyas (cf. Smith 1996b: 96), and as such they should be connected with the sky and jagat metre. However,
in this particular cosmogony their connection with directions of space (d) implies their connection with time:
because it is space that enables movement measured by time. It is confirmed by identification of the All-gods
with the seasons in atapatha Brhmaa 7.1.1.43, connection of the All-gods with quarters is also expressed in
atapatha Brhmaa 6.5.2.6).
33 Cf. Jurewicz 2010: 8689.
84 JOANNA JUREWICZ

[4] Jaiminya Brhmaa 1 (Bodewitzs translation with my minor changes):34

Prajpati [and nothing else] existed here in the beginning. Now Prajpati
was the mind.35 He desired: May I become manifold. May I procreate.
May I become abundant. He created from his top, from his head, the
Triv{t laud, the Gyatr metre, the Rathantara melody, the deity Agni,
the human being Brahmin, the animal goat. Therefore the Brahmin has
the Gyatr as his metre and Agni as his deity. And therefore also he is
the mouth of creatures. For he [Prajpati] created him from his mouth.
He desired: May I procreate. He created from both his arms and from
his breast the Fifteenfold laud, the Triubh metre, the B{hat melody,
the deity Indra, the human being Katriya, the animal horse. Therefore
the Katriya has the Triubh as his metre and Indra as his deity. And
therefore he displays his force with his arms. For he [Prajpati] created
him from his two arms, from his breast, from his force.
He desired: May I procreate. (68)
He created from his belly, from his middle, the Seventeenfold laud, the
Jagat metre, the Vmadevya melody, the deity All-gods, the human being
Vaiya and the animal cow. Therefore the Vaiya has the Jagat as his
metre and the All-gods as his deity. And therefore he is always intent on
producing (procreating). For he [Prajpati] created him from his belly,
from his generative organ.
He desired: May I procreate. He created from his two feet, from his
support, the Twenty-onefold laud, the Anuubh verse, the human being
dra and the animal sheep. Therefore the dra has the Anuubh as
his metre and the landlord as his deity. And therefore he desires to earn
his living by washing feet. For he was born from the support (feet) of
Prajpati. (69)

Prajpati, conceived of as the mind, creates language, which consists of lauds (stoma)
composed in particular metres (chandas) and has particular melodies (sman). Then,
Prajpati creates gods (Agni, Indra, All-gods) and human beings divided into four states.
As it has been argued above, the creation of groups of gods (Vasus, Rudras, dityas)
can be seen as the creation of Prajpatis language. In the same way, the creation of
particular gods can be seen. There are cosmogonies that present correspondences between
various gods and elements of language, such as vyh{tis (bh, bhuva, sva), metres
(chandas), lauds (stoma), melodies (sman).36 So the recipient can understand creation
of particular gods as the creation of language too.

34 For variants of this cosmogony in other Brhmaas, cf. Smith 1994b: 6566, 69. Cf. also Smith 1994a: 80ff.
35 Bodewitz 1990: 38: (vital) power mind.
36 Smith 1994b: 75, 79, 81, 129, 138139.
WHAT DO ANCIENT INDIAN COSMOGONIES TELL US ABOUT LANGUAGE? 85

What is more, as Smith (1994b) shows, the Brhmaas clearly imply the homology
between social states and elements of language.37 Taittirya Brhmaa 3.12.9.2 is one
of the most lucid passages in this respect, where it is said that three social states were
born of three Vedas.38 So the recipient can also interpret the creation of human beings
as the creation of Prajpatis language, similarly to the creation of gods.
This is attested by the very term denoting the social states, which is vara. Traditionally,
in the context of the society, vara is understood as colour.39 But the word vara also
means a sound of speech.40 In this sense, vara is used in the following cosmogony
which conceives creation as an extraction of the linguistic essence of the three parts of
the cosmos:
[5] Aitareya Brhmaa, 5.32 (Keiths translation with my minor changes):

Prajpati desired May I be propagated; may I be multiplied. He became


heated, having become heated41 he created these worlds; the earth, the
atmosphere, the sky. He heated (abhyatapat)42 these worlds; from these
worlds when heated (tebhyo bhitaptebhyas) these luminaries (jyotii)
were born; Agni was born from the earth, Vyu from the atmosphere,
ditya from the sky. He heated these luminaries. From these heated the
three Vedas were born; The gveda was born from Agni, the Yajurveda
from Vyu, the Smaveda from ditya. He heated these Vedas; from these
[Vedas] when heated three pure [sounds, ukra] were born: bh from
the gveda, bhuva from the Yajurveda, sva from the Smaveda. He
heated these pure ones; from them when heated the three sounds (vara)
were born; the letter a, the letter u, the letter m.

I would argue then that the use of the word vara to denote social classes could be
motivated not only by the anthropological features of the people but also by thinking
about language as a social phenomenon. The social states are the varas, the sounds
of Prajpatis language which, reversing the order of creation presented in the Aitareya
B{hmaa (example [5]), can be expanded into words (bh, bhuva, sva) and texts
(gveda, Yajurveda, Smaveda) pronounced by human beings.

37 Cf. Smith 1994b: 288ff. It is the homology between social states, on the one hand, and three vyh{tis, or
metres, melodies (smans), three Vedas, on the other.
38 Gonda 1979: 129, Smith 1994b: 288ff., the usual order is reversed: Vaiyas come from the gveda and

Brahmins from the Smaveda.


39 Already in the gveda, cf. Renou 1958: 1415.
40 Thieme 1985: 562; Coward, Kunjunni Raja 2008: 4, passim.
41 Keith 1920: 256: practised fervour; having practised fervour.
42 Keith 1920: 256: brooded over.
86 JOANNA JUREWICZ

4. Sacrificial place as the space of discourse

Let us come back to the cosmogony of the Jaiminya Brhmaa (example [4]).
Prajpati expresses himself and the sounds that he pronounces become the language
(which consists of lauds, metres and melodies). In accordance with the basic assumption
of the ontic results of cognitive activity, the sounds become speakers, both divine and
human. As Smith shows, each social state has a corresponding groups of gods: Agni
and the Vasus are Brahmin deities; Vyu, Indra, and the Rudras are Katriya divinities;
and ditya, Srya, Varua, and the dityas are Vaiya gods (1994: 292). These are the
gods enumerated by the atapatha Brhmaa 6.1.2 and the Jaiminya Brhmaa quoted
above (examples [1] and [4]). I would argue then that in this way Prajpati creates space
in which his language can be realised. This space is sacrifice (yaja).43 Animals created
together with gods and human beings will be offered as sacrificial victims.
Gods inhabit the heavenly world (svarga), human beings the earth. Both gods and
human beings have language. But, as Malamoud (1996) shows, the language of gods is
different from the language of men. The language of gods is true in that it is identical
with thought and being. In the language of gods, the triad thinking saying becoming
is reduced to one. This is the language of Prajpati himself who can create the world
just because his thoughts expressed in language become elements of creation.
But when language and the world appear, the elements of the triad become separated.
And contrary to the language of gods, the language of human beings (even Sanskrit!) is
false44 in that words are not identical with thoughts that they express and things that they
name. The words agni, indra, etc. are used in the language of human beings, while agri,
indha, etc. in the language of gods.45 These words of human language do not express
thoughts properly and cannot create anything, because, as I have shown elsewhere, they
do not name the essence of things that is named in the language of gods.46
In everyday life, the language of gods, as of Prajpati himself, is unintelligible
(anirukta) and out of sight (paroka).47 However, it can be restored and activated in
sacrifice. In the sacrificial place human beings address gods, the gods are expected to
answer. The Vedic philosophers emphasise the emotional background of sacrifice, which is
the desire for obtaining various goods. If obtained thanks to sacrifice, the goods confirm
the truth of the words of human beings who ask for them, and of the gods who give the
goods. The priests in the sacrificial place express the sacrificers thoughts about future
sons, wives, rule, wealth, long and healthy life in imperfect human language. The gods
are expected to answer: to put these thoughts into words of their perfect language in
which thinking speaking becoming are the same. When they do, the sacrificers
thoughts become true in that their designates really appear. We could say that the Vedic
43 In atapatha Brhmaa 3.2.1.38 sacrifice itself is identified with language (vg vai yajo).
44 Cf. atapatha Brhmaa 6.3.1.34.
45 Malamoud 1996: 195206, Gonda 1955, Jurewicz forthc.
46 Jurewicz forthc.
47 Malamoud 1996: 196197.
WHAT DO ANCIENT INDIAN COSMOGONIES TELL US ABOUT LANGUAGE? 87

texts propose the correspondent theory of truth in its most literal dimension. True thought
not only corresponds to the actual state of affairs, but also actually becomes reality. But
this can happen only in the sacrificial space.
On the surface level, sacrifice is communication between human beings themselves
who have to cooperate in the sacrificial place. So, to enter the sacrificial space is to begin
to communicate with other human beings. But human beings are not the final respondents
of this communication. The gods and Prajpati himself are. This is the human point of
view. However, from the point of view of Prajpati, sacrifice is the space in which his
linguistic competence realises itself through his speaking manifestations. During creation
Prajpati manifests himself in sounds that become parts of the cosmos. He also manifests
himself in sounds that become gods and human beings endowed with languages. And
then, he begins to talk to himself in the imperfect language of human beings and the
perfect language of gods, and thus he confirms the truth of his words and the ontic
reality of his manifestations.48
It is worth noting that in the sacrificial space, human beings reverse the order of
creation. They begin with language in order to find thoughts expressed by it and reach
Prajpatis mind encoded in the language of gods in this way, while during creation
Prajpati first thinks and then speaks. This bi-directional understanding of communication
reflects everyday experience: when one is a speaker, one begins with thoughts in order
to express them in words, when one is a hearer, one begins with words to understand
thoughts conveyed by them. This was the reason of creation: Prajpati wanted to become
the speaker and hearer, not one but many. We should remember that the reason for
creation in the cosmogonies is always the desire of Prajpati to multiply himself. Why?
Because he wanted to have others to talk to.

5. Conclusion

The concept of language reconstructed in this paper on the basis of the Brhmaas
is similar to the contemporary one. Manifesting himself in language Prajpati could
externalise himself. But if Prajpati really wanted to talk and make his language really
meaningful, he had to create many speakers. Creation of the society is the next phase of
creation of language when it becomes the means of social communication. In the sacrificial
sphere of communication, Prajpati talks to himself via a multitude of divine and human

48 This concept of language seems to be motivated by another human characteristics, which is the ability to
play: in some descriptions the concept of the game of hide-and-seek can be evoked (e.g. atapatha Brhmaa 6.1.2,
example [2], which is later elaborated by B{hadrayaka Upaniad 1.4.1-5). In terms of the game of hide-and-seek,
the character of linguistic communication can be specified, which is looking for and finding the hidden meanings
of words. This concept of language motivates the understanding of the existence of the world as a process during
which God plays hide-and-seek with himself: he hides himself in the multitude of divine and human speakers and
he finds himself again when the truth of his words is confirmed by goods obtained in sacrifice. The full analysis
of this problem needs a separate analysis.
88 JOANNA JUREWICZ

speakers and confirms the truth of his words by goods obtained in sacrifice. We could
then reformulate Clifford Geertzs words quoted in the beginning of my paper: Human
thought is basically both social and public () its natural habitat is the sacrificial place.

Bibliography

Primary sources

Aitareya Brhmaa. 19851906. Ed. by Satyavrata Smaram. 4 vols. Bibliotheca Indica, no. 134.
Calcutta: Asiatic Society of Bengal.
Das Gopatha Brhmaa. 1919. Ed. By Dieuke Gaastra. Leiden: E.J. Brill.
Jaiminya Brhmana of the Smaveda. 1986. Ed. By Raghu Vira, Lokesh Candra. Delhi: Motilal
Banarsidass.
Tya Mahbrahmaa. 1869. Ed. by nandachandra Vedntavgia. 2 vols. Calcutta: New Sanskrit
Press.
Taittirya Brhmaa. 1898. Ed. by Nryaa str Goabole. 3 vols. nandrama Sanskrit Series,
no. 37.
The atapatha Brhmana in the Mdhyandina kh. 1924. Ed. by Albrecht Weber (red.). Leipzig:
Otto Harrassowitz.

Secondary sources

Bernrdez, E. 2005. Social cognition: variation, language, and culture in a cognitive linguistic
typology. In: de Mendoza, Ibez F.J.R.; Cervel, M.S.P. (eds.), pp. 191224.
Cienki, A. 2007. Frames, Idealized Cognitive Models, and Domains. In: Geeraerts, Cuyckens (eds.),
pp. 170187.
Covill, L. 2009. A Metaphorical Study of Saundarananda. Delhi: Motilal Banarsidass.
Coward, H.G.; Kunjunni, Raja K. 2008. The Philosophy of the Grammarians. Delhi: Motilal Banarsidass
(first edition 1990).
Eggeling, J. 1994. The atapatha-brhmaa According to the Mdhyandina School. 5 vols. Delhi:
Motilal Banarsidass (first edition 1900).
Evans, V.; Pourcel, S. (eds.). 2009. New Directions in Cognitive Linguistics. Amsterdam, Philadelphia:
John Benjamins Publishing Co.
DesCamp, M.T.; Sweetser, E. 2005. Metaphors for God: why and how do our choices matter for
humans? The application of contemporary cognitive linguistic research to the debate on God and metaphor.
Pastoral Psychology, 53, 3, pp. 207238.
Fairclough, N. 2001. Language and power. Harlow: Pearson Education Limited.
Geertz, C. 1973. The interpretation of cultures, New York: Basic Books.
Geeraerts, D. 2005. Lectal variation and empirical data in Cognitive Linguistics. In: de Mendoza
Ibez F.J.R.; Cervel, M.S.P. (eds.), pp. 163190.
Geeraerts, D.; Cuyckens, H. (eds.). 2007. The Oxford Handbook of Cognitive Linguistics. New York:
Oxford University Press.
Gonda, J. 1955. The etymologies in the ancient Indian brhmaas. Lingua 5, pp. 6183.
Gonda, J. 1976. Triads in the Veda, Amsterdam.
Gonda, J. 1983. The creator and his spirit (Manas and Prajpati). Wiener Zeitschrift fr die Kunde
Sd- und Ostasiens 27, pp. 542.
WHAT DO ANCIENT INDIAN COSMOGONIES TELL US ABOUT LANGUAGE? 89

Holdrege, B.A. 1994. Veda in the Brhmaas: Cosmogonic Paradigms and the Delimitation of Canon.
In: Patton, L.L. (ed.). Authority, Anxiety and Canon: Essays in Vedic Interpretation, Albany: State University
of New York, pp. 3566.
Holdrege, B.A. 1996. Veda and Torah. Transcending the Textuality of Scripture. New York: State
University of New York Press.
Jurewicz, J. 2006. Combing a horse and licking thinking. The metaphor in the gveda. Rocznik
Orientalistyczny 58, 2, pp. 132151.
Jurewicz, J. 2010. Fire and Cognition in the gveda. Warszawa: Dom Wydawniczy Elipsa.
Jurewicz, J. forthc. Etymologies in the Brhmaas and the gveda; A Case Study of Fires Precedence.
In: Parpola, A.; Fujii, M.; Insler, S. (eds.).
Keith, A.B. 1920. Rigveda Brahmanas. The Aitareya and Kautaki Brhmaas of the Rigveda.
Cambridge, Massachusetts: Harvard University Press.
Lakoff, G.; Johnson, M. 1980. Metaphors We Live By. Chicago: The University of Chicago Press.
Lvi, S. 1898. La doctrine du sacrifice dans les Brhmaas. Paris: Ernst Leroux.
Malamoud, Ch. 1996. Cooking the World. Ritual and Thought in Ancient India. Delhi: Oxford University
Press (translation of Cuire le Monde. Rite et Pense dans lInde ancienne 1989).
de Mendoza, Ibez F.J.R.; Cervel, M.S.P. (eds.). 2005. Internal Dynamics and Interdisciplinary
Interaction. Berlin, New York: Mouton de Gruyter.
Parpola, A.; Fujii, M.; Insler, S. (eds.) forthc. Vedic Investigations. Series Papers of the 12th World
Sanskrit Conference Held in Helsinki, Finland, 1318 July, 2003 (gen. eds. A. Parpola and P. Koskikallio),
vol. 1.
Patton, L.L. 1994. Authority, Anxiety, and Canon: Essays in Vedic Intepretation. New York: State
University of New York Press.
Putz, M. (ed.). 1992. Thirty Years of Linguistic Evolution. Studies in Honour of Ren Dirven on the
Occasion of his Sixtieth Birthday. Amsterdam, Philadelphia: John Benjamins Publishing Co.
Radden, G. 1992. The cognitive approach to natural language. In: Putz, M. (ed.), pp. 513541.
Radden G., Kvecses Z. 1999. Towards a Theory of Metonymy. In: Metonymy in Language
and Thought, Panther, K.; Radden, G. (eds.), AmsterdamPhiladelphia: John Benjamins Publishing Co.,
pp. 1759.
Renou, L. 1958. Etudes sur le vocabulaire du gveda. Pondichry: Institut Francais dIndologie.
Sinha, Ch. 2009. Language as a biocultural niche and social institution. In: Evans, V.; Pourcel,
S. (eds.), pp. 289309.
Smith, B.K. 1994a. The Veda and the Authority of Class. Reduplicating Structures of Veda and Vara
in Ancient Indian Texts. In: Patton, L.L. (ed.), pp. 6593.
Smith, B.K. 1994b. Classifying the Universe. The Ancient Indian Vara System and the Origins of
Caste. New York, Oxford: Oxford University Press.
Sweetser, E. 1990. From Etymology to Pragmatics. Metaphorical and Cultural Aspects of Semantic
Structure, Cambridge: Cambridge University Press.
Thieme, P. 1985. The first of the Triaptiyam (AV, 1.1~AV, P 1.6) and the beginnings of Sanskrit
linguistics. Journal of the American Oriental Philosophy 105, 3, pp. 559565.
R O C Z N I K O R I E N T A L I S T Y C Z N Y, T. LXV, Z. 1, 2012, (s. 90106)

KARNINA KOLLMAR-PAULENZ

The Invention of Shamanism


in 18th Century Mongolian Elite Discourse

Abstract

The paper concentrates on the transformation of the religious field in 17th and 18th
century Mongolia, focusing on the discourse formations which accompanied the missionary
strategies the Tibetan Buddhist monks employed to win over the Mongols to Buddhism.
By drawing on a variety of Mongolian sources, from biographies and chronicles up to
legal documents and terminological dictionaries, the author argues that the encounter
between shamanic and Buddhist religious specialists led to the creation of a religious
other and the reification of a bge-ner-n sasin, a teaching of the shamans, most
likely influencing European Enlightenment and post-Enlightenment constructions of so
called shamanism.

Keywords: Shamanism, Mongolia, Buddhism, Buddhist proselytism

Introduction

In recent years, as a consequence of the debate on Orientalism1 and in the wake of


Postcolonial Studies,2 Shamanism as the original religion of the peoples of Northern

1 Edward Said, Orientalism. Western Conceptions of the Orient, London 1978. Critical evaluations of Saids
thesis are numerous. His monolithic distinction between the West and the East has drawn sharp criticism, see,
among others, Jrgen Osterhammel, Wissen als Macht. Deutungen interkulturellen Nichtverstehens bei Tzvetan
Todorov und Edward Said. In: J. Osterhammel, Geschichtswissenschaft jenseits des Nationalstaates. Studien zu
Beziehungsgeschichte und Zivilisationsvergleich, Gttingen 2001, pp. 240265. For a critical evaluation of Saids
conceptual polarity of religous and secular culture see William D. Hart, Edward Said and the Religious Effects
of Culture, Cambridge 2000, pp. 6287.
2 Compare Robert J.C. Young, Postcolonialism. An historical introduction, Malden and Oxford 2010, pp. 337394.
THE INVENTION OF SHAMANISM IN 18TH CENTURY MONGOLIAN ELITE DISCOURSE 91

Eurasia and the Mongols has been deconstructed as a European invention, which has its
origins in the 18th century Enlightenment discourse. Scholars like Ronald Hutton3 and
Andrei Znamenski4 have analysed the historical and intellectual origins and respective
discourse formations of the European construction of Shamanism. Kocku von Stuckrad
has aptly summarized the present state of research concerning Shamanism: Of special
importance [] is the statement that Shamanism as a distinctive type of religion owes
its existence to the appropriation by western observers.5 The authors mentioned here
discuss the discursive field of Shamanism to be solely constructed on the basis of
European knowledge formations. According to them the distinctive type of religion
called Shamanism owes its existence to the hegemonic dominance of the European
protestant model of religion which in the wake of European imperialism and colonialism
has been exported and led to the standardization of indigenous religious and cultural
traditions after this model.
It is certainly true that early Mongolian sources before the late 16th century6 do
not name a religion or teaching called shamanism, but only mention the bge and
iduan, that is, certain specialists shamanizing. Since the late 16th century, however, with
the advent of Buddhism in the Mongolian regions, these descriptive modes changed,
and as early as the 17th century the texts start talking about a teaching or view of
the shamans. The assertion that Shamanism as a distinctive discourse formation and
in consequence as a distinctive type of religion is a European invention ignores emic
Asian discourses. Based on an analysis of the Mongolian elite discourses on religion
of the 17th to 19th centuries, which deal with so-called Shamanism, I argue that in
early modern Mongolia the transformation of the religious field on the discourse level
led to the reification of a teaching of the shamans which in turn probably influenced
the formation of the European discourse on Shamanism.

3 Shamans. Siberian Spirituality and the Western Imagination, Hambledon and London 2001.
4 The Beauty and the Primitive. Shamanism and the Western Imagination, Oxford 2007.
5 Schamanismus und Esoterik. Kultur- und wissenschaftsgeschichtliche Betrachtungen, Leuven 2003, p. 56

(my translation from the German).


6 Most notably the Secret History of the Mongols from the 13th century, see Paragraphs 181 and 272,

which contain the only occurrences of the word be (mong. bge). For the text of the Secret History see
Erich Haenisch, Manghol un niuca tobcaan (Yan-chao pi-shi). Die geheime Geschichte der Mongolen
aus der chinesischen Transkription (Ausgabe Ye Th-hui) im mongolischen Wortlaut wiederhergestellt.
Teil I: Text. Wiesbaden 1962. The best annotated translation gives Igor de Rachewiltz, The Secret History
of the Mongols. A Mongolian Epic Chronicle of the Thirteenth Century. Translated with a historical and
philological commentary. 2 vols. Leiden/Boston 2004. About the shamans in the Secret History see vol. 2,
pp. 652654.
92 KARNINA KOLLMAR-PAULENZ

I. The Advent of Tibetan Buddhism in the Mongolian Steppes

When in 1578 Altan Qaan of the Tmed Mongols and bSod nams rgya mtsho,7 the
then abbot of Bras spungs monastery, met at the temple of abiyal at Kke nor lake,
their meeting marked the beginning of the Buddhist conversion of the Mongol regions.8
Soon after the meeting of Altan Qaan and bSod nams rgya mtsho, Tibetan lamas began
to spread the Dharma among the different Mongolian tribes, and within a time span of
not much more than forty years the Mongols had taken over Tibetan Buddhist concepts
and practices.9 The reasons why the Mongolian nobility so willingly adopted Tibetan
Buddhism and pressured their subjects into doing likewise, can only be speculated about.
Scholars of Mongolian Studies often assume that Altan Qaan needed an ideological
backing for his position as the then most powerful ruler in the Mongolian territories,
and Tibetan Buddhism with its religio-political concept of universal ruler, cakravartin,
provided him with just such an ideology.10 Be that as it may, the subsequent measures
taken to spread Buddhism among the Mongols aptly demonstrate that the persons in
power, the rulers and the nobility, took an active interest in implementing the new religion
among their subjects. From the beginning the encounter was not so much a contest of
different world views, but a struggle about power between two opposing groups of
religious specialists, the male and female shamans and the lamas. The local rulers in
late 16th century Mongolia issued laws that prohibited shamanizing and the indigenous
religious practices that were connected with it. These practices included the worship of
the ongod,11 the powerful ancestor spirits and spiritual helpers of the shamans, both

7 For a biographical sketch of bSod nams rgya mtsho see Karnina Kollmar-Paulenz, The Third Dalai Lama

Snam Gyatso and the Fourth Dalai Lama Ynten Gyatso, in: Martin Brauen (ed.), The Dalai Lamas. A Visual
History. Chicago 2005, pp. 5259.
8 For a historical evaluation of this meeting in the context of the religio-political relations between Tibetans

and Mongolians see Karnina Kollmar-Paulenz, Erdeni tunumal neret sudur. Die Biographie des Altan qaan der
Tmed-Mongolen. Ein Beitrag zur Geschichte der religionspolitischen Beziehungen zwischen der Mongolei und
Tibet im ausgehenden 16. Jahrhundert. Wiesbaden 2001, pp. 112147.
9 Buddhism was well known among the Mongols since the 13th century and its practice was widespread

among the Mongolian elites during the Yuan dynasty (12711368). After the fall of the Yuan dynasty it retained
its presence in the Mongolian regions. Tibetan chronicles of the 15th and 16th centuries comment on the frequent
visits of Buddhist dignitaries to Mongolian nobles, see for example Gos lo tsa ba gZhon nu dpals report about
the journey of the 4th Zva dmar pa Cod pan dzin pa in the year 1470 to the Mongols and his preaching activities
there, in the Deb ther sngon po of 1478 (reprint, Chengdu 1984, p. 651). See also Henry Serruys, Early Lamaism
in Mongolia, Oriens Extremus 10 (1963), pp. 181216, and his Additional Note on the Origin of Lamaism in
Mongolia, Oriens Extremus 13 (1966), pp. 165173, as well as Klaus Sagaster, The History of Buddhism among
the Mongols, in: Ann Heirman/Stephan Peter Bumbacher (eds.), The Spread of Buddhism. Leiden/Boston 2007,
pp. 397399.
10 See, for example, Charles R. Bawden, The Modern History of the Mongols, London and New York 1968,

p. 30, and Liu Jin Se, Mongol-un quriyangui teke, Kkeqota 1998, p. 149.
11 Uno Harva, Die religisen Vorstellungen der altaischen Vlker, Helsinki (without year), 371ff., provides

detailed information about the Ongod, and Walther Heissig, Die Religionen der Mongolei. In: G. Tucci; W.Heissig,
Die Religionen Tibets und der Mongolei, Stuttgart 1970, pp. 312314, gives an overview about their function and
ritual use. The Mongolian shamanic chronicle ongod qara sakiusun teke sudur biig orosiba, written by an
THE INVENTION OF SHAMANISM IN 18TH CENTURY MONGOLIAN ELITE DISCOURSE 93

male and female, and blood sacrifices.12 Most other indigenous religious practices, for
example the obo cult,13 the cult of the mountain14 or the worship of the deity of the
hearth,15 were neither forbidden nor prosecuted.
The juridical measures taken to assert the preference of Buddhism focussed on the
select group of the bge, the male shamans, and the iduan, the female shamans. The
local rulers did not only prohibit the practice of shamanizing, but actively persecuted
the male and female shamans, as the Erdeni tunumal neret sudur, composed around
the year 160716 states:

After they had set on fire the outer17 ongod-images, they weakened and
eliminated the ecstatic and ignorant male and female shamans.18

unknown author and preserved in the Royal Library at Copenhagen (Mong. 41; I wish to thank research librarian
Bent Lerbk Pedersen from the Oriental and Judaica Collections for his help to make the text available to me),
deals extensively with the ongod and even offers an explanation for the origin of the ongod who according
to this source originally have been the souls of powerful male and female shamans. A part of the text is given
in Roman transcription in Walther Heissig, A Mongolian Source to the Lamaist Suppression of Shamanism,
Anthropos 48 (1953), pp. 501503.
12 Blood sacrifices were forbidden because they violate the first of the five Buddhist precepts which are binding

for lay-people as well as for monks and nuns. One of the most important topics of the Tibetan Buddhist discourse
on the cultural and religious other was the reproach of these being people who eat the flesh and drink the blood
of living beings, see Karnina Kollmar-Paulenz, Religionslos ist dieses Land: Das Mongolenbild der Tibeter,
Asiatische Studien/ Etudes asiatiques LIV/4, 2000, pp. 875905, and Uncivilized Nomads and Buddhist Clerics:
Tibetan Images of the Mongols in the 19th and 20th Centuries. In: Monica Esposito (ed.), Images of Tibet in the
19th and 20th Centuries, Vol. II, Paris 2008, pp. 707724.
13 A detailed description of the obo-cult, including both shamananic and Buddhist practices, give L. Qurabaatur/

. jm-e, Mongol-un bge mrgl-n tayil-a takil-a-yin soyol, 1991, pp. 263283. See also Charles R.
Bawden, Two Mongolian Texts Concerning Obo-worship, Oriens Extremus 5/1 (1958), pp. 5761.
14 Magdalena Tatr, Two Mongol Texts Concerning the Cult of the Mountains, Acta Orientalia Academiae

Scientiarum Hungaricae 30 (1), 1976, pp. 158.


15 For the cult of the fire and the hearth deities see Dominique Dumas, Aspekte und Wandlungen der Verehrung des

Herdfeuers bei den Mongolen. Eine Analyse der mongolischen Feuergebete, Bonn 1987, and Nima, Mongolud-un
al, Kkeqota 2003.
16 The Erdeni tunumal neret sudur, Stra called the jewel translucent one, a verse biography of Altan

Qaan of the Tmed, is one of the earliest known Mongolian historiographical sources. There exists only
one manuscript of the chronicle that is preserved in the library of the Institute for History and Literature of
the Inner Mongolian Academy of Social Sciences. I use a Xerox copy of the original manuscript. The text is
made available in a modern Uiguro-Mongolian edition alongside the facsimile in Jorung-a, Erdeni tunumal
neret sudur orosiba, Beijing 1984. In the meantime translations into Japanese (T. Morikawa, Study of the
Biography of Altan Khan, Kyushu University, Fukuoka 1987), Chinese (Zhu rong ga, A-le-tan han zhuan,
Hohot 1991), German (Karnina Kollmar-Paulenz, Erdeni tunumal 2001) and English (Johan Elverskog, The
Jewel Translucent Stra. Altan Khan and the Mongols in the Sixteenth Century, Leiden 2003) have been
published.
17 The term outer points to the Tibetan differentiation between inner ones (tib. nang pa), i.e. Buddhists,

and outer ones (tib. phyi pa), i.e. Non-Buddhists.


18 Fol. 29r12-15.
94 KARNINA KOLLMAR-PAULENZ

The burning of the ongod is reported in Mongolian,19 and also Tibetan,20 sources.
Besides trying to stop the activities of the shamans by these purges and by law, the
secular authorities provided material incentives to employ Buddhist rituals, thereby
undermining the power of the shamans. Vivid descriptions of such incentives are given
in the indamani erike,21 the biography of the famous monk Neyii Toyin, a Torud
Mongol from Western Mongolia who in the first half of the 17th century was mainly
active in Eastern Mongolia, among the Ongniud and Qorin Mongols. But not only the
secular authorities, also the monks themselves took to bribery in order to win the people
over, as the indamani erike reports:

From all different kinds of jewels, gold and silver, and from different
kinds of things like silk and sable, which had been presented [to him] by
many donors, princes as well as great and petty nobles, [Neyii Toyin]
gave in all directions and everywhere one ounce of gold to everybody who
could memorize the Yamandaga, and one ounce of gold and silver each
to everybody who could memorize the sdhana and the bja-mantra of
Guhyasamja, and to the poor people, according to their wishes, goods and
cattle. Thereupon there were generally very many [people] who memorized
the Yamandaga and the Guhyasamja.22

This look at the juridical measures and the economic incentives gives us a vital
clue to the nature of the initial interaction between the indigenous religious specialists
and the newly arrived Tibetan lamas and monks. The interaction was very much
characterised by economic competition for the limited material resources of the
lay population available to both groups. In this competition the Tibetans were on

19 The most dramatic report is given in the Boda neyii toyin dalai majusryi-yin domo-i todorqai-a

geyiglgi indamani erike kemegdek orosiba, the biography of the monk Neyii Toyin, written in 1739 by
Prajsagara. I have at my disposal a copy of the xylograph preserved in the Royal Library at Copenhagen
(Mong. 506), and wish to thank Dr. Helmut Eimer for his help in obtaining the copy. The passage is found on
fol. 54r5-13. A German translation is given by Walther Heissig, Neyii Toyin. Das Leben eines lamaistischen
Mnches (15571653). Aus seiner mongolischen Biographie bersetzt und mit einer Einleitung, in Sinologica, III,
(1953), 42, and Karnina Kollmar-Paulenz, Der Buddhismus als Garant von Frieden und Ruhe. Zu religisen
Legitimationsstrategien von Gewalt am Beispiel der tibetisch-buddhistischen Missionierung der Mongolei im spten
16. Jahrhundert, Zeitschrift fr Religionswissenschaft, 11, 2003, p. 186. See also Isibaldan, Erdeni-yin erike (1835),
fol. 28v8 (Walther Heissig (ed.), Erdeni-yin erike. Mongolische Chronik der lamaistischen Klosterbauten der
Mongolei von Isibaldan (1835). In Faksimile mit Einleitung und Namensverzeichnis herausgegeben, Kopenhagen
1961).
20 For example in the biography of the 3rd Dalai Lama, rJe btsun thams cad mkhyen pa bsod nams rgya mtshoi

rnam thar dngos grub rgya mtshoi shing rta zhes bya ba bzhugs so, composed by the 5th Dalai Lama, reproduced
in Phags pa jig rten dbang phyug gi rnam sprul rim byon gyi khrungs rabs deb ther nor bui phreng ba, Vol.2,
Dharamsala 1984, p. 150, 25.
21 See note 19.
22 indamani erike, fol. 74r16-74v5.
THE INVENTION OF SHAMANISM IN 18TH CENTURY MONGOLIAN ELITE DISCOURSE 95

the winning side, because they had been able to secure the support of the secular
authorities.

II. Tibetan Buddhist Strategies to win over the Mongols

However, juridical measures and economic incentives alone do not explain the rapid
success of Tibetan Buddhism among the Mongols. The success is due to the specific
missionary strategies, employed by the Tibetan Buddhist monks, which directly affected
religious practices. What did it mean to spread the dharma (nom-i delgerek)? It first
and foremost meant to practice Buddhist rituals, to perform the dharma. The monks
strategies proved to be successful because the Mongols and the Tibetans shared a similar
socio-religious habitus in a common religious field.23 Conditioned by society and history,
the habitus is constituted by individual and collective experiences, and in this sense
it may be characterized as embodied history. Social meaning is tied to the human
body, and the body in turn is shaped by the habitual schemes which are acted out
implicitly, on an unconscious level.24 By performing the dharma the Buddhist monks
were able gradually to incorporate Buddhist world concepts into indigenous ones, which
thus infiltrated the indigenous religious field and became an essential part of Mongolian
religious life. This was achieved mainly by ritual performance, which very much focused
on bodily performed patterns of behaviour. This focus is mirrored in some of the terms
applied to Buddhism and Shamanism in our sources. In Mongolian texts from the 17th
century onwards the terms (among others, as we shall see) asin mrgl for Buddhism
and bge mrgl25 for Shamanism were used. Mongolian mrgl literally denotes
the act of bowing, stressing the bodily performance of venerating the Dharma or,
respectively, the shamans. Both terms emphasize the visible and performative aspects
of the Mongolian discursive and embodied construction of religion. In early modern
Mongolian societies the acting out of embodied socio-religious norms and roles proved to
be of crucial importance. Therefore, rather than bluntly attempting to exchange existing
religious concepts and practices, the Buddhist monks put considerable effort into giving
new or additional meaning to the bodily engrained patterns of behaviour, thereby slowly
transforming these practices and rituals.

23 According to Pierre Bourdieu, forms of habitus are systems of enduring dispositions, structured structures,
which are suitable to operate as structuring structures, in other words: as generating and structuring principles of
practice forms and representations (Pierre Bourdieu, Entwurf einer Theorie der Praxis auf der ethnologischen
Grundlage der kabylischen Gesellschaft, Frankfurt am Main 1976, p. 165; my own translation).
24 This is further elaborated by Pierre Bourdieu in his Sozialer Sinn. Kritik der theoretischen Vernunft, Frankfurt

am Main 1987, especially in the chapter Glaube und Leib (Belief and Body), pp. 122146.
25 See for example Qori kiged auyin buriyad-nar-un urida-daan bolusan anu, composed by Tuultur Toboev

in 1863, p. 17.
96 KARNINA KOLLMAR-PAULENZ

III. Mongolian Buddhist self-descriptions: The terms nom and asin


in 16th/17th century Mongolian sources

The competitive encounter between the newly arrived Tibetan Buddhist monks and
the Mongolian religious specialists, the male and female shamans, gradually led to the
categorisation of the religious opponents on the discourse level. As is well known, the
shamans did not adhere to a world-view different from the community they belonged to.
Moreover the shamanizing practices and rituals were only part of a multitude of traditional
Mongolian religious concepts and practices. Ritual practices like the mountain cult, the
fire cult and the veneration of the hearth deity, the cult of Caan ebgen, Geser Qan
etc., as well as the various groups of gods, demons and spirits believed to enliven the
world played an important role in everyday religious life. These practices and beliefs
were not persecuted, but gradually transformed or simply incorporated into Buddhist
practices and beliefs.
The nearly complete illiteracy of the male and female shamans gave the
Buddhist monks an advantage on the discourse level. They brought with them a well
differentiated analytical terminology, that had been developed in inner Buddhist polemical
debates and which provided a terminological instrument that could also be used
comparatively. In the Mongolian Buddhist texts of the time a standardized terminology
(in part going as far back as to the early Uighur translations of Buddhist texts) is
used regarding Buddhism. Two terms are omnipresent, nom (from Greek nomos) and
asin/sasin (from Sanskrit sana). Both translate Tibetan chos and bstan pa.26 They
are often used together with the attributive burqan-u, of the Buddha: burqan-u nom/
burqan-u asin, Dharma of the Buddha, teaching of the Buddha. Apart from these
meanings, in the Erdeni-yin tobi from 1662 the term nom is also used in the meaning
of rule, norm, and in this connotation we find the term for the two great rules of
the spiritual and secular orders: burqan-u nom and kmn- nom27 or sansar-un nom.
Although the earlier aan teke also uses asin to denote the two rules, in this text28

26 Whereas the early Ganjur translations and early chronicles often indiscriminately translate chos and bstan pa
as nom or asin, in the 18th century a standardisation process sets in. According to the Tibeto-Mongolian dictionary
Dag yig mkhas pai byung gnas/ Merged arqu-yin oron, compiled in 1741/42 by a translation committee under
the direction of the 2nd lCang skya Qututu Rol pai rdo je, bstan pa is to be translated by asin, chos by nom,
see chapter ka, fol. 6v2, 7r1, 10r2. I used the Beijing block print in the possession of Professor Richard Ernst to
whom I convey my sincere gratitude.
27 This latter term probably translates the Tibetan mi chos.
28 See Arban buyantu nom-un aan teke (I consulted the facsimile reproduction of a manuscript in Walther

Heissig, Die Familien- und Kirchengeschichtsschreibung der Mongolen. I. 16.18. Jahrhundert. Wiesbaden 1959,
pp. 124), where nom is used in a variety of combinations, for example nom-un ejen (3v3/4, 6v7), nom-un jasa,
nom-un tr (6v3, 7r7), nom-un yosun (6v5), nom-un jerge (7r12/13), nen nom (15v1), nen nom-un jasa (20r1).
Nom is also used in concrete sense as teaching, in the phrase qamu nom-ud (fol. 7r2), all teachings. In contrast
to the ubiquitous use of nom in its various meanings, ranging from rule, norm over teaching to religion,
THE INVENTION OF SHAMANISM IN 18TH CENTURY MONGOLIAN ELITE DISCOURSE 97

and the Erdeni-yin tobi29 the use of the term nom (in the narrow sense of dharma)
prevails. Interestingly, one of our oldest sources, the Erdeni tunumal neret sudur, mostly
uses sasin to denote Buddhism (either simply called sasin or often also burqan-u sasin),30
whereas nom is mostly used in its meaning as either scripture/book,31 dharma,32
in the triad of Buddha, Dharma, Sangha,33 or in composite terms like nom-un qaan,
Dharmarja.34 The Altan tobi anonymus35 speaks of ajin, denoting Buddhism, and
sometimes also of ongkaba-yin sajin,36 as also the slightly later Asarai neret-yin
teke does.37
Taking into account the colophons of the hand-written Ganjur preserved in St.
Petersburg,38 in these early translations dating back to 1628/29, but partly drawing on
even earlier translations, we mostly find the term asin and the combination asin nom.39
Nom alone is sparingly used.

i.e. Buddhism, asin in the aan teke is only used sparingly, either denoting Buddhism (see for example fol.
3v3: deged asin) or in the term qoyar asin (see fol. 15v4 and 15v6) the two teachings of the spiritual and
temporal orders. For a translation and thorough investigation of this important source see Klaus Sagaster, Die
Weisse Geschichte (aan teke). Eine mongolische Quelle zur Lehre von den Beiden Ordnungen Religion und
Staat in Tibet und der Mongolei, Wiesbaden 1976.
29 In the Erdeni-yin tobi the term nom is used 42 times against asin which is used 24 times. I used the

Urga-manuscript, see Erich Haenisch, Eine Urga-Handschrift des mongolischen Geschichtswerks von Secen Sagang
(alias Sanang Secen), Berlin 1955.
30 For example in fol. 20v2, 20v8/9, 20v18, 21v11, 21v15/16, 21v23, 22r1, 22v22, 25r10/11, 25v5, 25v18.

Only once burqan nom (fol. 23r20) is used.


31 In fol. 2r22, 30v8, 31v8.
32 For example fol. 21r2, 23r20.
33 For example fol. 17v13.
34 For example fol. 29r10.
35 Charles R. Bawden, The Mongol Chronicle Altan tobi. Text, translation and critical notes. Wiesbaden 1955,

p. 13.
36 See Qad-un ndsn quriyangui altan tobi, Kkeqota 1980, p. 121 and p. 122.
37 Written by the Qalqa noble Byamba erke dayiing in 1677. For the relevant text passages see Pringlai (ed.),

Byamba. Asarai neret-yin teke, Ulan-Bator 1960, pp. 67ff. There also exists an Inner-Mongolian edition
of this chronicle with an extensive commentary, see B. Baan-a (ed.), Asarai neret-yin teke, Beijing 1984.
A description and translation of the chronicle provides Hans-Rainer Kmpfe, Das Asarai neret-yin teke
des Byamba Erke Daiing alias amba Jasa. (Eine mongolische Chronik des 17. Jahrhunderts). Wiesbaden
1983.
38 Zoya K. Kasyanenko, Katalog peterburgskogo rukopisnogo Ganzhura. Sostavlenie, vvedenie, transliteraciya

i ukazateli. Moskva 1993.


39 For example in the colophon of the Bilig-n inadu krgsen jaun mingan to-a-tu, see Kasyanenko,

Katalog, No. 524, 136138. The redaction committee of the Kangxi era who prepared the print edition of 17181720
did not change the terminology of either asin or nom, compare the same colophon in Louis Ligeti, Catalogue du
Kanjur mongol imprim, Vol. 1, Catalogue, Budapest 1942, No. 746, pp. 166168. Compare also the colophons
of Nos. 510, 539, 545, 599, 616, 669, 727 in the handwritten Ganjur.
98 KARNINA KOLLMAR-PAULENZ

The 18th and 19th century sources tend to favour the term sasin/asin, as an
exemplification of the terminology used in the Altan krdn mingan kegest biig,40
the Bolor erike41 and the Bolor toli42 brought about.

IV. Speaking about the religious Other

Mongolian sources of the 17th 19th centuries rarely deal with the shamanic adversaries
of the Buddhist monks.43 Still, in the few texts that mention the shamans and their activities
we note the formation of a discourse about the true teaching, namely Buddhism, and
the wrong view/doctrine, which the shamans possess. In Mongolian Buddhist writings
of the time the activities of the shamans were gradually turned into a homogenous entity
that first received the standard appellation buruu jel, wrong view, or buruu nom,
wrong dharma. In a text fragment about the turning back of bad omens found in
Xarbuxyn Balgas and dating to the early 17th century,44 I found the first mention of the
appellation buruu jel-t bge [idu]45an, the male and female shamans possessing
awrong view. This term is often used in opposition to burqan-u asin, the teaching of
the Buddha, as can be noticed in the statement: In this way the wrong view was brought
to an end and the teaching of the Buddha turned out to be immaculate. (indamani erike,
fol. 54r11-13). Our source, however, is vague about what particular doctrines, practices
and rituals the term buruu jel refers to. Buruu jel translates the Tibetan term lta

40 This chronicle was composed in 1739 by Sireget guosi dharma, see Heissig, Familien- und
Kirchengeschichtsschreibung, I, pp. 134159. I used the manuscript in Walther Heissig (ed.), Altan krdn
mingan gegest biig. Eine mongolische Chronik von Sireget Guosi Dharma (1739). Herausgegeben und
mit Einleitung und Namensverzeichnis versehen. Kopenhagen 1958. Sireget guosi dharma almost exclusively
writes asin, see III, fol. 4r9 and 10, IV, fol. 14v5 and 6, 20v1, or V, fol. 2v8 and 13r2. Buddhism in Mongolia
is further often specified as bConggaba-yin asin (in the 4th and 5th blg), demonstrating the dominant dGe lugs
pa-discourse.
41 Written by the nobleman Rasipungsu in 1774/75, see the monograph by Walther Heissig, Bolur Erike

Eine Kette aus Bergkristallen. Eine mongolische Chronik der Kienlung-Zeit von Rasipungsu (1774/75).
Literaturhistorisch untersucht. Peiping 1946, and Walther Heissig, Familien- und Kirchengeschichtsschreibung,
op. cit., pp. 198200. I consulted the edition of Mostaert/Cleaves, see Antoine Mostaert/ Francis W. Cleaves (eds.),
Bolor Erike. Mongolian Chronicle by Rasipungsu. 5 vols, Cambridge/Mass., 1959. Rasipungsu mostly speaks
of burqan-u ain, bCongkaba-yin deged ain, or simply asin/ sajin, for example book V, p. 684,7.
42 Composed between 1834 and 1837 by the Buddhist monk Jimbadorji, see Walther Heissig (ed.), Bolur Toli

Spiegel aus Bergkristall von Jimbadorji (18341837). Buch III: Geschichte der Mongolen, Kopenhagen 1962.
Jimbadorji treats the conversion of the Mongols to Buddhism similiarly to Saang Seen (Urga-Ms., fol. 76v14ff.),
compare fol. 54-58 of Book III.
43 This fact was already noted by Walther Heissig, A Mongolian Source to the Lamaist Suppression of

Shamanism, Anthropos, 48, 1953, p. 518.


44 Text XBM 150, published in Elisabetta Chiodo, The Mongolian Manuscripts on Birch Bark from Xarbuxyn

Balgas in the Collection of the Mongolian Academy of Sciences, Part 2, Wiesbaden 2009, 182.
45 Chiodo, 2009, op. cit., Manuscripts, p. 182, note 11, suggests to read iduan for the missing part of the

word. The variant udaan, of course, is also possible. The indamani erike, fol. 46r26/27, for example, reads
buruu jel-t bge udaan.
THE INVENTION OF SHAMANISM IN 18TH CENTURY MONGOLIAN ELITE DISCOURSE 99

log and in this way is closely connected to the Tibetan-Buddhist polemical discourse.
Tibetan lta log, respectively chos log, is used in Buddhist polemics to denounce the
doctrinal adversaries. The terms chos and chos log are closely related to each other and
are commonly applied in Inner-Buddhist discourse. Apart from buruu jel we also
find the terms buruu nom46 and qaucin ba buruu jel.47 The buruu jel-t bge
iduan are also described as adherents of the adaadu ongod, the outer ongod.
As mentioned before, Mongolian adaadu translates the Tibetan phyi pa, outer, in
opposition to nang pa, inner, denoting somebody who is positioned within the Dharma,
e.g. a Buddhist.48 The outer inner dichotomy refers to collective social identities in
the respective societies.
The use of the terms nom, asin and buruu jel (translating the respective Tibetan
terms) attest to an Inner-Buddhist polemic discourse about the religious other. Their
first appearance on the one hand illustrates the reifying processes in the Buddhist
intellectual discourse of the time, on the other hand they testify to the fact that the idea
of aplurality of different religious teachings during that period was not yet present. The
practices of the shamans were looked upon from a normative and exclusivist Buddhist
viewpoint. Religion in this intellectual context was synonymous to Buddhism, and
the shamanic practices were judged to be wrong from the only standpoint imaginable,
they were Buddhism turned wrong.

V. Reifying the religious opponents

The collective term buruu jel testifies to the fact that in the course of the Buddhist
encounter with the Mongolian shamans the loosely connected and extremely localized
shamanic ritual practices which centred on the male and female shamans as ritual actors
were viewed as a single system and thus reified. Further evidence of this development is
provided by the employment of the term qara asin, black teaching, which derives from
the self-descriptive sira asin, yellow teaching, as the dGe lugs pa-tradition described
itself in the Mongol regions. Already in the colophon of an early Mongolian translation
of the Dzangs blun49,which was not included in the Ganjur, the term qara jg occurs.
The term is also used in the colophon of the o-tu/ulaan/erlig-n dayisan-u qaan

46 See Bolor erike, book V, 683,10, and Erdeni-yin erike, fol. 28r11.
47 indamani erike, fol. 13v24-25.
48 The term dotoadu nom-tan-u jel, view of the adherents to the inner teaching, is also found in Saang

Seens Erdeni-yin tobi, see Urga-Ms., fol. 34r19-21.


49 Commonly known under the title liger-n dalai, as it is called in the printed Ganjur-edition (Ligeti,

Catalogue, No. 1103). The original title of Sireget gsi orjis translation was Siluun onul-tu kemegdek sudur,
see Kasyanenko, Katalog, No. 839. The translation referred to here, however, was prepared by Toyin guosi under
the title Siluun budaun ye onoqui neret sudur sometime between 1578 and 1612. The colophon is given in
Roman transcription with a German translation by Walther Heissig, Toyin guosi ~ Guisi alias otu guisi: Versuch
einer Identifizierung, Zentralasiatische Studien 9, 1975, pp. 361446, especially pp. 391408. For qara jg see
p. 398.
100 KARNINA KOLLMAR-PAULENZ

neret included in the hand-written Ganjur, but not in the printed edition.50 The use of
these diverse terms marks the birth-hour of the invention of Shamanism as a reified
system in 18th century Mongolia.
Qara jg as well as qara asin were at first clearly derogatory terms used by the Tibetan
Buddhists to belittle their religious opponents. They were dependent on the juxtaposition
sira qara, and still mirror the Buddhist discourse on shamans as heretics. Interestingly,
very soon the shamans used the polemic term qara asin as a self-descriptive term.
Further evidence of the ongoing reification process can be found in the first law code
for the Mongols to be commissioned by the Qing which was published sometime after
1694.51 This law code of all in all 152 articles contains one article on the community of
the Buddhist lamas and the community of the male and female shamans, entitled lam-
a-nar-yin ayima. Bge iduan-u ayima (Fol. 39v-41r). The article gives instructions
how to deal with lamas or shamans who do not follow the prescribed rules (yosun) of
their respective communities.52 The Qing administration thus dealt with the indigenous
religious specialists of the Mongols in the same way as they dealt with the Tibetan-
Buddhist lamas. They acknowledged both groups as juridical bodies.
The article in the Kangxi law code reveals the extent of the reification process the
indigenous Mongolian religious practices centered on the shamans had undergone already
at the end of the 17th century. Furthermore, the use of asin and also surtaun53 together
with a defining attribute led to the development from an exclusivist to a comparative
terminology. From the 18th century onwards asin was no longer used exclusively to denote
Buddhism, but emerged as a comparative term, signifying concepts which were deemed to
be equivalent to the Buddhist teaching and thus denoting religion as a general concept.
Our sources now speak of bge-ner-n asin, the teaching of the shamans, to denote
the practices (and concepts) of the male and female shamans.54 Still later examples of
the comparative use of sasin are keristos-n asin55 for Christianity or Lalu56-yin asin
for Islam. Still, however, other distinctions, also purely Buddhist ones, like dotoadu
asin and adaadu asin, were used alongside this emerging comparative terminology.
50 See Kasyanenko, Katalog, No. 106, pp. 5152.
51 This collection of individual laws is kept in the State Library in Ulan Bator under the title adaadu
mongol-un tr-yin jasaqu yabudal-un yamun-un engke amuulang-un y-e-d 1693 on-du keblegsen, dotor-a
1629 on-aa ekileged uda-a dara-a qaad-un y-e-d jarlasan aaja-du-i jasamjilasan mongol-un aajan-u
biig, compare Dorothea Heuschert, Die Gesetzgebung der Qing fr die Mongolen im 17. Jahrhundert anhand des
Mongolischen Gesetzbuches aus der Kangxi-Zeit (16621722), Wiesbaden 1998, 47. As Heuschert points out, the
publication date 1693 mentioned in the title cannot be correct, because the last article of the collection dates from
the year 1694.
52 Heuschert, Gesetzgebung, 136. The Mongolian text is given on pp. 215216, accompanied by a German

translation.
53 Ongod qar-a sakius-un teke sudur biig orosiba, p. 4.
54 For example, to quote just one chronicle, in the Qori-yin arban nigen eige-yin jun-u u ijaur-un tuuji,

p. 92ff. The chronicle was composed in 1875 by Vandan Yumsunov.


55 Thus in Qori kiged auyin buriyad-nar-un urida-daan bolusan anu, p. 15.
56 From the Tibetan kla klo, which translates Sanskrit mleccha. In early Tibetan Buddhist texts like the Klacakra-

tantrarja and the Vimalaprabh the term denotes Muslims.


THE INVENTION OF SHAMANISM IN 18TH CENTURY MONGOLIAN ELITE DISCOURSE 101

VI. The teaching of the shamans

The terms jel and nom suggest a bundle of concepts, especially when used in binary
opposition to burqan-u asin. Yet none of the texts from the 17th and 18th centuries elaborate
which concepts or which practices can be subsumed under the term buruu jel/ adaadu
nom, or even bge-ner-n asin. In the end these labels are decidedly unspecific as to
their contents. From the texts we only get to know that the wrong view of the male
and female shamans denotes the actors of this wrong view and their activities. They
act mainly as healers and as exorcists, and the ongod, the helpers of the shamans, are
likewise often mentioned. The first texts to give a definition of a shaman and an overview
of shamanic practices date to the second half of the 19th century. The Buryad-Mongolian
chronicles Qori-yin arban nigen eige-yin jun-u u ijaur-un tuuji and Qori kiged
auyin buriyad-nar-un urida-daan bolusan anu provide us with asystematic overview
of shamanic practices.57 They enumerate healing techniques, including the prevention of
death, and exorcism of evil spirits (qourlaju cidkr) as characteristic activities. In the
latter case the shaman bans the cidkr into a substitute (joli) which is burnt, or into
a living animal which is slaughtered. Furthermore, according to these sources shamans
are adept in divination from the shoulder blades of a sheep. This divination method is
already well attested in the 13th century when the Franciscan brothers at the court of the
great Qans at Karakorum reported this custom. Expulsion (araul) of an evil spirit who
is responsible for harm to ones material possessions (ed tavar) is also performed by the
shaman. Finally, the male and female shamans invoke spells, blessings and formulas to
protect the family, the children, the cattle etc. against evil spirits and contagious diseases
and to bring blessings and luck to them.
The systematic enumeration and description of different religious doctrines and practices
is well known in Tibetan Buddhist doxographical literature which had its heyday in the
19th century. Therefore the definitions and detailed descriptions of shamanic concepts
and activities point to a Tibetan Buddhist influence. The religion of the shamans is
treated in these chronicles as the Bon-religion is treated in the Tibetan doxographies.58
The Buryad chronicles, however, also show Russian influence.

57 Both chronicles devote whole chapters to the description of the bge-ner-n mrgl, see Qori kiged auyin

buriyad-nar-un urida-daan bolusan anu, pp. 1621, and Qori-yin arban nigen eige-yin jun-u u ijaur-un
tuuji, pp. 5868.
58 Compare the Grub mtha shel gyi me lon, composed in 1802 by the Mongolian author Thuu bkvan Blo bzang

chos kyi nyi ma, which devotes chapter nine to the Bon-religion and chapter ten to the philosophical systems of
China. For a description of this work see Andrei I. Vostrikov, Tibetskaya istoricheskaya literatura, Moskva 1962,
pp. 97100. This work was well known in Mongolia, and chapter twelve, dealing with the history of Buddhism
in Mongolia, was translated into Mongolian. In Buryatia a block print under the title Deged asin erdeni ber
mongol oron-i tgeglgsen u-i jeglgsen irau kelen- krkirel neret was prepared, which led to its wide
circulation, compare Boris Ya. Vladimircov, Nadpisi na skalakh khalkhaskogo Coktu-taidzhi, I. Izvestiya Akademii
Nauk SSSR, 1926, pp. 12721273.
102 KARNINA KOLLMAR-PAULENZ

VII. Emic discourses and the shamanism of the early ethnographers

In 1846 the Buryad scholar Dorji Banzarov published a work entitled The Black
Faith or Shamanism among the Mongols,59 thus introducing the emic term qara asin
to a wider scholarly audience. Dorji Banzarov was the first scholar to write a monograph
about the so-called shamanism, although in the early 19th century shamanism already
was a well established construct in European intellectual discourse. When tracing the
origin of shamanism back to the external world, nature, and the internal world, the
soul of man, Banzarov was deeply influenced by the European Romantic movement. But
his explanation of shamanism as arising out of mans attachment to the surrounding
landscape is also found in the Buryad-Mongolian chronicles of the time. According to
the already quoted chronicle Qori-yin arban nigen ecige-yin jun-u u ijaur-un tuuji,
the shamanic powers originate in the transformation of the souls of the male and female
shamans into the masters, ongod and demons of these mountains, streams, brooks,
lakes and forests etc. (p. 95).
In the debate about the European construction of shamanism, the question whether
the European discursive construct could have been influenced by Inner Asian discourses
has not yet been raised. Taking a closer look at the ethnographic accounts from which
the European thinkers of the Enlightenment and post-Enlightenment romantic renaissance
drew their knowledge about native religiosity we should reconsider our assumptions
about the European origins of the reification of shamanism. Ethnographers like Johan
Gottlieb Georgi show in their reports that they were acquainted with the resentment the
Buddhist monks nurtured against the shamans. Peter Simon Pallas even mentions the
persecution of the shamans:

They still have secret sorcerers of both sexes among themselves,


who, despite being persecuted and cursed by the lamas, trifle with the
superstitious mob.60

The Russian and German ethnographers who travelled among the Mongols in the
late 18th and early 19th centuries have been well aware of the Buddhist attitude towards
the shamans and their treatment as a homogenous religious group and teaching opposed
to Buddhism. Moreover, they also took over the Buddhist self-perception of having
a civilizing influence on the barbarian natives, as the following statement by Pallas
shows:

59 Dorji Banzarov, Chernaya vera ili shamanstvo u Mongolov, Kazan 1846.


60 Peter Simon Pallas, Sammlungen historischer Nachrichten ber die mongolischen Vlkerschaften, 2 Theile.
Um eine Einfhrung vermehrter Nachdruck der 1776 und 1801 bei der Kaiserlichen Akademie der Wissenschaften
in St. Petersburg erschienenen Ausgabe. Mit einer Einfhrung von S. Hummel, Graz 1980, pp. 341. The English
translation is my own.
THE INVENTION OF SHAMANISM IN 18TH CENTURY MONGOLIAN ELITE DISCOURSE 103

Still one notices among the Kalmucks and more so among the Mongols
who confess to the lamaist religion, that their customs, partly through
the community with the Chinese, partly through the Tangut clergy, have
become immeasurably softer than those one even now finds among the
Burts who adhere to the shamanic superstition and who are, as it were,
the exact likeness of that which their brothers used to be.61

VIII. Conclusion

To sum up: A close reading of the available Mongolian sources of the 17th to 19th
centuries reveals an ongoing reification process of a teaching of the shamans. This
process unfolds in several steps. First, the Mongolian indigenous religious specialists
were described as possessing a wrong view, compared to the true Buddhist teaching.
The discourse on the shamanic adversaries of the Buddhist monks was thus situated
within the well known Inner-Buddhist culture of polemic debate. In later texts we note
an increasing use of objectifying vocabulary to characterise the shamans. They are
no longer those who possess an old and wrong [world]view, but are described as
adhering to a teaching different from Buddhism, without judging this to be inferior.
The emergence of a neutral descriptive mode that described different religious teachings
with a comparative terminology also brought about the need to define and classify these
teachings. The Buryad-Mongolian chronicles of the late 19th century testify to this. The
reification processes described here should, however, not be understood as a singular
and monolithic discourse. Even as late as the 19th century derogatory terms to denote
the shamans were still in use. The discourse on the teaching of the shamans is thus
a polyphonic discourse.
The analysis of the Buddhist way into Mongolia leads us to some general conclusions.
The emergence of an analytical vocabulary to categorize the religious opponents
demonstrates that religions as entities often emerge out of confrontational situations
and in their wake of polemics, and that religious practices and concepts are given names
and treated as reified entities when cultural boundaries are crossed and rival traditions are
encountered. Abstract terms that bundle up practices, rituals, ideas, concepts and persons in
one teaching are created in situations, when differences are encountered and negotiated.
An encounter situation within a social-religious field can thus lead to the distinction of an
autonomous field of religion on the discourse level and the development of a respective
analytical terminology. As a result of our examination of the Mongolian sources from
the late 16th up to the 19th century we can sum up that a religion called shamanism
was invented in 17th/18th century Mongolia. What implications does this result have on
61 Pallas, Sammlungen, pp. 103. See also Karnina Kollmar-Paulenz, Zur europischen Rezeption der
mongolischen autochthonen Religion und des Buddhismus in der Mongolei. In: Peter Schalk et al. (eds.). Religion
im Spiegelkabinett. Asiatische Religionsgeschichte im Spannungsfeld zwischen Orientalismus und Okzidentalismus,
Uppsala 2003, pp. 243288, especially pp. 267275.
104 KARNINA KOLLMAR-PAULENZ

the European discourse on shamanism? The assumption of Kocku von Stuckrad and
other scholars that Shamanism as a distinct type of religion owes its existence to the
appropriation by western observers needs to be modified. In the light of the Mongolian
evidence we can reformulate Stuckrads statement thus: Shamanism as a distinctive type
of religion owes its existence to the appropriation by Buddhist observers.

Bibliography

Primary Sources

Anonymus. Erdeni tunumal neret sudur. Xerox-copy of the manuscript preserved in the library of
the Institute for History and Literature of the Inner Mongolian Academy for Social Sciences, Hohot/ PRC.
Anonymus. Ongod qara sakiusun teke sudur biig orosiba. Mong. 41, Royal Library, Copenhagen.
Arban buyantu nom-un aan teke, facsimile reproduction in Walther Heissig, Die Familien- und
Kirchengeschichtsschreibung der Mongolen. I. 16.18. Jahrhundert. Wiesbaden 1959, pp. 124.
Baan-a, B. (ed.). 1984. Asarai neret-yin teke. Beijing.
Bawden, Charles R. 1955. The Mongol Chronicle Altan tobi. Text, translation and critical notes.
Wiesbaden.
Dag yig mkhas pai byung gnas zhes bya ba bzhugs so/ Merged arqu-yin oron neret totaasan
dayi orosiba. Beijing block print, private collection of professor Richard Ernst.
Deged asin erdeni ber mongol oron-i tgeglgsen u-i jeglgsen irau kelen- krkirel neret.
Buryad block print.
adaadu mongol-un tr-yin jasaqu yabudal-un yamun-un engke amuulang-un y-e-d 1693
on-du keblegsen, dotor-a 1629 on-aa ekileged uda-a dara-a qaad-un y-e-d jarlasan aaja-du-i
jasamjilasan mongol-un aajan-u biig. Xerox copy of the text preserved in the State Library, Ulan Bator.
Gos lo tsa ba gZhon nu dpal. Deb ther sngon po. Chengdu 1984.
Haenisch, Erich (ed.). 1955. Eine Urga-Handschrift des mongolischen Geschichtswerks von Secen
Sagang (alias Sanang Secen). Berlin.
Haenisch, Erich. 1962. Monghol un niuca tobcaan (Yan-chao pi-shi). Die geheime Geschichte
der Mongolen aus der chinesischen Transkription (Ausgabe Ye Th-hui) im mongolischen Wortlaut
wiederhergestellt. Teil I: Text. Wiesbaden.
Heissig, Walther (ed.). 1958. Altan krdn mingan gegest biig. Eine mongolische Chronik von
Sireget Guosi Dharma (1739). Herausgegeben und mit Einleitung und Namensverzeichnis versehen.
Kopenhagen.
Heissig, Walther (ed.). 1961. Erdeni-yin erike. Mongolische Chronik der lamaistischen Klosterbauten
der Mongolei von Isibaldan (1835). In Faksimile mit Einleitung und Namensverzeichnis herausgegeben.
Kopenhagen.
Heissig, Walther (ed.). 1962. Bolur Toli Spiegel aus Bergkristall von Jimbadorji (18341837). Buch
III: Geschichte der Mongolen. Kopenhagen.
Jorung-a (ed.). Erdeni tunumal neret sudur orosiba. Beijing 1984.
Mostaert, Antoine; Cleaves Francis W. (eds.). 1959. Bolor Erike. Mongolian Chronicle by Rasipungsu.
5 vols., Cambridge, MA.
Ngag dbang blo bzang chos kyi rgya mtsho. rJe btsun btsun thams cad mkhyen pa bsod nams rgya
mtshoi rnam thar dngos grub rgya mtshoi shing rta zhes bya ba bzhugs so, in Phags pa jig rten dbang
phyug gi rnam sprul rim byon gyi khrungs rabs deb ther nor bui phreng ba, Vol. 2, Dharamsala 1984,
pp. 1171.
Prajsagara. Boda neyii toyin dalai majusryi-yin domo-i todorqai-a geyiglgi indamani erike
kemegdek orosiba. Blockprint.
THE INVENTION OF SHAMANISM IN 18TH CENTURY MONGOLIAN ELITE DISCOURSE 105

Pringlai (ed.). 1960. Byamba. Asarai neret-yin teke. Ulan Bator.


Qad-un ndsn quriyangui altan tobi. Kkeqota 1980.
Siluun onul-tu kemegdek sudur. Translated by Sireget gsi orji. St. Petersburg Kanjur, No. 839.
Siluun budaun ye onoqui neret sudur. Translated by Toyin guosi.
Thuu bkvan Blo bzang chos kyi nyi ma. Grub mtha thams cad kyi khungs dang dod tshul ston pa
legs bshad shel gyi me long. Indian reprint in pothi format,
Toboev, Tuultur. 1935. Qori kiged auyin buriyad-nar-un urida-daan bolusan anu. In: N. Poppe
(ed.), Letopisi khorinskikh buryat. Vyp. I. Khroniki Tuultur Toboeva i Vandana Yumsunova. Trudy instituta
vostokovedeniya, IX. Moskva, Leningrad, pp. 147.
Yumsunov, Vandan. 1935. Qori-yin arban nigen eige-yin jun-u u ijaur-un tuuji. In: N. Poppe
(ed.), Letopisi khorinskikh buryat. Vyp. I. Khroniki Tuultur Toboeva i Vandana Yumsunova. Trudy instituta
vostokovedeniya, IX. Moskva, Leningrad, pp. 48172.

Literature Cited

Banzarov, Dorji. 1846. Chernaya vera ili shamanstvo u Mongolov. Kazan.


Bawden, Charles R. 1968. The Modern History of the Mongols, London and New York.
Bawden, Charles R. 1958. Two Mongolian texts Concerning Obo-worship, Oriens Extremus 5/1
(1958), pp. 5761.
Bourdieu, Pierre. 1976. Entwurf einer Theorie der Praxis auf der ethnologischen Grundlage der
kabylischen Gesellschaft, Frankfurt am Main.
Bourdieu, Pierre. 1987. Sozialer Sinn. Kritik der theoretischen Vernunft, Frankfurt am Main.
Chiodo, Elisabetta. 2009. The Mongolian Manuscripts on Birch Bark from Xarbuxyn Balgas in the
Collection of the Mongolian Academy of Sciences, Part 2, Wiesbaden.
Dumas, Dominique. 1987. Aspekte und Wandlungen der Verehrung des Herdfeuers bei den Mongolen.
Eine Analyse der mongolischen Feuergebete. Bonn.
Elverskog, Johan. 2003. The Jewel Translucent Stra. Altan Khan and the Mongols in the Sixteenth
Century. Leiden.
Hart, William D. 2000. Edward Said and the Religious Effects of Culture, Cambridge.
Harva, Uno. Die religisen Vorstellungen der altaischen Vlker, Helsinki (without year).
Heissig, Walther. 1946. Bolur Erike Eine Kette aus Bergkristallen. Eine mongolische Chronik der
Kienlung-Zeit von Rasipungsu (1774/75). Literaturhistorisch untersucht. Peiping.
Heissig, Walther. 1953; 1954. Neyii Toyin. Das Leben eines lamaistischen Mnches (15571653).
Aus seiner mongolischen Biographie bersetzt und mit einer Einleitung. In: Sinologica, III, 1953, pp.144;
IV, 1954, pp. 2138.
Heissig, Walther. 1953 A Mongolian Source to the Lamaist Suppression of Shamanism, Anthropos,
48, pp. 229, 493536.
Heissig, Walther. 1959. Die Familien- und Kirchengeschichtsschreibung der Mongolen. I. 16.18.
Jahrhundert. Wiesbaden.
Heissig, Walther. 1970. Die Religionen der Mongolei. In: G. Tucci; W. Heissig, Die Religionen Tibets
und der Mongolei. Stuttgart.
Heissig, Walther. 1975. Toyin guosi ~ Guisi alias ortu guisi: Versuch einer Identifizierung,
Zentralasiatische Studien 9, pp. 361446.
Heuschert, Dorothea. 1998. Die Gesetzgebung der Qing fr die Mongolen im 17. Jahrhundert anhand
des Mongolischen Gesetzbuches aus der Kangxi-Zeit (16621722). Wiesbaden.
Hutton, Ronald. 2001. Shamans. Siberian Spirituality and the Western Imagination, Hambledon and
London.
Kmpfe, Hans-Rainer. 1983. Das Asarai neret-yin teke des Byamba Erke Daiing alias amba
Jasa. (Eine mongolische Chronik des 17. Jahrhunderts), Wiesbaden.
106 KARNINA KOLLMAR-PAULENZ

Kasyanenko, Zoya K. 1993. Katalog peterburgskogo rukopisnogo Ganzhura. Sostavlenie, vvedenie,


transliteraciya i ukazateli. Moskva.
Kollmar-Paulenz, Karnina. 2000. Religionslos ist dieses Land: Das Mongolenbild der Tibeter,
Asiatische Studien/ Etudes asiatiques LIV/4, pp. 875905.
Kollmar-Paulenz, Karnina. 2001. Erdeni tunumal neret sudur. Die Biographie des Altan qaan der
Tmed-Mongolen. Ein Beitrag zur Geschichte der religionspolitischen Beziehungen zwischen der Mongolei
und Tibet im ausgehenden 16. Jahrhundert. Wiesbaden.
Kollmar-Paulenz, Karnina. 2003. Der Buddhismus als Garant von Frieden und Ruhe. Zu religisen
Legitimationsstrategien von Gewalt am Beispiel der tibetisch-buddhistischen Missionierung der Mongolei
im spten 16. Jahrhundert, Zeitschrift fr Religionswissenschaft, 11, pp. 185207.
Kollmar-Paulenz, Karnina. 2003. Zur europischen Rezeption der mongolischen autochthonen
Religion und des Buddhismus in der Mongolei. In: Peter Schalk et al. (eds.), Religion im Spiegelkabinett.
Asiatische Religionsgeschichte im Spannungsfeld zwischen Orientalismus und Okzidentalismus, Uppsala,
pp. 243288.
Kollmar-Paulenz, Karnina. 2005. The Third Dalai Lama Snam Gyatso and the Fourth Dalai Lama
Ynten Gyatso. In: Martin Brauen (ed.), The Dalai Lamas. A Visual History. Chicago, pp. 5259.
Kollmar-Paulenz, Karnina. 2008. Uncivilized Nomads and Buddhist Clerics: Tibetan Images of the
Mongols in the 19th and 20th Centuries. In: Monica Esposito (ed.), Images of Tibet in the 19th and 20th
Centuries, Vol. II, Paris, pp. 707724.
Ligeti, Louis. Catalogue du Kanjur mongol imprim. Vol. 1, Catalogue. Budapest 1942.
Liu Jin Se, Mongol-un quriyangui teke, Kkeqota 1998.
Morikawa, T. Study of the Biography of Altan Khan. Fukuoka 1987.
Nima, Mongolud-un al. Kkeqota 2003.
Osterhammel, Jrgen. Wissen als Macht. Deutungen interkulturellen Nichtverstehens bei Tzvetan
Todorov und Edward Said. In: J. Osterhammel, Geschichtswissenschaft jenseits des Nationalstaates. Studien
zu Beziehungsgeschichte und Zivilisationsvergleich, Gttingen 2001, pp. 240265.
Pallas, Peter Simon. 1980. Sammlungen historischer Nachrichten ber die mongolischen Vlkerschaften,
2 Theile. Um eine Einfhrung vermehrter Nachdruck der 1776 und 1801 bei der Kaiserlichen Akademie
der Wissenschaften in St. Petersburg erschienenen Ausgabe. Mit einer Einfhrung von S. Hummel, Graz.
Qurabaatur, L.; jm-e, . 1991. Mongol-un bge mrgl-n tayil-a takil-a-yin soyol.
Rachewiltz, Igor de. 2004. The Secret History of the Mongols. A Mongolian Epic Chronicle of the
Thirteenth Century. Translated with a historical and philological commentary. 2 vols. Leiden/Boston.
Sagaster, Klaus. 1976. Die Weisse Geschichte (aan teke). Eine mongolische Quelle zur Lehre von
den Beiden Ordnungen Religion und Staat in Tibet und der Mongolei, Wiesbaden.
Sagaster, Klaus. 2007. The History of Buddhism among the Mongols. In Ann Heirman; Stephan
Peter Bumbacher (eds.), The Spread of Buddhism. Leiden/Boston, pp. 379432.
Said, Edward. 1978. Orientalism. Western Conceptions of the Orient, London.
Serruys, Henry. 1963. Early Lamaism in Mongolia, Oriens Extremus 10 (1963), pp. 181216.
Serruys, Henry. 1966. Additional Note on the Origin of Lamaism in Mongolia, Oriens Extremus
13 (1966), pp. 165173.
Stuckrad, Kocku von. 2003 Schamanismus und Esoterik. Kultur- und wissenschaftsgeschichtliche
Betrachtungen, Leuven.
Vladimircov, Boris Ya. 1926. Nadpisi na skalakh khalkhaskogo Coktu-tajdzhi, vol. I. Izvestiya
Akademii Nauk SSSR, pp. 12721273.
Vostrikov, Andrei I. 1962. Tibetskaya istoricheskaya literatura. Moskva.
Young, Robert J.C. 2010. Postcolonialism. An historical introduction, Malden and Oxford.
Znamenski, Andrei. 2007. The Beauty and the Primitive. Shamanism and the Western Imagination,
Oxford.
Zhu rong ga. 1991. A-le-tan han zhuan. Hohot.
R O C Z N I K O R I E N T A L I S T Y C Z N Y, T. LXV, Z. 1, 2012, (s. 107119)

NORBERT KORDEK

Segmentotactics of Mandarin Chinese

Abstract

The aim of this paper is to propose an extension of Baczerowskis phonotactic theory


and introduce a research project on Mandarin Chinese segmental linguistics in the proposed
theoretical framework. The proposal includes a brief summary of Baczerowskis original
framework and a more detailed description of its extension to various levels of linguistic
analysis. The application of the consequent extended framework to the analysis of the
phonetics, phonology (phonemic and syllabic levels), morphology and writing system of
Mandarin Chinese is proposed in the main part of the paper.

Keywords: linguistics, Mandarin Chinese, segmentotactics

1. Baczerowskis framework

In his only published paper to date on his phonotactological theory, Baczerowski


shows its application to a fragment of Mandarin Chinese (MC) phonotactics, using the
example of the pinyin transliteration.1 In the introduction to his research he provides
a non-formal account of the axiomatic phonotactological theory. Such an introduction
is also necessary for our purposes, but is should be stressed that the theory in question
is in fact a complex axiomatic system. More importantly, in this paper we treat the
phonotactological theory as a departure point for a more flexible segmentotactological
framework which is capable of multi-level analysis. In this approach, phonotactology is
a member of the class of segmentotactological theories.

1 Baczerowski 2009.
108 NORBERT KORDEK

Conformity with Baczerowskis work requires us to distinguish between:


segmentotactology (a class of linguistic theories)
segmentotactics (the subject matter of segmentotactological theories).
The subject matter of segmentotactology may be briefly defined as a word grammar
understood as a calculus that in research practice produces relevant results by the
means of computational analysis.
An analysis of this type requires a certain type of data to be available for computational
processing. Baczerowski lists four conditions for a database to be considered suitable
for phonotactological analysis:
(i) it should be sufficiently representative of the vocabulary of a given language;
(ii) the entries should be solely words (not including syntagms composed of more
than one word);
(iii) it should be accessible in an electronic form;
(iv) the word-entries should be given in phonetic transcription.2
Confronted with reality of Chinese electronic dictionaries, such conditions turn out
to be rather demanding and present the most challenging task in conducting the research.
For example the MDGB English-Chinese Dictionary (CC-CEDICT) that was used by
Baczerowski contains numerous syntagm-entries, which means that it does not satisfy
condition (iv).

1.1. Terminology

The uniqueness of Baczerowskis phonotactological theory is reflected in the terms


he uses. That is one reason why the terminology needs to be explained here. Another
reason is that the framework expanded to any other level of linguistic analysis will utilize
analogons of the term coined for the need of phonotactics. The definitions are quoted
after Baczerowski.
An utterance is a spatio-temporal physical object, individual and concrete, produced
hinc et nunc by a definite speaker in a definite time and space In a certain sense an
utterance is a linear object consisting of phonical substance, having its beginning, duration
and termination in time, and immediately preceded and succeeded by pauses.3
A vocabulon (actual word) is a maximal unit of linear, that is, sequential, ordering
of an utterance. Putting it differently, the linear structure of an utterance may be imagined
as a sequence consisting of vocabulons as always linearly continuous and relatively easy
distinguishable units within utterances.4
A phonaton is any subvocabulonic part or segment of various size, provided it is
linguistically relevant. Each phonaton is also as individual and concrete as its corresponding
vocabulon and it is always a linearly continuous unit. Needless to say, every vocabulon

2 Ibid., p. 9.
3 Ibid., p. 10.
4 Ibid.
SEGMENTOTACTICS OF MANDARIN CHINESE 109

will be treated as a particular kind of phonaton. Thus, every vocabulon is also its own
subphonaton.5 Further Baczerowski distinguishes two kinds of phonatons:
proper,
virtual,
the latter being an asubstantial object (zero segment, a pause, a moment of silence).6
A phonon is a minimal phonaton; this term is preferred to sound or actual phone
for technical reasons.7
A phone is a set of homophonous (auditory indistinguishable) phonons.
A vocable is a set of homophonous and homosignificant vocabulons. The term word
would be ambiguous in this terminological setting.
To the notions of a phonon and a phone there correspond two possible linear
representations of vocabulons and vocables:
phononotacton,
phonotacteme,
the former being the linear representation of vocabulons in terms of phonons, the latter
the linear representation of vocables in terms of phones.8 In research practice the notion
of phonotacteme will be utilized much more frequently.
So far no new types of linguistic segments or units have been defined, the new terms
being coined for the sake of precision and for technical reasons to avoid ambiguity. At
this point, however, the introduction of theory-specific terms is necessary.

1.2. Tactophoneme

The vocables consist of sequences of phones; a different way of putting it is to say


that certain sets of phones sequentialize (tactify) in the vocables. A tactophoneme will
be conceived as a set of phones that tactify in a phonotacteme. We shall avail ourselves
of the example of Polish tactophoneme {A, K, T}9 which is a set of three phones; out
of all possible permutations eight result in phonotactemes representing the corresponding
Polish vocabulons: AKT, TAK, KAT, TKA, AKTA, KATA, TAKA, ATAK.
The properties of a tactophoneme may be described in terms of:
(i) phonicity: the number of phones which are its elements;
(ii) phonotactemic range: the set of all phonotactemes generated out of it;
(iii) phonotactemicity (phonotactemic load): the number of all phonotactemes generated
out of it.10

5 Ibid.
6 Ibid.
7 Ibid.
8 Ibid., p. 11.
9 For reasons of illustration - at this point Polish serves as a better example than Chinese, as will be

explained later.
10 Baczerowski 2009, p. 13.
110 NORBERT KORDEK

The characteristics of the tactophoneme in the above example are as follows:


(i) phonicity: 3
(ii) phonotactemic range: {AKT, TAK, KAT, TKA, AKTA, KATA, TAKA, ATAK}
(iii) phonotactemicity: 8.
Other important properties of phones are described by their:
(i) tactophonemic dispersion the set of all tactophonemes to which a given phone
belongs;
(ii) tactophonemic dispersion number the number of all tactophonemes to which
a given phone belongs;
(iii) phonotactemic dispersion the set of all phonotactemes in which a given phone
occurs;
(iv) phonotactemic dispersion number the number of all phonotactemes in which
a given phone occurs.11
The other important property of tactophonemes is described by their phonotactemic
efficiency the ratio between the phonotactemicity and the phonicity of a given
tactophoneme.12 The phonotactemic efficiency of the exemplary tactophoneme {A, K, T}
equals 2.6 (its phonotactemicity is 8, and its phonicity 3). The notion of phonotactemic
efficiency may be extended to the whole family of tactophonemes. In the extended
interpretation the most obvious understanding, but not the only one possible, involves
treating it as the ratio between the number of all phonotactemes and the number of all
tactophonemes.13
The above terms are sufficient for our main purpose, i.e. introducing the concept of
extending the framework to different levels of linguistic signs structure.

1.2. Phonotactics of Mandarin Chinese

The detailed results of phonotactemic analysis of MC and Polish are presented in


Baczerowski 2009. Due to the unavailability of electronic word databases (dictionaries)
with phonetic or phonematic transcriptions, the results of Baczerowskis research in
fact pertain to the domain of orthographic systems of Polish and pinyin transliteration
of Chinese. However it seems justified to assume that the results in Baczerowskis
paper are a fairly close approximation of the MC and Polish word tactics in terms of
the phonematic structure, or at least it gives us some insight into the tactical structure
of words.
The lack of appropriate databases is less problematic for MC than it is for Polish.
Due to the syllable structure and the small number of syllables it is possible to create
an algorithm which automatically converts syllables represented in pinyin transliteration
into the phonetic or phonematic transcriptions.

11 Ibid., p. 14.
12 Ibid., p. 15.
13 Ibid.
SEGMENTOTACTICS OF MANDARIN CHINESE 111

The restrictions on MC syllable structure make the possible linear alignments of phones
or phonemes rather predictable. The small size of the MC syllabary, which consists of
404 syllables not taking tones into account, and the lack of influence of morphology on
the syllable structure , also make the task much easier than in the case of Polish.
The MC phone inventory is a controversial issue.14 In Dyczkowski et al. we proposed
an inventory for MC consisting of 76 phones,15 but in the most updated version we have
significantly reduced the inventory to 52 phones:
[p p b b b b t t d d d d k g g f s x m m
n n l j w i y u e a o ]. Conducting the computational analysis on
the transcribed syllabary, and consequently applied to the lexical entries in an electronic
dictionary, instead of on the transcribed texts, probably requires further reduction of the
inventory, but this issue is beyond the scope of this paper.
For the sake of terminological clarity it should be noted that the terms tactophoneme
and phonotacteme both pertain to tactical analysis in terms of phones, which in the
former case might be misleading in suggesting a phonemic analysis. It seems justified
to adjust the terminology by applying the terms tactophone and phonotacteme for the
phonotactical level of analysis, while for the phonemotactical level the terms tactophoneme
and phonemotacteme should be used.
The terminological issues, and the fact that Baczerowski had to inquire into
orthographical systems rather than phonotactic ones, paradoxically prove the versatility
of the framework.
This paper aims to explore this versatility by expanding Baczerowskis proposal
beyond phonotactics.

2. Beyond phonotactics

Baczerowski was well aware of the fact that he was exemplifying his phonotactical
framework with an inquiry into a different level of language structure, which he calls
graphotactic.16
It is a common and prevailing practice in Chinese linguistics to reduce the phonetic
and phonological studies of MC to a subfield of psycholinguistics, cognitive linguistics
or cognitive psychology; Myers and Tsay, for example, place the phonetic studies within
the field of experimental psycholinguistics phonetic studies test hypotheses about
how phonological knowledge (competence) is actually used (performance), and as such
they can be highly relevant to phonological theory and to our understanding of how the
mind works.17 One of the consequences of this approach is the assumption that positing

14 The issue of the phone inventory controversy was addressed in Dyczkowski et al. 2009.
15 Ibid., p. 90.
16 Baczerowski 2009, p. 17. For this level we use a different term, for reasons to be explained in the following

sections.
17 Myers, Tsay 2003, p. 30.
112 NORBERT KORDEK

segmental phonetic units is not necessary in MC. Segmentotactics explicitly presumes


the opposite that the segmental phonetic units as a representation of the phonological
system are a valid and interesting subject of linguistic inquiry.

2.1. Orthotactics

Orthotactics is used as a replacement term for the graphotactics mentioned above.


The reason for this becomes apparent when we are confronted with the diversity of the
writing systems of world languages. It is probably justified to assume that in the case of
languages using alphabetical writing systems the two terms could be used synonymously,
since it seems difficult to associate them with two different levels of tactical analysis in
those languages there is no relevant graphical level of the writing system other than
orthography. However, the same cannot be said of languages with non-alphabetical writing
systems, such as Chinese. The graphic aspect is inherently associated with the Chinese
script; on the other hand it is not immediately obvious what orthography means in reference
to MC. The units that tactify into the written representations of vocables in the two types
of writing systems are of a very different nature. The letter type units in alphabetical
systems more or less directly reflect the phonetic or phonemic properties of a vocable,
while in the case of the Chinese logographic script the internal structure of individual
characters is not restricted by such properties of vocables. In other words, if we accept
this terminological distinction, orthotactics would pertain to writing systems dependent
on the phonetic and phonological properties of a given language, while graphemotactics
would refer to systems independent of phonetics and phonology.
The orthotactics of non-alphabetical writing systems, including Chinese script, is not
then a direct inquiry into the writing system, but instead into its alphabetical transliteration.
The proposed terminology is analogous to the phonotactical case. The orthotacteme will
be the linear representation of vocables in terms of letters. The tactorthoneme will be
conceived as a set of letters that tactify in a orthotacteme. The following terms relate
to a tactorthoneme:
(i) orthocity: the number of letters which are its elements;
(ii) orthotactemic range: the set of all orthotactemes generated out of it;
(iii) orthotactemicity (orthotactemic load): the number of all orthotactemes generated
out of it;
(iv) orthotactemic dispersion the set of all orthotactemes in which a given letter
occurs;
(v) orthotactemic dispersion number the number of all orthotactemes in which
a given letter occurs;
(vi) orthotactemic efficiency the ratio between the orthotactemicity and the orthocity
of a given tactorthoneme.
As was already mentioned, in the case of analysis of the Chinese script, the orthotactic
analysis is an inquiry into the transliteration system. The results based on the pinyin
transliteration presented by Baczerowski (2009) reflect the relevant properties of
SEGMENTOTACTICS OF MANDARIN CHINESE 113

MC. For example the orthotactemic efficiency is expected to be lower than in Polish.
The reason for this is the syllable and word structure of MC and the related issue
of the syllable-morpheme-word correspondence.18 The typical word in MC consists of
two syllables. Every syllable is subject to rigorous restrictions on its linear structure.
Typically only one permutation of the elements of a tactophoneme is allowed (the same
is true for a tactorthoneme). For example the tactorthoneme {L, O, N, G} tactifies
into one orthotacteme only: {LONG}. The only theoretical possibility of increasing
the orthotactemic efficiency of most MC tactorthonemes is the existence of a vocable
consisting of a duplicated syllable {LONGLONG} in this particular example. In the
case of tactorthonemes that can tactify into bisyllabic vocables, for example {S, H, I,
H, E}, the typical efficiency equals one, with the exception of cases where there exist
orthotactemes representing the vocables with reversed syllabic linear order. In the above
example the orthotactemic efficiency equals 2, since both orthotactemes SHIHE and
HESHI represent vocables of MC. The restrictions on the linear order of syllables and the
related small number of syllables in MC are the main factors which reduce phonotactemic
and orthotactemic efficiency. On the other hand the possibility of syllable duplication
and permutations in the syllabic linear order a phenomenon non-existent in Polish
increase the efficiency. In extreme cases the efficiency may increase to values not seen
in Polish:
{N, A, I}: {NAI, NIAN, NAINAI, NIANNIAN, NINA, NANI, NAINA, NANAI,
NANAI, AINAN, NANAI, NAINAN, NANNAI, NINIAN, NIANNI, AINAI, NAIAI,
AINA, NAAI, AINIAN, NIANAI, NIANAI}. Intuitively, out of the properties having
an opposite effect on efficiency, the number of syllables and the restrictions on linear
order within the syllable are expected to dominate the tactical properties of MC vocables.
This intuition is confirmed by the results obtained by Baczerowski. The orthophonemic
efficiency of Polish is 1.36 while that of MC is only 1.11.

2.2. Phonemotactics

The orthotactic analysis of pinyin transliteration was conducted mostly for the reason
of its accessibility in terms of the existing databases. The linguistic importance of an
inquiry into the transliteration system is perhaps questionable, but the research output can
be at least considered as an approximation of the phonemotactical one. Phonemotactics
is of course understood as a tactical analysis of vocables in terms of phonemes.
The phoneme inventory is much less controversial in MC than the repertoire of
phones. For the purpose of the computational tactical analysis, the inventory or even the
complete phonemic transcription of all MC syllables in the referential work of Duanmu19
can be adapted.

18 These properties have significance for every type of tactical analysis of MC, not only orthotactical.
19 Duanmu 2007.
114 NORBERT KORDEK

The proposed terminology is analogous to the phonotactical and orthotactical cases. The
phonemotacteme will be the linear representation of vocables in terms of phonemes. The
tactophoneme will be conceived as a set of phonemes that tactify in a phonemotacteme.
The following terms are related to a tactophoneme:
(i) phonemicity: the number of phonemes which are its elements;
(ii) phonemotactemic range: the set of all phonemotactemes generated out of it;
(iii) phonemotactemicity (phonemotactemic load): the number of all phonemotactemes
generated out of it;
(iv) phonemotactemic dispersion the set of all tactophonemes to which a given
phoneme belongs;
(v) phonemotactemic dispersion number the number of all tactophonemes to which
a given phoneme belongs;
(vi) phonemotactemic efficiency the ratio between the phonemotactemicity and the
phonemicity of a given tactophoneme.
The numbers themselves show that phonemotactic analysis should render different
results than the orthotactical one. In the pinyin system there are 29 symbols, while
there are 24 phonemes in the MC inventory.20 Straightforward logic would tempt one to
speculate that in MC the phonemotactemic efficiency will be higher than the orthotactemic
efficiency, but the issue is probably more complicated than that.

2.3. Syllabotactics

One of the most characteristic typological features of MC is its syllable prominence.


Due to the one-to-one correspondence of syllables and morphemes,21 the syllabic structures
also shape and determine the morphological and lexical structures.
The following terminology is proposed: the syllabotacteme will be the linear
representation of vocables in terms of syllables, and the tactosyllable will be conceived
as a set of syllables that tactify in a syllabotacteme. The following terms relate to
a tactosyllable:
(i) syllabocity: the number of syllables which are its elements;22
(ii) syllabotactemic range: the set of all syllabotactemes generated out of it;
(iii) syllabotactemicity (syllabotactemic load): the number of all syllabotactemes
generated out of it;
(iv) syllabotactemic dispersion the set of all tactosyllables to which a given syllable
belongs;
(v) syllabotactemic dispersion number the number of all tactosyllables to which
a given syllable belongs;

20 Ibid. The number of phonemes differs according to different accounts.


21 Also the Chinese characters are in direct correspondence to morphemes, and thus are in the same kind of
correspondence with syllables.
22 Syllabicity, which is a more obvious terminological choice, is already in use in phonology.
SEGMENTOTACTICS OF MANDARIN CHINESE 115

(vi) syllabotactemic efficiency the ratio between the syllabotactemicity and the
syllabocity of a given tactosyllable.
A detailed description of the syllabic system of MC is beyond the scope of this paper;
some very basic properties must, however, be addressed. It was already mentioned that
the syllabary of MC consists only of slightly more than 400 syllables, if the tones are
ignored. The average ratio of syllables per word is also very low, since most Chinese
words are bisyllabic. In theory this suggests the following:
the low syllabocity of tactosyllables and their relatively high efficiency;
the high efficiency of the family of all tactosyllabons.
The latter point may be understood in different ways, namely as:
the ratio of all tactosyllables to all syllabotactemes;
the average efficiency of tactosyllables;
the permutational efficiency of tactosyllables (the ratio of existing tactosyllabons to
all potential ones).
Considering just the number of syllables and the average ratio of syllables per word
in MC, it is quite obvious that some other factors must be also taken into account,
otherwise there would not be enough building material for the lexical level. MC copes
with the problem by way of an extremely high homonymy of morphemes and words. On
one hand, then, the system itself is potentially highly efficient, while on the other there
is a phonological-lexical mechanism which decreases the efficiency. The other important
factors influencing the results of syllabotactic analysis are some of the word-formation
strategies in MC, which include:
morphological reduplication,
inversion of linear order of morphemes23 in bisyllabic words,
some types of grammatical affixation,
all contributing to increasing the syllabotactemic efficiency. Some examples of
tactosyllables with corresponding syllabotactemic range are:
{YI}: {YI, YIYI} (duplication)
{FENG, MI}: {MIFENG, FENGMI} (inversion)
{YI, DE}: {YIDE, DEYI, YIDEDE, DEYIDE} (affixation + inversion).
The results of the syllabotactic analysis of MC are at this point only a speculation,
but it is worth noticing that simply the fact that such analysis is interesting and relevant
as linguistic research makes MC (and languages that share similar features of the syllabic
system) unique. For example, in the case of Polish, syllabotactic analysis in the vein
proposed here does not seem to be relevant. The above proposal ignores the tones in
MC, but an actual analysis should take the tones into account; interesting results should
be obtained by comparing the tonal and toneless analyses.

23 Inversion of syllables is even more common, but in the case where the morphological identity of a syllable

is not retained it cannot be called a word-formation strategy.


116 NORBERT KORDEK

2.4. Morphotactics

The analysis of word structure, word-formation, word semantics, etc. in research


practice is almost synonymous with a morphological analysis of some kind. Tactical
analysis of words in terms of morphemes is usually understood as a study of their
occurrence and ordering restrictions in different phonological environments. Due to the
general properties of the morphological system, the analysis that is proposed here cannot
be expected to render any interesting results in most languages. In MC, however, there are
reasons to believe that morphotactical analysis in the vein of Baczerowskis framework
is justified.
The following terminology is proposed: the morphotacteme will be the linear
representation of vocables in terms of morphemes. The tactomorpheme will be conceived
as a set of morphemes that tactify in a morphotacteme. The following terms relate to
atactomorpheme:
(i) morphemicity: the number of morphemes which are its elements;
(ii) morphotactemic range: the set of all morphotactemes generated out of it;
(iii) morphotactemicity (morphotactemic load): the number of all morphotactemes
generated out of it;
(iv) morphotactemic dispersion the set of all tactomorphemes to which a given
morpheme belongs;
(v) morphotactemic dispersion number the number of all tactomorphemes to which
a given morpheme belongs;
(vi) morphotactemic efficiency the ratio between the morphotactemicity and the
morphemicity of a given tactomorpheme.
By stating that morphotactical analysis in most cases would not render any interesting
result, we mean that a large number of very inefficient tactomorphemes is the expected
result. MC offers more promising possibilities. The reasons are in general the same as
in the case of syllabotactics:
morphological reduplication (for example { man}: { man, people});
inversion of the linear order of morphemes in bisyllabic words (for example { m
honey fng bee}: { mfng bee fngm honey}.
The expected results are relatively high morphotactemicity and morphotactemic
efficiency of tactomorphemes.

2.5. Graphemotactics

Probably the most unique tactical analysis in MC refers to one of its most unique
features the script. Even an introductory description of Chinese script is beyond the
limitations of this paper, so we will only mention the most relevant features which directly
determine the proposed framework:
discreteness the characters are composed of recurring elements, except for a small
number of one-stroke characters which are not compositional;
SEGMENTOTACTICS OF MANDARIN CHINESE 117

lack of obligatory information neither phonetic nor semantic information is required


as a component part of a character;
the spatial arrangement (including linear sequence) of components is relevant for
distinguishing the characters.
The complexity of the Chinese writing system presents us with the problem
of determining the most basic concepts of the tactical analysis of characters. We
provisionally propose that the grapheme is a component part of a character. We have yet
to determine the exact meaning of the term it is possible to understand it as a radical-
type component or as a stroke-type component this problem will be discussed later.
The remaining basic terminology is as follows: The graphotacteme will be the spatial
representation of vocables in terms of graphemes. The tactographeme will be conceived
as a set of graphemes that tactify in a graphotacteme. The following terms relate to
a tactographeme:
(i) graphemicity: the number of graphemes which are its elements;
(ii) graphotactemic range: the set of all graphotactemes generated out of it;
(iii) graphotactemicity (graphotactemic load): the number of all graphotactemes
generated out of it;
(iv) graphotactemic dispersion the set of all tactographemes to which a given
grapheme belongs;
(v) graphotactemic dispersion number the number of all tactographemes to which
a given grapheme belongs;
(vi) graphotactemic efficiency the ratio between the graphotactemicity and the
graphemicity of a given tactographeme.
The average efficiency of the tactographemes is not expected to be high, since the
majority will have efficiency equal to 1. This is due to the fact that in most cases the
same set of components makes up a single character; however the character formation
rules allow variations in the spatial arrangement of components resulting in different
characters, and another important mechanism of character formation is the recurrence of
components. The following examples of tactographemes and their graphotactemic range
illustrate these properties:
{}: {m tree ln woods sn forest} (recurrence)
{}: { dn dawn gn continuous} (recurrence)
{,}: { dng cubes, ch footstep} (recurrence)
{, }: { gu enough, gu enough} (linear rearrangement)
{, }: { tin heaven; day, f man} (spatial rearrangement).
The components in the examples are the radicals, which is the natural, but not the
only possible analysis. The decomposition of characters reveals a multi-layer component
structure. The component structure is best represented by IC trees. The following example
is the decomposition of the character a lake:
118 NORBERT KORDEK

Every single component in the tree (water, g ancient, ru flesh,


sh ten, ku mouth) is a radical or a character with a compositional function ( h
beard in the above example), though at different branching levels. The first branching
is the most important in the sense that it corresponds to the traditional classification of
characters in terms of etymology, components type and spatial arrangement. If we continue
the decomposition to the simplest elements, regardless of the complexity of a character,
the branching nodes at the lowest possible levels will represent individual strokes.

The branchings of the more-than-two-stroke radicals functioning as character


components (( shu) water, g ancient, ru flesh, ku mouth in
the example) are more or less arbitrary. We propose to construct the tree in such a way
that in the result the lowest nodes are arranged from left to right in an order corresponding
to the stroke order in the whole character. In this way the arbitrariness is addressed by
referring to another compositional property.
A few conclusions can be drawn from the above tree. The nodes in this representation
do not differentiate between the component types. Not all of the nodes represent the true
SEGMENTOTACTICS OF MANDARIN CHINESE 119

components of a character, for example the left branching of the true component
()24 is a part of a true component and at the same time is composed of true ones,
but is not one itself. on the other hand is a complex character on its own that is used
as a true component of another character, but it is not a radical.25 This recalls the phrasal
structure of a sentence represented by X-bar syntactic trees distinguishing between the
intermediary and true phrasal components. That kind of representation would be suitable
for the purpose of character decomposition. In the case of complex characters (like the
one in the example) the components in the highest (radicals or complex characters)
and the lowest (individual strokes) nodes are a natural units of graphotactical analysis.
The unaddressed problem remains the graphotactical status of the intermediate radical
components (e.g. ) and intermediate non-radical ones (e.g. (, ).26
In the case of characters with simple structure (functioning as radicals) only the analysis
in terms of strokes is available.
It is premature at this point to speculate about the results of the graphotactemic
analysis of Chinese script, but it is certain that it will be both a challenging and interesting
project. The most serious problem in its realization is the availability of a properly tagged
database of Chinese characters.

Bibliography

Baczerowski, J. 2009. Aspects of Chinese Phonotactics Against a Comparative Background of Polish.


Scripta Neophilologica Posnaniensia X: 722.
Duanmu, S. 2007. The Phonology of Standard Chinese. Oxford University Press.
Dyczkowski, K.; Kordek, N.; Nowakowski, P.; Stroski, K. 2009. The Phonetic Grammar of Mandarin
Chinese a Computational Comparative Analysis. Rocznik Orientalistyczny LXII. 1: 8091.
(Huang Jinhong), (Zhang Zhengnan), (Zhang Xiaoyu), (Huang Jiading),
(Ye Deming). 2008. (Mandarin Chinese Phonetics), , .
() (Huang Pushu, ed.). 2006. .. , .
Myers, J.; Tsay, J. 2003. Investigating The Phonetics of Mandarin Tone Sandhi. Taiwan Journal of
Linguistics 1.1: 2968.
Packard, J.L. 2000. The morphology of Chinese. A linguistic and cognitive approach. Cambridge:
Cambridge University Press.
Rogers, H. 2005. Writing Systems: A Linguistic Approach. Oxford: Blackwell.
() (Song Yejin, Jia Jiaoyan, ed.). 2003. . .
() (Su Peicheng, ed.). 2001. . .
Wierzcho, P. 2004. Gramatyka diakrytologiczna. Studium ortograficzno-kwantytatywne. Pozna:
Wydawnictwo Naukowe UAM.

24 The shape is changed for typographical reasons.


25 The basic function of radicals is to function as components of complex characters; not all complex characters
are used in this function. Radicals may be defined as characters that can be decomposed only into strokes, while
complex characters are decomposable into radicals or less complex characters.
26 One solution to this problem is to allow ternary, quaternary, etc. branchings the radicals, regardless of their

complexity, would be decomposed directly into individual strokes. The binary trees have their advantages, though;
it is difficult at the present time to decide which solution is superior.
R O C Z N I K O R I E N T A L I S T Y C Z N Y, T. LXV, Z. 1, 2012, (s. 120139)

JOWITA KRAMER

Descriptions of Feeling (vedan), Ideation (saj),


and the Unconditioned (asaskta) in Vasubandhus
Pacaskandhaka and Sthiramatis Pacaskandhakavibh1

Abstract

This article focuses on three passages of Vasubandhus [Treatise] on the Five


Constituents of the Person (Pacaskandhaka) and its commentary by the 6th-century
Indian scholar Sthiramati, the Pacaskandhakavibh. The three parts dealt with here
comprise the sections on feeling, ideation, and the unconditioned and are compared
with parallel descriptions of these concepts in other Abhidharma texts, including the
Abhidharmakoabhya and the Abhidharmasamuccaya. The treatment of the unconditioned
factors is particularly notable since the lists of these factors vary strongly in the works
under discussion, ranging from three to nine entities. Among the unconditioned factors
the two entities cessation [obtained through] consideration (pratisakhynirodha) and
cessation not [obtained through] consideration (apratisakhynirodha) are of particular
interest and are analysed in detail in the present study.

Keywords: Buddhism, philosophy, Abhidharma, Vasubandhu, Sthiramati

1. Introduction

The present study continues a series of previously completed articles investigating


the contents of Vasubandhus Pacaskandhaka and its commentary by Sthiramati, the

1 I would like to thank Jens-Uwe Hartmann, Ralf Kramer, Lambert Schmithausen, and Jonathan Silk for

offering very helpful comments and corrections to previous drafts of this paper. I am also grateful to the German
Research Foundation (DFG) for funding my research which led to this article.
DESCRIPTIONS OF FEELING (VEDAN), IDEATION (SAJ)... 121

Pacaskandhakavibh.2 The main focus of these two works is the description of the
five constituents of the person (skandha). From the viewpoint of doctrinal development
the most relevant parts of this description are the sections on matter (rpa), impulses
(saskra), and perception (vijna). Therefore, I have dealt with these three skandhas
in three separate publications. The present investigation focuses on three other passages
of the Pacaskandhaka and its commentary that have not been considered so far. These
three passages include the treatment of the two skandhas feeling (vedan) and ideation
(saj) as well as the category of those factors that are considered to be unconditioned
(asaskta). The treatment of the two skandhas feeling and ideation comprises only
afew lines in the Pacaskandhaka and around three folios in the Pacaskandhakavibh.3
The discussion of the unconditioned is of similar extent.4 Despite the concision of
these descriptions the passages are, nonetheless, worthy of closer examination. As already
indicated in my study of the Pacaskandhakas saskra section, the comparison of the
Pacaskandhaka and the Pacaskandhakavibh with other Abhidharma texts, like the
Abhidharmasamuccaya and the Abhidharmakoabhya, reveals noteworthy parallels and
discrepancies. Therefore, I present an overview of the contents of the relevant passages
in the Pacaskandhaka and its commentary in the following, and compare them with
corresponding descriptions in the Abhidharmakoabhya, the Abhidharmasamuccaya,
and its commentary, the Abhidharmasamuccayabhya.5

2. Feeling (vedan)

In the Pacaskandhaka Vasubandhu explains feeling as experiencing (anubhava), of


which he distinguishes three kinds: pleasant (sukha), unpleasant (dukha), and neither
pleasant nor unpleasant. He then adds that a pleasant feeling is characterized by the fact
that one wishes to be (re)united with it when it has ceased. In contrast, the unpleasant
feeling leads to the wish of being separated from it, whereas the neutral feeling does
not result in either of these two desires.6 The enumeration of three alternative feelings
corresponds to the explanation of the Abhidharmakoabhya. However, the latter
adds a sixfold classification into the various kinds of feeling that arise from contact
of the eye, ear, nose, tongue, body, and the mental faculty with their objects.7 This

2 See Kramer 2008 and forthcoming A and B.


3 See PSk 3,10-4,2 and PSkV 14b2-18a1.
4 See PSk 18,12-19,7 and PSkV 61b4-64a1.
5 The quotations from the Abhidharmakoabhya, Abhidharmasamuccaya, and Pacaskandhakavibh given

below are available in English translation in Pruden (1988), Boin-Webb (2001), and Engle (2009). These translations
have been consulted for the present study and modified where it appeared necessary.
6 PSk 3,10-13: vedan katam / trividho nubhava sukho dukho dukhsukha ca / sukho yasya nirodhe

sayogacchando bhavati / dukho yasyotpdd viyogacchando bhavati / adukhsukho yasyotpdt tadubhaya


na bhavati.
7 AKBh 10,12-14: vedannubhava / trividho nubhavo vedanskandha / sukho dukho dukhsukha ca /

sa punar bhidyamna a vedanky, cakusasparaj vedan yvan manasasparaj vedaneti.


122 JOWITA KRAMER

sixfold categorization is also found in the Abhidharmasamuccaya. The latter does not
refer to feeling as anubhava within the general definition of feeling, but the term is
used in another passage of the text to describe the main characteristic of vedan.8 The
Abhidharmasamuccaya additionally explains that feeling can also be classified as physical
(kyika) or mental (caitasika), being either related to the five sense perceptions (and
thus based on the contact of the five sense faculties with their objects) or to the mental
perception (manovijna).9 Another distinction proposed by the Abhidharmasamuccaya
is between feeling associated with worldly pleasures (smia) and not associated with
worldly pleasures (nirmia) or between feeling based on craving (gredhrita) and
based on renunciation (naikramyrita). The feeling associated with worldly pleasures
is explained in the Abhidharmasamuccaya as being associated with desire for the self,
whereas the feeling that is based on craving is paraphrased as relying on greed for
the five sense objects.10 Interestingly, all the characterizations of feeling found in the
Abhidharmasamuccaya are not referred to by Vasubandhu in the Pacaskandhaka, but are
supplemented by Sthiramati in his commentary.11 Additionally, Sthiramati mentions two
interpretations of feeling by other scholars. The first specifies feeling as the experiencing
of a desirable or undesirable contact (spara), or contact that is different from both, and
is ascribed by Sthiramati to Saghabhadra.12 Sthiramati rejects this view by arguing that
in this case feeling as the experiencing of contact would either be used in the sense of
feeling accompanying contact or feeling having contact as its cause. But both these
statements are unsuitable for paraphrasing the distinct nature of feeling, as all the mental
factors share the quality of accompanying contact and because contact is the cause of
all mental factors.13 The second theory proposed by other scholars and opposed by
8 AST 52a6: tshor bai mtshan nyid ci zhe na / myong bai mtshan nyid de.
9 AST 54a5-b4: tshor bai tshogs drug ste/ mig gi dus te reg pa las byung bai tshor ba bde ba yang rung /
sdug bsngal yang rung / sdug bsngal yang ma yin bde ba yang ma yin pa yang rung ba dang / rna ba dang/ sna
dang / lce dang/ lus dang / yid kyi dus te reg pa las byung bai tshor ba bde ba yang rung / sdug bsngal yang
rung / sdug bsngal yang ma yin bde ba yang ma yin pa yang rung ste/ bde ba lus kyi yang rung / sdug bsngal
lus kyi yang rung / sdug bsngal yang ma yin bde ba yang ma yin pa lus kyi yang rung / bde ba sems kyi yang
rung / sdug bsngal sems kyi yang rung / sdug bsngal yang ma yin bde ba yang ma yin pa (D pa; P om.) sems
kyi yang rung / [] lus kyi gang zhe na/ rnam par shes pai tshogs lnga dang mtshungs par ldan pao // sems
kyi gang zhe na/ yid kyi rnam par shes pa dang mtshungs par ldan pao.
10 AS 54a8-b6: bde ba zang zing dang bcas pa yang rung / sdug bsngal dang / sdug bsngal yang ma yin
T
bde ba yang ma yin pa zang zing dang bcas pa yang rung / bde ba zang zing med pa yang rung (D yang rung;
P dang) / sdug bsngal dang / sdug bsngal yang ma yin bde ba yang ma yin pa zang zing med pa yang rung /
bde ba zhen pa rten (D rten; P brten) pa yang rung / sdug bsngal dang / sdug bsngal yang ma yin bde ba yang
ma yin pa zhen pa rten pa yang rung / bde ba mngon par byung ba rten (D rten; P brten) pa yang rung / sdug
bsngal dang / sdug bsngal yang ma yin bde ba yang ma yin pa mngon par byung ba rten pa yang rung bao
// [] zang zing dang bcas pa gang zhe na / lus la sred pa dang mtshungs par ldan pao // [] zhen pa rten
(D rten; P bsten) pa gang zhe na / dod pai yon tan lnga la sred pa dang mtshungs par ldan pao.
11 PSkV 16a4-b5.
12 PSkV 14b6-15a1: vedannubhava sparasyety cryasaghabhadra / iniobhayavipartn sparnm

anubhava s vedan. On Saghabhadra see Cox (1995), pp. 5355.


13 PSkV 15a4f.: sparnubhava ca tatsamprayoga kraabhvo v parikalpyeta / [...] tatra na bhavet samprayoga

sarvacaittn sparasamprayogviet / npi kraabhva sarvacaittn sparasya kraatvviet.


DESCRIPTIONS OF FEELING (VEDAN), IDEATION (SAJ)... 123

Sthiramati states that feeling is the experiencing of effects of maturation of previous


virtuous (ubha) and non-virtuous deeds. Sthiramati rejects this view by referring to
the assumption that only the layavijna and the neutral feeling accompanying it are
the results of maturation of virtuous and non-virtuous deeds and that the pleasant or
unpleasant feelings are actually arisen from the [result of] maturation (vipkaja), i.e.
are only secondary products of maturation.14

3. Ideation (saj)

Ideation is characterized in the Pacaskandhaka as the grasping of an objects


specific features (nimitta).15 This definition is similar to the explanation of the
Abhidharmakoabhya, except for the latters additional enumeration of six kinds of
ideation depending on the five sense faculties, the mental faculty (manas) and their
objects.16 The Abhidharmasamuccaya also mentions the phrase grasping of specific
features (mtshan mar dzin pa) when defining the characteristic nature (lakaa) of
saj,17 and it also explains that there are six kinds of saj, arising on the basis of
one of the five sense faculties or the mental faculty. However, it additionally classifies
ideation into another six kinds, namely ideation of an object associated with a specific
feature (sanimitta), of an object without a specific feature (animitta), of a limited (partta)
object, of a great (mahadgata) object, of an immeasurable (aprama) object, and of
the sphere of nothingness (kicanyyatana).18 Remarkably, the Tibetan version of the
Pacaskandhaka includes the statement that saj is of three kinds comprising ideation
of limited, great, and immeasurable objects.19 However, this explanation does not occur in
the Sanskrit manuscript of the text preserved in China, nor does it have a parallel in the
Chinese translation of the Pacaskandhaka.20 Sthiramati includes the discussion of all six
kinds of objects in his commentary without mentioning that Vasubandhu enumerated only
three of them.21 As these six kinds of saj are, moreover, not referred to in the form of
a quotation from the root-text, it is very likely that Sthiramati supplemented as he did
in the case of vedan the root-text with the description of the six objects, presumably
relying on the Abhidharmasamuccaya. Thus, the three kinds of ideation listed in the Tibetan

14 PSkV 15a6-b1: eva tu manyante ubhubhn karma phalavipka pratyanubhavanty anenety


anubhava / [...] atra clayavijnam eva ubhubhakarmavipka / tatsamprayuktaivopek paramrthata
ubhubhn karma vipka / sukhadukhayos tu vipkajatvd vipkopacra.
15 PSk 4,1: viayanimittodgrahaam.
16 AKBh 10,15-17: saj nimittodgrahatmik // [] sa punar bhidyamna a sajky vedanvat.
17 AS 52a6f.: du shes kyi mtshan nyid ci zhe na / [] mtshan mar dzin pa dang.
T
18 AS 15,21-23: a sajky / cakusasparaj saj rotraghrajihvkyamanasasparaj saj

/ yay sanimittam api sajnti, animittam api, parttam api, mahadgatam api, apramam api, nsti kicid ity
kicanyyatanam api sajnti.
19 PSk 13b2f.
T
20 See Li and Steinkellner 2008, p. 4, n. for line 2.
21 PSkV 17a5: lambana puna sanimittam, animittam, parttam, mahadgatam, apramam, kicanyyatana ca.
124 JOWITA KRAMER

text of the Pacaskandhaka are possibly a later addition. However, in case they were part
of the original Sanskrit text, the listing of only three items in contrast to the six kinds
of objects mentioned in the Abhidharmasamuccaya would provide another example of a
divergence between the latter and the Pacaskandhaka.22 As for Sthiramatis comments on
the six kinds of objects, they are probably influenced by the Abhidharmasamuccaya and
also show parallels with its commentary, the Abhidharmasamuccayabhya. One of the
common features of the latter and the Pacaskandhakavibh is the addition of the three
arguments because [the sphere of desire] is inferior, because [the material sphere] is
superior, and because [the spheres of unlimited space and unlimited consciousness] are
boundless. They have been introduced to explain the nature of the three kinds of objects
limited, great, and immeasurable. In contrast to the Abhidharmasamuccayabhya,
Sthiramati also mentions the reason why the object of the sphere of nothingness is called
nothingness: because there is no grasping at all.23 Sthiramati also includes explanations
of the terms sanimitta and animitta in his commentary. He defines the first class of
objects as follows:

Whichever [feature] is ascribed [to an objects nature] by means of aword[,


this is its specific feature]. The entitys intrinsic nature endowed with this
specific feature is associated with a specific feature. The ideation which
[recognizes that] this is the nature ascribed to the object [and] this is the
term [referring to it] this is the ideation which has an object associated
with a specific feature.24

According to Sthiramati, the object not associated with a specific feature (animitta)
refers to an object not having a feature that can be ascribed to it. This can apply to the
nature of an entity (vastusvarpa), nirva, or the peak of existence (bhavgra). In the
first case the object is not associated with specific features because there is no specific
feature, [in the case of nirva] because there is no specific feature of matter and feeling,
and [in the case of the peak of existence] because there is no clarity of the specific
feature.25 Sthiramati goes on to explain that in the first case someone who is inexperienced
in connecting words and objects has an ideation of the nature of matter but does not

22 For other examples of differences between these two texts see Kramer (forthcoming A).
23 PSkV 17b5-18a1: partta kmadhtur nikatvt / mahadgato rpadhtu, utkatvt / aprama
kavijnnantyyatane, aparyantatvt / parigrahakicanbhvd kicanyyatanam kicanyam / tasmt
kmadhtvdylamban saj parttdisajety ucyate. ASBh 4,14-16: partta kmadhtu nikatvt / mahadgato
rpadhtus tat utkatvt / aprame kavijnnantyyatane paryantatvt / tasmt tadlaban saj
parttdisaj veditavy.
24 PSkV 17a5-6: tatra abdadvrea yad artharpam adhyropitam, tena nimittena vastusvarpa sanimittam

(PSkVT 16af.: de la sgrai sgo nas gang don gyi ngo bor sgro btags pa de ni mtshan ma ste/ mtshan ma de dang
lhan cig pai dngos poi rang gi ngo bo ni mtshan ma dang bcas pao) / tasmin vastuny adhyropita rpam
etac chabdo sv iti y saj, s sanimittlamban.
25 PSkV 17a6-17b1: animitta punar anadhyropitkra vastusvarpam, nirvam, bhavgra ca.

tatrdhyropitanimittbhvd rpavedannimittbhvd apautvc caitat trayam animittam ucyate.


DESCRIPTIONS OF FEELING (VEDAN), IDEATION (SAJ)... 125

recognize it explicitly as [this is] matter. Therefore, the ideation having such an object
is to be considered an ideation whose object is not associated with a specific feature.26
As for the sphere of nirva, Sthiramati states that it is characterized by the cessation
of all specific features of the conditioned. Thus, the ideation accompanying the person
absorbed in contemplation having this object is an ideation which is not associated with
a specific feature. In the case of the peak of existence it is, according to Sthiramati,
due to its lack of clarity that it is not associated with specific features. The ideation of
someone who has entered this state does not conceptualize the object. Therefore, this
ideation is the one whose object is not associated with specific features.27 This condition
is compared with a thin woman who is described as having no waist. This state is not
referred to as not associated with specific features in the sense that there is no specific
feature at all.28 Otherwise, Sthiramati explains, it would follow wrongly that there is no
ideation at all (because saj has been defined as the grasping of specific features).
If there was no specific feature at all their grasping would be completely impossible.29
In the case of the thin woman the fact that she lacks the specific feature consisting in
a (fat) waist (and in this sense is animitta) makes others conceptualize her as a thin
woman. Except for the example of the thin woman, the explanations of the object not
associated with specific features found in the Pacaskandhakavibh closely resemble
the parallel definition found in the Abhidharmasamuccaya.30

4. The Unconditioned (asaskta)

The passages dealing with the category the unconditioned show remarkable
divergences in the Abhidharma works under discussion. While the Abhidharmakoabhya

26 PSkV 17b1f.: avyutpannaabdrthasambandhasya hi rpasvarpa eva saj bhavati, na tu rpam iti / atas

tadviayasaj animittlamban.
27 PSkV 17b2-4: nirvadhtur api sarvasasktanimittapratyastamitasvarpa iti tadlambanenpi samdhin

samprayukt saj animittlamban / bhavgram apautvd animittam / tatsampannasya sajlambana na


nimittkarotty animittlamban.
28 PSkV 17b4f.: [...] anudarkanynyyena (read: anudarakanynyyena), na punas tatra nimitta naivstty

animittam.
29 PSkV 17b5: anyath sajbhvaprasaga iti, nimittbhve nimittodgrahabhvt.
30 ASBh 4,11-4,14: avyavahrakualasyikitabhatay rpe saj bhavati na tu rpam iti / tasmd

animittasajety ucyate / animittadhtusampannasya rpdisarvanimittpagate nimitte nirve saj


animittasaj / bhavgrasampannasypautvenlabannimittkarad animittasaj (Someone who is not
experienced in conventional language because he has not learnt it has an ideation with reference to matter but does
not [conceptualize] it as [this is] matter. Therefore, this is called an ideation [of an object] not associated with
a specific feature. Someone who has attained the sphere without specific features has the ideation with respect to
the nirva as being without specific features because all the nimittas, like matter etc., are gone. [This ideation]
is an ideation [of an object] not associated with a specific feature. Someone having attained the peak of existence
has an ideation [of an object] not associated with a specific feature because it conceptualizes the object in an
unclear way.).
126 JOWITA KRAMER

enumerates three unconditioned categories, the Abhidharmasamuccaya and the


Pacaskandhaka(vibh) mention eight and four, respectively:
Abhidharmakoabhya31
1. space (ka)
2. cessation not [obtained through] consideration (apratisakhynirodha)
3. cessation [obtained through] consideration (pratisakhynirodha)

Abhidharmasamuccaya32
1. true reality of beneficial factors (kualadharmatathat)
2. true reality of non-beneficial factors (akualadharmatathat)
3. true reality of neutral factors (avyktadharmatathat)
4. space (ka)
5. cessation not [obtained through] consideration (apratisakhynirodha)
6. cessation [obtained through] consideration (pratisakhynirodha)
7. the state of motionlessness (nijya)
8. cessation of ideations and feelings (sajvedayitanirodha)

Pacaskandhaka33
1. space (ka)
2. cessation not [obtained through] consideration (apratisakhynirodha)
3. cessation [obtained through] consideration (pratisakhynirodha)
4. true reality (tathat)
The last two categories listed in the Abhidharmasamuccaya, that is the nijya
and the sajvedayitanirodha, correspond to the equipoise of non-conception
(asajisampatti) and equipoise of cessation (nirodhasampatti) respectively.34 It is
remarkable that these two entities are subsumed under the category the unconditioned
after they have already been defined as belonging to the factors dissociated from mind
(cittaviprayukt saskr) and thus as obviously being part of the conditioned.
A list of eight unconditioned factors identical with the list provided in the
Abhidharmasamuccaya appears in the Yogcrabhmi.35 This enumeration seems to be
related to a group of unconditioned factors ascribed to the Mahsakas (sa ston pai sde
pa) in the *Samayabhedoparacanacakra. According to this source the latter state that
the following nine factors are asaskta: cessation [obtained through] consideration,
cessation not [obtained through] consideration, space, the state of motionlessness, true
reality of beneficial factors, of unbeneficial factors, and of neutral factors, true reality of

31 AKBh 3,16-19.
32 AST 62a7f.
33 PSk 18,12f. and PSkV 61b4f.
34 See AS 18,23-27 and AS 62b6-63a1.
T
35 Y 69,4-6.
DESCRIPTIONS OF FEELING (VEDAN), IDEATION (SAJ)... 127

the path, and true reality of the conditioned arising.36 In the following, the definitions of
the four unconditioned factors mentioned in the Pacaskandhaka are investigated in more
detail and compared with the corresponding explanations in the other Abhidharma works.

4.1 Space (ka)

The Pacaskandhaka defines space as the one that [gives] room for matter (PSk
18,14: yo rpvaka). The explanation found in the Abhidharmasamuccaya is similar, but
at least the wording does not appear to be directly related to that of the Pacaskandhaka:
it is the absence of matter because it [gives] room for all kinds of activities.37 The
terminology used in the Abhidharmakoabhya to characterize space seems to differ
from both the Pacaskandhaka and the Abhidharmasamuccaya: Space is that which does
not hinder. Space has for its nature not hindering [matter]. It is where matter spreads.38
Later Vasubandhu adds the Sautrntika view on space: Space is the mere absence of
touchable things: for instance, when [people], in the dark, do not touch [anything that
possesses] resistance they say [there is] space.39 In his commentary on the four basic
elements in the Pacaskandhaka, Sthiramati emphasizes that space is not to be regarded
as a fifth basic element (mahbhta):40

That which is called space is nothing else than the mere absence of
impenetrable matter. An impenetrable thing makes room in its own location
for [another] impenetrable [thing] when it moves away from it. Space is
not [like this], because space [can]not move from a particular place in
the same way as matter [does]. Thus, it is not possible to say that [space]
makes room. Impenetrable matter prevents the arising of other matter
at its own place. Space [can]not [do this]. Therefore, there is no room
where there is [matter], and there is room where it is not there. Thus, the
departure [of matter] from a certain place is [what constitutes the action
36 P5639, fol. 175b8-176a2: dus ma byas kyi dngos po dgu ste / so sor brtags pa ma yin pai gog pa dang/
so sor brtags pai gog pa dang / (<so sor brtags pai gog pa dang /> em.; DP om.) nam mkha dang / mi g.yo
ba dang / dge bai chos rnams kyi de bzhin nyid dang (D nyid dang; P gshegs pa) / mi dge bai chos rnams kyi
de bzhin nyid dang / lung du mi ston pai chos rnams kyi de bzhin nyid dang / lam gyi de bzhin nyid dang / rten
cing brel bar byung bai de bzhin nyid do. Remarkably, four of the five kinds of tathat (excluding the true
reality of neutral factors) are also mentioned in the Pacaviatishasrikprajpramit (P5188, e.g., fols. 221a
and b, 258b, and 265a).
37 AS 62b5: gzugs med cing byed pa thams cad kyi go byed pai phyir ro.
T
38 AKBh 3,22f.: tatrkam anvti / anvaraasvabhvam ka yatra rpasya gati.
39 AKBh, 92,4f.: spraavybhvamtram kam / tadyath hy andhakre pratightam avindanta kam ity

hu. See also Kritzer 2005, p. 117, where a description of space found in the Vinicayasagraha is mentioned
which also brings forward the idea that space is not a real entity but only an expression.
40 In the section of the Pacaskandhakavibh in which the four unconditioned factors are explained only

a very short commentary on space is included: At which place impenetrable matter is not obstructed this is its
room. Therefore, it is [called] space because it gives room for material entities. (PSkV 61b6: sapratigha rpa
yatra pradee na pratihanyate, sa tasyvaka / ato rpi bhvnm avakadnd kam).
128 JOWITA KRAMER

of] making room by this [matter]; it is not [providing room] by space.


Therefore, space is not a basic element in the same way as earth etc. is.41

4.2 Two Kinds of Cessation

The second unconditioned factor mentioned in the Pacaskandhaka is


apratisakhynirodha, cessation not [obtained through] consideration. This and the
following category, the cessation [obtained through] consideration (pratisakhynirodha)
seem to be concepts that are of particular importance mainly within the framework of
the theory that all factors exist (sarvam asti) on the three time levels of present, past and
future. If one assumes that all future possibilities are existent, those of them that are not
to become present in the end have to be blocked somehow. This condition is achieved
through the application of the two entities pratisakhy- and apratisakhynirodha, both
of which are capable of preventing the arising of a future factor. In the case of the first
entity, the non-arising (of a contaminated factor) is attained by means of an antidote
produced in the personal continuum of a person, and of the insight related to it. The
Pacaskandhaka explains the pratisakhynirodha as that which is a cessation and that
[which] is a separation. It is the permanent non-arising of the constituents [caused] by the
antidote against a contamination.42 In the definition found in the Abhidharmakoabhya
the meaning of the term pratisakhy is specified in more detail:

Cessation [obtained through] consideration is a separation. Cessation


[obtained through] consideration is a separation from impure factors. The
analysis [or] consideration of the [four] noble truths of suffering etc. is
a particular insight (praj); the cessation obtained through this [insight]
is the cessation [obtained through] consideration. [The phrase obtained
through is not mentioned], because the middle word is elided as in the
expression ox-cart [used instead of ox-drawn-cart].43

Vasubandhu then goes on to explain in the Abhidharmakoabhya that there is more


than one pratisakhynirodha:

41 PSkV 4a4-b1: na hy ka nma kicid asty anyatra sapratigharpbhvamtrt / avaka ca saprati

gham eva vastu svasminn avake sapratighasyaiva tasmd apasara dadti, nkam / na hi tasmt praded
rpavad kam apastam iti tasyvakadna na yujyate / sapratigha ca rpa svadee nyarpasyotpatti
pratibadhnti nkam / tasmd yatra yasya yasmin saty avakbhva, yasminn asati tasya tatrvaka iti
tenaiva tasmt sthnd apasarat tasyvako datta, nkeneti / tasmn nka pthivydivan mahbhtam.
42 PSk 19,4f.: yo nirodha / sa ca visayoga / sa puna kleapratipakea skandhnm atyantam anutpda.

The first part of this definition is almost identical with the explanation of pratisakhynirodha found in AST 62b6:
gang gog la de ni bral bao.
43 AKBh 3,24-4,2: pratisakhynirodho yo visayoga / ya ssravair dharmair visayoga sa

pratisakhynirodha / dukhdnm ryasatyn pratisakhyna pratisakhy prajvieas tena prpyo


nirodha pratisakhynirodha / madhyapadalopd gorathavat.
DESCRIPTIONS OF FEELING (VEDAN), IDEATION (SAJ)... 129

Is there only one cessation [obtained through] consideration of all impure


factors? No. Instead, each [separation occurs] separately. The objects of
separation are as many as the objects of conjunction. If it were otherwise,
experiencing the cessation of the defilement which is abandoned by seeing
the [truth of] suffering would result in experiencing [at the same time] the
cessation of all defilements. It would be useless in this way to cultivate
the remaining antidotes.44

Sthiramatis comments on the definition of pratisakhynirodha offered in the


Pacaskandhaka are partly related to this explanation from the Abhidharmakoabhya.
Sthiramati also states that consideration is a particular insight (prajviea), adding
that this insight belongs to the uninterrupted path.45 He then continues by stating that

[the cessation obtained through consideration] is obtained by means of


consideration. [The phrase obtained is not mentioned] because the
middle word is elided. [As for the phrase] which is a cessation, [it
could lead to] the wrong consequence that it [is to be applied to] all
cessation. Therefore it is said: and that [which] is a separation. In this
way [this cessation] is distinguished from the other four cessations, since
they are not characterized as separation. [This] separation indicates [the
definition:] it is the permanent non-arising of the constituents [caused]
by the antidote against a contamination.46

Sthiramati also explains the way in which the contaminations are hindered from
arising one by one (and not all in a single moment):

This basis [of personal existence] arises accompanied by all seeds of


contaminations belonging to [the world of] the three spheres. Then,
when a certain moment of the path, which is an antidote for a particular
[contamination], appears on this path, this moment of the basis is not
capable of producing another moment of the immediately following
basis that would carry on the seed of the contamination that is being
abandoned by this [very moment of the path], because it is the condition
that hinders the arising of this [following moment of the basis]. However,

44 AKBh 4,3-7: ki punar eka eva sarve ssrav dharm pratisakhynirodha / nety ha / ki
tarhi / pthak pthak / yvanti hi sayogadravyi, tvanti visayogadravyi / anyath hi dukhadarana-
heyakleanirodhasktkarat sarvakleanirodhasktkriy prasajyeta / sati caiva eapratipakabhvan-
vaiyarthya syt.
45 PSkV 63a1: [...] nantaryamrgasaghta prajviea ity artha.
46 PSkV 63a1-3: pratisakhyay prpyo nirodho madhyapadalopt / yo nirodha iti sarvanirodhaprasaga ity ha

/ sa ca visayoga iti / eva hi eebhya caturbhyo nirodhebhyo vyavacchinno bhavati, tem avisayogtmakatvt
/ visayogam eva nirpayati / sa punar ya kleapratipakea skandhnm atyantam anutpda iti.
130 JOWITA KRAMER

[this moment] only becomes the cause of that state which is characterized
by the separation from the seed of a contamination that is being abandoned
by this [very antidote] since [other] conditions that are compatible with
the arising of those [seeds that are not abandoned by this moment of
the path] are [still] existent. It is to be understood that this moment of
the path is the condition that is incompatible with the arising of another
moment that would carry on the seed of the contamination that is being
abandoned by this [very moment of the path]. It is[, however,] compatible
with other [contaminations]. In this way, when the antidote for the smallest
of the small of contaminations of the kind that is to be abandoned through
[repeated] cultivation arises, this basis [of existence] becomes the cause
for another moment that is characterized by the separation from all seeds
of contaminations belonging to [the world of] the three spheres that are
to be abandoned through an insight or [repeated] cultivation. In this way,
is achieved the permanent non-arising [caused] by the antidote against
acontamination of which the seed has been removed in the sense that
the basis for the contamination does not appear and [the permanent
non-arising] of the factors related to it. And this is the cessation [obtained
through] consideration, [which] is described as separation.47

In the case of the apratisakhynirodha, the future factor does not arise due to other
reasons than pratisakhy, namely because of an insufficiency of causes.48 This situation
occurs, for instance, if a potentially perceivable object is not grasped by a sense perception
because the respective sense faculty is occupied with another object. In this case, the first
object is lost as a condition for the arising of its perception in the next future moment.
This is because it cannot be perceived anymore as soon as it passes from the present to
the past. The apratisakhynirodha apparently is the entity that blocks the arising of the
(potentially possible) perception of this object that did not get into the scope of the sense
faculty in the present moment and therefore did not become an appropriate condition for
the emergence of its perception in the future moment. This example of the functioning of

47 PSkV 63a3-b3: aya hy rayo niravaeatraidhtukakleabjnusyta pravartate / tatra yasya


yasya yo yo mrgakaa pratipaka, tasmin mrga utpanne sa rayakaas tatpraheyakleabjnugatam
anantarasyrayasya kantaram utpdayitu na aknoti, tadutpattiviruddhapratyayasnnidhyt / ki tarhi
tatpraheyakleabjavyvttytmakasyaiva kraa bhavati, tadutpattyanuguapratyayasadbhvt / sa eva mrgakaas
tatpraheyakleabjnugatasya kantarasyotpattaye vigua pratyaya, itarasynuguo veditavya / eva yvad
bhvanheyasya mdumdo kleaprakrasya pratipaka utpanne sa rayo niravaeatraidhtukadaranabhvan
prahtavyakleabjavyvttytmakasyaiva kantarasya krabhavati / eva yasya yasya kleasyraypravttito
bjam uddhta bhavati, tasya tasya tatsamprayuktn ca dharm kleapratipaketyantam anutpda
prpto bhavati / sa ca pratisakhynirodho visayoga cety ucyate.
48 AKBh 4,12: na hy asau pratisakhyay labhyate / ki tarhi / pratyayavaikalyt. See also PSkV 62a3: The

permanent non-arising of future factors due to an insufficiency of causes this is the cessation not [obtained through]
consideration (pratyayavaikalyd angatn dharm yo tyantam anutpda, so pratisakhynirodha).
DESCRIPTIONS OF FEELING (VEDAN), IDEATION (SAJ)... 131

the apratisakhynirodha appears in the Abhidharmakoabhya and is also mentioned


in Sthiramatis Pacaskandhakavibh. The Abhidharmakoabhya explains:

For example, for someone whose visual faculty and the mental organ
are occupied with one particular visible object, [other] visible objects,
sounds, odours, tastes and tangibles pass [from the present into the past].
[Therefore] the five kinds of perception which have the latter as their
objects cannot arise, since [these perceptions] are not able to grasp their
object when it is past. There is thus a cessation of these [perceptions],
which is not [obtained through] consideration [but] due to the insufficiency
of the cause [of arising].49

Sthiramati seems to have the same condition in mind (and to draw on the passage
in the Abhidharmakoabhya) when he explains:

For example, if for someone whose eye consciousness is engaged in


[perceiving] one particular kind of a visible object other objects and other
kinds of visible [entities] occur, then the five kinds of sense perceptions
that have [these other entities] as their objects cannot arise. [This is]
because, as for the present [entities], they cannot become immediately
preceding conditions and, as for the past [entities], they cannot become
objective conditions (for the present perceptions). Therefore, there is an
apratisakhynirodha of these [perceptions].50

In their definition of apratisakhynirodha the Pacaskandhaka and the


Abhidharmasamuccaya explain this category as that which is a cessation but not
aseparation, the first text adding that it is the permanent non-arising of the constituents
without the antidote against contaminations (klea).51 The Abhidharmasamuccayabhya
comments that this cessation is not a separation, because it does not eradicate evil

49 AKBh 4,12-15: yathaikarpavysaktacakurmanaso yni rpi abdagandharasaspraavyni ctyayante,


tadlambanai pacabhir vijnakyair na akya punar utpattum / na hi te saty atta viayam lambayitum iti
/ ata sa tem apratisakhynirodha pratyayavaikalyt prpyate. Another example for apratisakhynirodha is
given in the second chapter of the Abhidharmakoabhya: The non-arising [of factors obtained] entirely without
consideration, [merely] due to the absence of conditions this is what is called cessation not [obtained through]
consideration, as for instance [the non-arising] of a remainder of [life in] a [certain] homogeneous existential
class in case one dies in between (AKBh 92,7f.: vinaiva pratisakhyay pratyayavaikalyd anutpdo ya so
pratisakhynirodha / tadyath nikyasabhgaeasyntar marae).
50 PSkV 62b5f.: tadyathaikarpaprakravys aktacakurvijnasya yni viayntary utpattimanti

rpaprakrntari ca, tadlambanai pacabhir vijnakyair na akyam utpattum, vartamneu


samanantarapratyaybhvt, attev lambanapratyaybhvt / atas tem apratisakhynirodha prpyate.
51 PSk 19,1-3: yo nirodha / na ca visayoga / sa punar yo vin kleapratipakea skandhnm atyantam

anutpda; AST 62b6: gang gog la bral ba ma yin pao.


132 JOWITA KRAMER

propensities (anuaya).52 The Abhidharmakoabhya states that [it is] a different


[type of] cessation, not [obtained through] consideration, which consists of the absolute
hindering of arising: cessation not [obtained through] consideration is a cessation that is
different from separation [and] which consists of the absolute hindering of the arising of
future dharmas.53 The statement that this cessation (i.e. non-arising) is not a separation
refers to the concept that blocking of the arising of a certain future factor, evidently
a contamination, does not mean that this (contaminated) factor has been previously
removed by the application of its antidote, and the person is permanently separated
from it. Instead, it only indicates that a future factor is prevented from becoming present
because the conditions necessary for its arising are not available in a certain moment. In
his comments on this passage of the Pacaskandhaka, Sthiramati adds that

as for [the phrase] a separation, if a seed exists its association with


future factors is determined because it is their [general] nature to arise.
[Their] permanent non-arising when their antidote has removed their seed
from its place is the [pratisakhy]nirodha. It is said that the permanent
non-arising of future factors, [which occurs] even though no previous
mental process towards the removal of the seed has been developed due
to an insufficiency of causes, is the cessation not [obtained through]
consideration.54

Further on in his commentary, Sthiramati analyses the meaning of the phrase [cessation
not obtained through consideration] is the permanent non-arising of the skandhas without
the antidote against contaminations. He explains that there is permanent non-arising of
the arhats personal constituents (skandha), even though he still has seeds of beneficial
and neutral skandha associated with another (i.e. a future) existence, and that one could
assume that this is also a cessation not obtained through consideration. Sthiramati objects
to this idea by stating that this [non-arising of beneficial and neutral skandhas associated
with a future existence] in spite of the presence of seeds of beneficial and neutral skandhas
is not obtained without the antidote against contaminations. Thus, if the contaminations
had not ceased due to a pratisakhy, the conditions for the reappearance of these seeds
(of beneficial and neutral factors) in a future existence would not have been removed.

52 ASBh 15,3: yo nirodho na ca visayoga ity anuaysamuddhtt.


53 AKBh 4,10-12: utpdtyantavighno nyo nirodho pratisakhyay // angatn dharmm
utpdasytyantavighnabhto visayogd yo nyo nirodha, so pratisakhynirodha.
54 PSkV 62a5f.: <visayoga iti> sayogo hy angatair dharmair bje sati tem utpattidharmatm updya

vyavasthpyate / tatpratipakea svrayt te bje poddhte yo tyantam anutpda <so nirodha / tad etad
ukta bhavati / [...?] abuddhiprvako pi pratyayavaikalyd angatn dharm yo tyantam anutpda, so
pratisakhynirodha>. (After anutpda a longer passage has been omitted in the manuscript, which is rendered
in PSkVT 57b as de gog pao // de ni di skad du sa bon bton la blo sngon du ma btang yang rkyen ma tshang
bas ma ongs pai chos rnams gtan mi skye ba gang yin pa de so sor brtags pa ma yin pai gog pao zhes bstan
par gyur ro.)
DESCRIPTIONS OF FEELING (VEDAN), IDEATION (SAJ)... 133

Therefore, according to Sthiramati, this is a pratisakhynirodha, not an apratisakhy


nirodha. Sthiramati concludes that in this way both the state of nirva with remainder
(sopadhiea) and nirva without remainder (nirupadhiea) are a pratisakhynirodha.55
The nirva with remainder corresponds to the state of the arhat having seeds of beneficial
and neutral skandhas, the one without a remainder is the cessation of the arhats skandhas
after his death. Finally, Sthiramati explains the single components of the phrase [cessation
not obtained through consideration] is the permanent non-arising of the skandhas without
the antidote against contaminations in more detail:

The antidote against contaminations is understood here as being the pure


path. As for permanent non-arising, the term permanent [is used] in
order to differentiate [this non-arising] from the cessation [which follows]
the arising [of factors] (i.e. impermanence) and from the one which is
the equipoise [of cessation] (nirodhasampatti). The [latter] is non-
arising, but it is not permanent because the mind and mental factors arise
again. Non-arising [is different] from cessation that is impermanence
(i.e. the cessation that follows the arising of a factor). Cessation that
is impermanence is [the cessation] of a present [factor]. However, [the
apratisakhynirodha] is not the non-arising of a present, but only of
a future [factor].56

The Abhidharmakoabhya includes some further remarks on the two kinds of


cessation which are not mentioned in the other texts. It explains, for instance, that there
are four possibilities of occurrence of pratisakhy- and apratisakhynirodha:
1. Factors of which only the pratisakhynirodha is obtained, namely the impure factors
which are past, present and certain to arise;
2. Factors of which only the apratisakhynirodha is obtained: pure conditioned factors
which are not destined to arise;
3. Factors of which both is obtained, pratisakhynirodha and apratisakhynirodha:
impure factors which are not destined to arise;

55 PSkV 62a6-b2: <arhato nikyasabhgntarasambaddhn kualvyktn skandhn saty api bje

tyantam anutpda> (omitted in the manuscript and rendered in PSkVT 57b as dgra bcom pa la ni ris gzhan
dang brel bai dge dang / lung du ma bstan pai phung po rnams kyi sa bon yod bzhin du gtan mi skye ba)
pratilabdha iti so py apratisakhynirodha prpnotty ata ha sa punar ya kleapratipakea vin skandh
nm atyantam anutpda iti / etad ukta bhavati aya hi saty api kualkualvykta-(read: kualvykta-)
skandhabjasadbhve na vin kleapratipakea labhyata iti pratisakhynirodha evyam, npratisakhynirodha
/ eva sopadhieo nirupadhiea ca nirvadhtu pratisakhynirodha ity ukta bhavati.
56 PSkV 62b2-4: kleapratipako trnsravo mrgo bhipreta / atyantam anutpda ity atyantagrahaam

upapattisampattinirodhavyavacchedrtham / sa hy anutpdo bhavati, na tv atyantam, puna cittacaittotpdasadbhvt


/ anutpda ity anityatnirodht / anityatnirodho hi vartamnasya, na ca vartamnasynutpdo sti, ki tarhy
angatasyaiveti.
134 JOWITA KRAMER

4. Factors of which neither pratisakhynirodha nor apratisakhynirodha is obtained:


pure factors which are past, present and destined to arise.57
This classification shows that pratisakhynirodha hinders only the occurrence of
impure factors. This is because it is not reasonable to assume that a pratisakhy, an
insight, hinders the arising of pure factors. In contrast, the apratisakhynirodha can
stop the coming into existence of both, pure and impure, entities (the impure only if they
are not destined to arise). In the first case the factors have either already arisen (being
past or present) or they are destined to arise (because their conditions are available).
Therefore, only a pratisakhynirodha achieved through the application of an antidote
can stop their future existence (that is, can prevent them from becoming present). An
apratisakhynirodha, which would be a non-arising due to the lack of conditions,
is impossible in this case. Factors that have already arisen or which are automatically
going to arise due to the existence of appropriate conditions cannot be hindered by an
apratisakhynirodha. The second category involves pure factors for which there are
no causes enabling them to arise.58 The impure factors referred to in the third category
are not destined to arise and therefore must have been removed before by means of an
antidote. Therefore, there is a pratisakhynirodha of their occurrence in the future.
As for their apratisakhynirodha, it probably occurs in the moments following their
pratisakhynirodha because then there are no causes anymore for their future arising.
In the last case, pratisakhynirodha is not possible because the factors are pure, and
apratisakhynirodha is not applicable since they have already arisen or are certain to
arise due to appropriate conditions.59
Other notable explanations found in the Abhidharmakoabhya with regard to the
two cessations are concerned with the presentation of divergent positions on this topic.
In this context Vasubandhu mentions that

another school says: The capacity of insight [is decisive] for the non-
arising of the evil propensities (anuaya). Therefore this [non-arising]
is cessation [obtained through] consideration. Cessation not [obtained
through] consideration is the non-arising of suffering, which is only an
insufficiency of evil propensities as conditions for the arising [of suffering]
and therefore not a capacity of insight.

57 AKBh 4,15-19: catukoika ctra bhavati / santi te dharm ye pratisakhynirodha eva labhyate / tad-
yathttapratyutpannotpattidharm ssravm / santi yem apratisakhynirodha eva / tadyathnutpatti-
dharmm ansravasasktnm / santi te yem ubhayam / tadyath ssravm anutpattidharmm / santi
ye nobhayam / tadyathttapratyutpannotpattidharmm ansravm iti.
58 AKVy 18,28-30 gives the example of someone abiding on one of the six stages consisting of angamya,

dhynntara, and the four dhynas, who in that moment possesses the apratisakhynirodha of the remaining
five stages.
59 For further explanations see AKVy 18,20-19,24.
DESCRIPTIONS OF FEELING (VEDAN), IDEATION (SAJ)... 135

Vasubandhu objects that this non-arising of suffering is not established without


consideration therefore it is cessation [obtained through] consideration. After this he
mentions another divergent viewpoint: Cessation not [obtained through] consideration
is the subsequent non-existence of an arisen [factor] due to its spontaneous destruction.
According to Vasubandhu, this hypothesis results in a cessation not [obtained through]
consideration which is impermanent since it is non-existent as long as the factor has not
perished.60 In the same chapter of the Abhidharmakoabhya the divergent viewpoint
of the Sautrntikas with regard to the nature of the unconditioned factors is discussed.
At first Vasubandhu presents the following view:

The nature of this [factor called cessation obtained through consideration]is


to be recognized only by the noble ones, each for himself. It is only
possible to say that it is a distinct real entity which is eternal and beneficial
and which is called separation from this [contamination] and cessation
[obtained through] consideration.
However, according to Vasubandhu, the Sautrntikas [say] that all these
unconditioned [factors] are not real entities. They are not distinct entities
like matter, feeling etc. [] Cessation [obtained through] consideration is
the cessation of the (future) arising of already produced evil propensities
[and] the non-arising of any other by reason of the force of consideration.
The non-arising independent of [the force of] consideration [and only]
due to an insufficiency of causes is the cessation not [obtained through]
consideration.61

4.3 Suchness (tathat)

The last unconditioned factor to be discussed here is tathat. Unsurprisingly, this


category is not mentioned in the Abhidharmakoabhya. As already mentioned above,
in the Abhidharmasamuccaya three kinds of the tathta are indicated: the suchness of
beneficial, of non-beneficial, and of neutral factors. Only the first, however, is described
in more detail, whereas the other two are simply said to be understood in a parallel way:

60 AKBh 92,8-13: nikyntary punar hu / anuaynm <an>utpattau (Tib. mi skye bar bya ba la) prajy

smarthyam ato sau pratisakhynirodha / yas tu puna dukhasynutpda sa utpdakranuayavaikalyd eveti


na tasmin prajy smarthyam asty ato sv apratisakhynirodha iti / so pi tu nntarea pratisakhy sidhyatti
pratisakhynirodha evsau / ya evotpannasya pacd abhva sa eva svarasanirodhd apratisakhynirodha ity
apare / asy tu kalpanym anityo pratisakhynirodha prpnoty avinae tadabhvt.
61 AKBh, 92,2-7: ryair eva tatsvabhva pratytmavedya / etvat tu akyate vaktu nitya kuala

csti dravyntaram / tadvisayoga cocyate pratisakhynirodha ceti / sarvam evsasktam adravyam iti
sautrntik / na hi tad rpavedandivad bhvntaram asti / ki tarhi / [] utpannnuayajanmanirodha prati-
sakhybalennyasynutpda pratisakhynirodha / vinaiva pratisakhyay pratyayavaikalyd anutpdo ya so
pratisakhynirodha. For another definition of apratisakhynirodha, provided in the Vinicayasagraha which
also emphasizes the concept that this category is not a real entity but only an expression, see Kritzer 2005, p.121.
136 JOWITA KRAMER

What is the suchness of beneficial factors? It is the twofold essencelessness


(nairtmya), emptiness (nyat), the [realm] not associated with specific
features (animitta), limit of existence (bhtakoi) and the ultimate
(paramrtha). It is also the true reality of the factors (dharmadhtu).
Why is suchness called suchness? Because it does not become anything
else. Why is it called emptiness? Because pollution does not occur [there].
Why is it called not associated with specific features? Because it is the
tranquility of specific features. Why is it called the limit of existence?
Because it is the object of the correct [insight]. Why is it called ultimate?
Because it is the object of true knowledge of the noble ones. Why is it
the true reality of the factors? Because it is the cause of all factors of
the rvakas, pratyekabuddhas and buddhas. The suchness of the non-
beneficial and of the neutral factors is to be understood in the same way
as the suchness of beneficial factors.62

The definition offered in the PSk is far less extensive than the one in the
Abhidharmasamuccaya: [It is that] which is the true reality (dharmat) of factors, the
essencelessness of factors (dharmanairtmya).63 Notably, the Pacaskandhaka does not
mention the twofold essencelessness (but only the dharmanairtmya), nor the terms
nyat, animitta, bhtakoi, paramrtha, or dharmadhtu. Thus, we may assume that
Vasubandhu did not draw on the wording of the Abhidharmasamuccaya in this case.
In Sthiramatis commentary only the very beginning seems to rely on the explanation
given in the Abhidharmasamuccaya: It is suchness because it does not become anything
else.64 The remaining comments do not appear to be related to the Abhidharmasamuccaya.
Notably, Sthiramati does not mention the threefold classification of the tathat into the
suchness of beneficial factors and so on. Obviously this system of presenting the true
reality did not assert itself throughout the Yogcra tradition. Sthiramati mainly paraphrases
the single terms of Vasubandhus definition in his commentary:

Of factors means of matter, feeling, ideation, impulses, and consciousness.


True reality (dharmat) is the [true] being of factors, [their] true essence

62 AST 62a8-b5: chos dge ba rnams kyi de bzhin nyid gang zhe na / bdag med pa rnam pa gnyis dang / stong
pa nyid dang / mtshan ma med pa dang / yang dag pai mtha dang / don dam pa ste / chos kyi dbyings kyang de
yin no // cii phyir de bzhin nyid la de bzhin nyid ces bya zhe na / gzhan du mi gyur bai phyir ro // cii phyir
stong pa nyid ces bya zhe na / kun nas nyon mongs pa mi rgyu bai phyir ro // cii phyir mtshan ma med pa zhes
bya zhe na / mtshan ma nye bar zhi bai phyir ro // cii phyir yang dag pai mtha zhes bya zhe na / phyin ci
log med pai dmigs pa yin pai phyir ro // cii phyir don dam pa zhes bya zhe na / phags pai ye shes dam pai
spyod yul yin pai phyir ro // cii phyir chos kyi dbyings zhes bya zhe na / nyan thos dang / rang sangs rgyas
dang/ sangs rgyas kyi chos thams cad kyi rgyu yin pai phyir ro // chos dge ba rnams kyi de bzhin nyid ji lta ba
bzhin du chos mi dge ba rnams dang / lung du ma bstan pa rnams kyi de bzhin nyid kyang de bzhin (D kyang
de bzhin; P om.) du blta bar byao.
63 PSk 19,7f.: y dharm dharmat dharmanairtmyam.
64 PSkV 63b4: ananyathtvena tathat.
DESCRIPTIONS OF FEELING (VEDAN), IDEATION (SAJ)... 137

(ythtmya), [their] ultimate being (bhtaprakara). Since factors also have


the nature of being effects, being impermanence, and so on, therefore
[the root text says:] the essencelessness of factors. The characteristic
feature through which entities become objects of knowing (= mind) and
speech is their self. Therefore, they are free from a self because they do
not have that [self] as [their] intrinsic nature (svabhva). Due to this,
they are selfless (or essenceless).65

Sthiramatis interpretation of this essencelessness is of particular interest. He


understands it as the fact that the factors are empty of their imagined nature: [This
essenceless] state of these [factors] is the essencelessness of the factors (dharmanairtmya).
This is the emptiness of the factors of the imagined nature (kalpitasvabhvanyat).66

5. Conclusions

The passages of the Pacaskandhaka and its commentary, the Pacaskandhakavibh,


discussed in the present paper show parallels with the Abhidharmakoabhya, the
Abhidharmasamuccaya, and the Abhidharmasamuccayabhya. At the same time they
also include notable divergences from these other works. When defining the category
feeling (vedan), for instance, Vasubandhu restricts himself to the explanation of the
three kinds of pleasant, unpleasant, and neutral feelings. He does not mention any of
the other characterizations, as for instance physical and mental feeling or feeling
based on craving, described in the Abhidharmasamuccaya. As already suggested by
Sthiramati, a possible explanation for this omission could be that Vasubandhu tried to
give a very condensed presentation of the topic in his Pacaskandhaka, and therefore
limited himself to the most important information.67 Vasubandhus description of ideation
(saj) in the Pacaskandhaka also does not fully correspond to the definitions found
in the other texts under discussion. The version transmitted in the Sanskrit manuscript
available through the copy preserved in the China Tibetology Research Centre is
extremely short and, notably, does not contain the classification of saj into various
kinds (as found e.g. in the Abhidharmasamuccaya), like, for example, the ideation of
objects associated with specific features (sanimitta) etc. Three of these classes of saj

65 PSkV 63b4-6: dharmm iti rpavedansajsaskravijnnm / dharm bhvo dharmat,


ythtmya bhtaprakara / dharm krynityatdayo pi dharmat vidyanta ity ata ha dharmanairtmyam iti
/ dharm hi yena rpea jnbhidhnayor viaybhavanti, tat tem tm / tasmd tmano nirgat atatsvabhvatvd
iti nirtmna.
66 PSkV 63b6-64a1: tadbhvo dharmanairtmyam / tat punar bhvn kalpitasvabhvanyat.
67 See PSkV 16b5f.: iha sakepasya vivakitatvt sarvabhedn ca svarpd apthaktvd raydibhedena

nokta iti (Since [Vasubandhu] intended to give a concise [presentation] and because all the classifications [of
feeling] are not different with regard to their intrinsic nature, he did not discuss the classifications with respect to
[their] basis etc.).
138 JOWITA KRAMER

appear in the Tibetan translation of the Pacaskandhaka. Vasubandhus enumeration of


four unconditioned entities in the Pacaskandhaka is obviously a compromise between
the three unconditioned factors listed in the Abhidharmakoabhya on the one hand
and the eight entities as found in the Yogcrabhmi and in the Abhidharmasamuccaya
on the other. He probably adopted this group of three and adjusted it to the Yogcra
context of the Pacaskandhaka by adding tathat to it, a central philosophical concept
of the Yogcra tradition. However, Vasubandhu seems not to have relied directly on
the explanations of the Abhidharmasamuccaya when he defined the tathat in the
Pacaskandhaka. The definition of the unconditioned factor space (ka) is similar
in the works under discussion with regard to its contents. However, the wording does not
seem to be directly related. Of particular interest are the explanations of the two cessations,
pratisakhy- and apratisakhynirodha, which are very similar in the Pacaskandhaka
and the Abhidharmasamuccaya. In the Abhidharmakoabhya the discussion of these
categories, which prevent future entities from becoming present, is far more extensive.
This might be due to the fact that the two cessations are of crucial importance in the
context of the Sarvstivda theory that entities exist on all three levels of time. Remarkably,
Sthiramatis comments on this topic are also rather lengthy and include, for instance, the
notable remark that the nirva with remainder (sopadhiea) and the nirva without
remainder (nirupadhiea) are both to be classified as cessation [obtained through]
consideration.

Bibliography

Primary Sources

AKBh Abhidharmakoabhyam of Vasubandhu. Ed. Pralhad Pradhan. Patna: K.P. Jayaswal Research
Institute, 1967.
AKVy Sphurth Abhidharmakoavykhy of Yaomitra. Ed. Unrai Wogihara, Tokyo: Sankibo Buddhist
Bookstore, 1971 (reprint; first published in 19321936).
AS Fragments from the Abhidharmasamuccaya of Asaga. Ed. V.V. Gokhale in Journal of the
Bombay Branch of the Royal Asiatic Society 23, 1947, pp. 1338.
AST Abhidharmasamuccaya, Tibetan translation, P5550.
ASBh Abhidharmasamuccaya-bhyam. Ed. Nathmal Tatia. Patna: K.P. Jayaswal Research Institute, 1976.
D Derge
P Peking
PSk Pacaskandhaka. Ed. in: Li and Steinkellner.
PSkT Pacaskandhaka, Tibetan translation, P5560.
PSkV Pacaskandhakavibh. Ed. Jowita Kramer. Beijing/Vienna: China Tibetology Publishing House/
Austrian Academy of Sciences Press (forthcoming).
PSkVT Pacaskandhakavibh, Tibetan translation, P5567.
Y The Yogcrabhmi of crya Asaga. Ed. Vidhushekhara Bhattacharya, Calcutta: University of
Calcutta, 1957.
DESCRIPTIONS OF FEELING (VEDAN), IDEATION (SAJ)... 139

Secondary Sources

Boin-Webb, Sara. (2001). Abhidharmasamuccaya: The Compendium of the Higher Teaching


(Philosophy) by Asaga, Fremont: Asian Humanities Press (English translation of Walpola Rahulas French
translation).
Cox, Collett. (1995). Disputed Dharmas: Early Buddhist Theories on Existence, Tokyo: The International
Institute for Buddhist Studies.
Engle, Artemus. (2009). The Inner Science of Buddhist Practice, Ithaca: Snow Lion.
Kramer, Jowita. (2008). On Sthiramatis Pacaskandhakavibh: A Preliminary Survey. In: Nagoya
Studies in Indian Culture and Buddhism: Sabh 27, pp. 149171.
Kramer, Jowita. (forthcoming A). A Study of the Saskra Section of Vasubandhus Pacaskandhaka
with Reference to Its Commentary by Sthiramati. In: The Yogcrabhmi and Its Adaptation in India,
East Asia, and Tibet, Ulrich Timme Kragh (ed.), Harvard Oriental Series, Harvard University Press.
Kramer, Jowita. (forthcoming B). Indian Abhidharma Literature in Tibet: The Section on Vijna
in Sthiramatis Pacaskandhakavibh. In the Proceedings of the Conference Buddhism Across Asia:
Networks of Material, Intellectual and Cultural Exchange, (Singapore 2009), Tansen Sen (ed.), Institute
of Southeast Asian Studies.
Kritzer, Robert. (2005). Vasubandhu and the Yogcrabhmi: Yogcra Elements in the
Abhidharmakoabhya. Tokyo: The International Institute for Buddhist Studies.
Li, Xuezhu and Steinkellner, Ernst. (2008). Vasubandhus Pacaskandhaka, Beijing/Vienna: China
Tibetology Publishing House/Austrian Academy of Sciences Press.
Pruden, Leo M. (1988). Abhidharmakoabhyam, vol. 1, Berkeley: Asian Humanities Press (English
translation of Louis de La Valle Poussins French translation).
R O C Z N I K O R I E N T A L I S T Y C Z N Y, T. LXV, Z. 1, 2012, (s. 140148)

MAREK MEJOR

Tantric Exposition of the Dependent Origination according to the


Caamahroaatantra, Chapter XVI: prattyasamutpda-paala

Abstract

The Caamahroaatantra, or Tantra of Fierce and Greatly Wrathful One, belongs


to the class of Highest Tantras (anuttarayoga, rnal byor chen po bla med). The text
which has been preserved in the Sanskrit original and in Tibetan translation consists of
twenty five chapters (paala). The 16th chapter entitled prattyasamutpda-paala is an
exposition of the doctrine of dependent origination. The present author is preparing a
critical edition of this chapter from Sanskrit and Tibetan, provided with an annotated
translation. In this paper is offered a working translation alone with occasional references
the readings of the oldest Sanskrit palm leaf manuscripts, compared with the Tibetan
translation (Wanli edition).

Keywords: Buddhism, Tantra, doctrine of causality, Sanskrit manuscripts, Tibetan Kanjur

1. The Caamahroaatantra (henceforth abbreviated CMT), or Tantra of Fierce


and Greatly Wrathful One, belongs to the class of Highest Tantras (anuttarayoga, rnal
byor chen po bla med).1 According to the fourfold classification in Bu stons Catalogue
of Tantras (Rgyud bum gyi dkar chag), CMT is farther classified as belonging to the
Vairocana cycle (Vairocana-kula).2 CMT has been preserved in the Sanskrit original3 and
1 George 1974: xxxvi: According to formal Tibetan classification, this work is a Vykhytantra, or Explanatory

Tantra, belonging to the school of the Guhyasamja Tantra, which in turn is one of the five Mlatantras, or Basic
Tantras in the class of Anuttarayogatantras. See also Skorupski 1996.
2 Eimer 1989: 32: 2.3. rnam snang gi rigs gtso bor ston pai rgyud.
3 George 1974 described his source materials in his Introduction: pp. xlviiilvii - 17 Sanskrit Mss; p. lxivff.

- Tibetan blockprints; p. lxiii - Table 3: Chronology of primary and secondary affinity groups.
TANTRIC EXPOSITION OF THE DEPENDENT ORIGINATION ACCORDING TO... 141

in Tibetan translation.4 It consists of twenty five chapters (paala).5 First eight chapters
were edited and translated into English by George in 1974.6 The Sanskrit text of the
chapter sixteen entitled prattyasamutpda-paala was edited for the first time by La
Valle Poussin (LVP 1897) from the three manuscripts: 1. Cambridge Ms. Add. 1319,
42b-46a (Bendall Cat., p. 45) [= George: K]; 2. London Ms. 115, 39b-43a (fol. 41
missing) (Asiatic Society, Hodgson 46) [= George: A]; and 3. Paris (Soc. Asiatique), fol.
61a1-66a4 (Filliozat No. 13) [= George: M]. The basis of LVPs edition was the palm
leaf London Ms. (= George A), undated but placed at ca. 1380 A.D. (George 1974: 6
& n. 31). Re-edition of the text (with some omissions and corrections) is found in LVP
1913, pp. 125127. The Tibetan version from the Wanli Kanjur was published in Mejor
2010: 151153 (on pp. 154155 chapter division of CMT). The Sanskrit text is at places
obscure and LVPs edition requires a thorough revision in comparison with the Sanskrit
manuscripts and the Tibetan translation which differs here and there.7 The text is very
interesting since it contains a Tantric explanation of the so-called dependent origination
which is the fundamental doctrine of the Buddhists.
2. The content of the 16th chapter prattyasamutpda-paala is an exposition of the
doctrine of dependent origination given by Bhagavan, i.e. Caamahroaa (Fierce
and Greatly Wrathful), at the request of his spouse, Bhagavat (Queen of the Vajra-
Sphere).8 The exposition resembles in its form that of the Prattyasamutpdastra, i.e. it
is following the so-called di-vibhaga pattern, to which is added a general explanation
of the meaning.9 Thus the first part (= di, beginning) presents the standard formula
of the twelve-membered chain of dependent origination in its anuloma and pratiloma
sequence; the second part (= vibhaga, explanation) gives definitions of the following
members of the chain; and the third part offers an overall explanation of the meaning of
the exposition (artha), in which to the Abhidharmic presentation of the twelve members
is applied the theory of gandharva.10 This presentation of the prattyasamutpda has

4 Translated by Indian paita Ratnar and Tibetan lotsava Grags pa rgyal mtshan (11471216); the latter

also compiled a catalogue of Tantras (Harrison 1996: 77). See the biography of Grags pa rgyal mtshan at: http://
www.treasuryoflives.org/biographies/view/Drakpa-Gyeltsen/2915; cf. also http://tbrc.org/link/?RID=
P1614#library_person_Object-P1614. For the colophon cf. George 1974: lxivf.; cf. also Eimer 1989: 71; the complete
text of the colophon following Wanli Kanjur in Mejor 2010: 154f.
5 Listed in George 1974: xxxviiif.; Mejor 2010: 154.
6 George 1974 is the most comprehensive study of CMT to date; cf. some references in Dasgupta 1974 and

Snellgrove 2002.
7 Critical edition of the Sanskrit text of the 16th chapter, based mainly on the oldest palm leaf manuscripts from

London (Asiatic Society, Hodgson 46 = George A, no date, ?ca. 1380 A.D.) and Kathmandu (National Archives
A 994/4 = George B, date: 1427 A.D.), together with the Tibetan translation based on the several versions of the
Kanjur, is under preparation by the present author. The Padmavat-nma Caamahroaapajik dated from
1297 A.D. (Kathmandu, National Archives B 31/7 = George Comm; cf. also Isaacson 2008: 2 & n. 4.) does not
comment on the 16th chapter.
8 The name Caamahroaa is associated with iva, Snellgrove 2002: 153, 287.
9 Cf. Mejor 1994b.
10 Translated in part in Stablein 1980: 212f. and Filliozat 1971: 146f. (Filliozat wrongly refers to chapter XVII

instead of XVI).
142 MAREK MEJOR

much in common with the Sarvstivdin-Vaibhika exposition found in Vasubandhus


Abhidharmakoa, since it combines the Abhidharmic in their form definitions and the
theory of gandharva. 11
3. The peculiar fusion of the theory of antarbhava and gandharva and the theory
of the twelvefold dependent origination is described in Vasubandhus Abhidharmakoa12
(AKBh ad III.15a-b; III.19), and later it is found also in the 16th chapter of the
Caamahroaatantra.13 Its earliest application, however, is found in the Pli canon.
3.1. In the canonical Pli Mahtahsakhayasutta (Majjh. N. 38 = PTS ed., I:
256271) there are mentioned four kinds of nourishment (hra) 14 which are acting for
the persistence of living beings and for taking up those beings which are seeking (re)
birth (MN I: 261). The four nourishments are, in their turn, conditioned by , originating
in , descended from , and produced from desire (tah-nidn, tah-samuday, tah-
jtik, tah-pabhav). Then into the causal sequence are included the members following
from tah to avijj (nos. 8-1 of the twelvefold chain of prattyasamutpda), and next
follow the regular anuloma and pratiloma formulas of the twelvefold prattyasamutpda.
However, what makes the text most interesting is juxtaposition of the theorem of dependent
origination and another explication of entanglement in the painful world, viz. the theory
of gandhabba (Skt. gandharva), an intermediate immaterial being, which presides over
conception (MN I: 265f.).15

11 See Mejor 1991: 96ff.


12 Cf. Schayer 1934: 62(466): In Abhidharmakoa, the celebrated treatise of Vasubandhu, we find an interesting
passage describing the reincarnation process in the following way: The Gandharva seeing from a distance its father
and mother united in the act of procreation, is overpowered by passion for its mother and hatred for its father,
when it is a male Gandharva, or with passion for its father and hatred for its mother, if it is a female Gandharva.
Under the influence of these conflicting sentiments, it loses his presence of mind and becomes affected by the
illusion that it takes itself an active part in the act, mixes with the secretions, and penetrates its mothers womb
as an embryo. It is clear, from this curious anticipation of Freudian theories, that the Gandharva is both a child
spirit and a lover of its mother.
13 Cf. LVP, Koa, III, p. 5 n. 4; LVP 1913: 125ff.; Filliozat 1971;.
14 The theory of four nourishments was explained in the Abhidharmakoa III.38d &ff. It is said that the world

of living beings subsists on food. There are four kinds of food: 1. material food (kavakra) [lit. by mouthful],
which is further divided into gross (audarika) and fine, subtle (skma); material food is in its nature the three
yatanas, i.e. odors (gandhyatana), tastes (rasyatana), and tangible things (spraavya); 2. contact (spara) is
what originates from the cooperation of the three: an organ, its object, and an appropriate class of consciousness;
3. mental cognition (or volition) (cetan) is a mental action (mnasa karman); 4. consciousness (vijna) is the
aggregate of consciousness (vijna-skandha).
15 Cf. Schayer 1934: 61: At the moment of death, the individual loses his psycho-physical apparatus and becomes

a so-called Gandharva, a separate being in the intermediary state, which sets forth into the world to look for the
womb of its future mother. As soon as it finds its proper parents proper meaning that they belong to the class
of beings in which it is to be born in accordance with its karmic destination it keeps ready (pratyupasthita)
and, on the occasion of its parents coition enters its mothers vulva. Schayer long ago drew attention to the fact
that the popular Buddhist theory of reincarnation (...) should be distinguished from its philosophical exposition
on the ground of the dharma-theory. (ibid., p. 61 n. 2).
TANTRIC EXPOSITION OF THE DEPENDENT ORIGINATION ACCORDING TO... 143

The Mahtahsakhayasutta describes the consecutive stages of conception and


development of an embryo, then birth, growing up, and sensual contact with the outer
world, which result in origination of the whole mass of suffering:

Conception (lit. entering into a womb, gabbhassakkanti)16 takes


place when the three [factors] join together, viz. when [the prospective]
mother and father are united, when the mother has a period, and when
the gandhabba is present.

When the child becomes fully grown up, in consequence of maturation of its senses
it amuses oneself, being possessed of, endowed with the five pleasures of the senses: it
amuses with the forms which come through the visual consciousness, with the sounds
which come through the auditory consciousness, etc. up to with the ideas which come
trough the mental consciousness. It becomes attached to the vision of pleasant forms and
becomes vexed with the unpleasant forms. In consequence of such experiences it cultivates
the pleasant feelings and becomes attached to them. This delight in feelings (vedan) is
called attachment (updna). Conditioned by attachment is becoming (bhava), conditioned
by becoming is birth (jti), conditioned by birth is old age and death (jarmaraa),
and also sorrow (soka), lamentation (parideva), pain (dukkha), grief (domanassa), and
despair (upysa). Such is the origin (samudaya) of this whole aggregate of suffering
(dukkha-khandha) alone (kevala).
4. Now the prattyasamutpdapaala of CMT continues the two trains of Buddhist
thought it combines the twelvefold chain with the theory of gandharva in the form of
a sermon of the di-cum-vibhaga type, in which it imitates the Prattyasamutpdastra.
The most interesting point is the peculiar active role played by avidy in the process of
conditioning of the subsequent members of the chain of prattyasamutpda.17
4.1. First it said that {i} ignorance (avidy) means nescience (ajna) of what is
to be abandoned and what is to be accepted (heyopadeya), a thought (citta) following
immediately the moment of death and having a weak form (dhandha-rpa)18, which
takes the figure of a body (rrkra bhavati).
4.2. Next after ignorance come {ii} karmic formations (saskra), which are threefold;
among these, (1) karmic formation of body (kyasaskra) is breathing in and out
(vsa-pravsa)19, (2) karmic formation of speech (vc) is reflection and judgement
(vitarka-vicra), and (3) karmic formation of mind (manas) is passion, hatred, and delusion
(rga-dvea-moha). Ignorance (avidy), united with these (ebhir yukt), is breathing in
and breathing out; it is reflecting (vitarkayati) and grasping a gross object of thought
(sthla ghti); it is making judgement (vicrayati) and grasping a subtle object of

16 Cf. Garbhvakrntistra quoted in AKBh, p. 24.10; Kritzer 2000.


17 Cf. also Mejor 1994a.
18 BHSD sub dhandha slow, weak, dull; Tib. nub pa to decay, decline. References to the Tibetan follow
the Wanli Kanjur, see Mejor 2010.
19 BHSD p. 110: m. dual or pl., breath.
144 MAREK MEJOR

thought (skma), respectively. Thus, it becomes attached passionately (anurakta), hostile


(dvia), and bewildered (mugdha).
4.3. After these comes {iii} consciousness (vijna), which is sixfold, viz. the
eye consciousness, ear consciousness, nose consciousness, tongue consciousness, body
consciousness, and mind consciousness. Ignorance (avidy), united with these, is seeing,
hearing, smelling, tasting, touching, and thinking, respectively.
4.4. From consciousness comes {iv} name and form (nma-rpa). Name means four
aggregates, viz. the aggregate of feeling (vedan-skandha), and the rest; form means
just material form (rpa rpam eva); thus, having collected and having put together
(piayitv20, sdom nas) the two, it is called name and form. It means that ignorance
(avidy) is maturing (pariamati)21 in the form of the five aggregates of grasping
(updna pacaskandharpea). Among these (1) feeling (vedan) is threefold, viz.
pleasant, unpleasant, neither-pleasant-nor-unpleasant22; (2) apprehension (saj) means
desire (abhila)23 which immediately follows grasping of the own form (svarpa) of the
objects (vastu)24; (3) formations (saskra) consist of mind and mental states (citta-caitta)
which are grasping the general and particular states (smnyavievasthghina) 25; (4)
six kinds of consciousness (vijna) were explained earlier; (5) matter (rpa) is in its
nature the four gross elements (bhta); (i) the earth (pthv) is heaviness (kakkaatva)26, (ii)
the water (pas) is fluidity (dravatva), flowing (abhisyanditatva)27; (iii) the fire (tejas) is
warmth (umatva)28, bringing to maturity (paripcanatva)29; (iv) the wind (vyu) is bending,
streching out, gently setting in motion (kucana-prsaraa-laghusamudraatva)30.

4.5. Then come {v} six bases of cognition (ayatana)31, viz. of the eye, ear, nose,
tongue, body, and mind. Ignorance (avidy), which is united with them, like previously,
is seeing, etc.

20 So read the oldest palm leaf Mss consulted by me (London = George A, Kathmandu = George B); LVP

viundhitv (sic).
21 BHSD p. 323: pariamana, change; ripening, maturing; Tib. yongs su gyur ba, completely change

[mature].
22 Skt. Mss A, B: sukh-dukhdukhsukh; LVP 1897, p. 468, l. 16f.: sukh dukhsukh; LVP 1913, p.126,

l. 9: dukh sukhdukhsukh. Tib. bde ba dang/ sdug bsngal dang / sdug bsngal yang ma yin bde ba yang ma
yin pa.
23 BHSD p. 56: n. (Skt. m.); Tib. reads mngon par brjod pa = abhilpa speech; announcement.
24 Skt. Mss vastun (LVP), read: vastun; Tib. dngos po rnams kyi rang bzhin gzung bai mthar mngon par

brjod pa.
25 Tib. spyii khyad par gyi gnas skabs la dzin pa can gyi sems dang sems las byung ba rnams.
26 Skt. Mss A, B: kakkaatva; LVP gurutva (sic); cf. BHSD sub kakkhaatva. Tib. lci ba nyid heavy,

weighty.
27 Mss A, B; LVP abhispanditvam (after Cambridge Ms [= George K], against Paris and London Mss); Tib.

mngon par dzag pa nyid dripping, trickling.


28 Tib. dro ba nyid.
29 BHSD p. 326: -na, ripening, bringing to maturity; Tib. om. equivalent.
30 Tib. bskum brgyad ba dang / yang ba dang gyo ba nyid.
31 BHSD p. 101: sense, organ of sense.
TANTRIC EXPOSITION OF THE DEPENDENT ORIGINATION ACCORDING TO... 145

4.6. From them comes {vi} contact (spara), i.e. attainment (sampatti)32 of the
forms, sounds, smells, tastes, tangibles, and the constituent elements (dharma-dhtu).
4.7. The text both Sanskrit and Tibetan omits the definition of {vii} feeling
(vedan) and proceeds to the next item.
4.8. Then comes {viii} desire (t), i.e. craving for pleasure (sukhbhila).
4.9. Next comes {ix} attachment (updna), i.e. an action leading to obtain that
(tatprpaka karma).33
4.10. Next comes {x} existence (bhava), i.e. entrance into mothers womb
(garbhapravea).34
4.11. Next follows {xi} birth (jti), i.e. a visible appearance (prakakarabhinipatti)35,
attachment, i.e. obtaining five aggregates.36
4.12. Then comes {xii} old age (jar), i.e. becoming old (purtanbhva)37, and {xii}
death (maraa), i.e. annihilation of the mind and mental elements (citta-caitta-nirodha).
4.13. Then when there is old age and death he is overcome by the thought38 of sorrow
(okkula)39: I have not been seeking40 liberation (mukti), so he is lamenting.41 And
due to oppression by disease and other calamities (vydhydyupadravatas) he becomes
miserable (dukhin). Thus, because of attaching (niyojand)42 it again and again to the
mind [he] becomes dejected. And being disspirited (durmanas), he becomes irritated
(upys bhavati)43 by any kind of misfortune (upadrava).
5. In the following third part of the chapter 16th the true meaning of the above is
explained (ayam artha).44 Here we find the application of the theory of gandharva to
the twelvefold chain of dependent origination (see above).

32 Mss. A, B: -sampatti, LVP: -samvarttaye (sic); Tib. yang dag par thob pa to acquire or gain or obtain

completely.
33 Ms. B; LVP: tata prpaka karma; Tib. de las thob pai las.
34 Cf. Kritzer 2000.
35 Tib. rab tu gsal bar byed pa mngon par rgyu thun (D: mthun) par byed pa.
36 This is supported by Tib. D; P has: mngal gyi phung po lnga thob po, *garbha pacaskandhalbha. Tib.

PD adds: de nas mngal nas skye bar gyur ro, Then [it] originates from the womb.
37 The whole sentence up to here is missing from the Tibetan.
38 LVP: jarmaraacitta yena; Mss. A, B: jarmarae citta yana (LVP 1897: 469 n. 1 reads: maraa

citta yana).
39 Tib. chi bai sems dang sems las byung ba gog pas / de nas rga shii sems mya ngan dang khrug pa med

par gyur ro / (When [his] thought of death and the mental factors are destroyed, then [his] thought of old age
and death does not become afflicted by suffering).
40 praveiteti paridevati (LVP 469, n. 2: sic Camb.); Ms. B: paryyeiteti paridevate; Ms. Tokyo No. 69

(= George D, date: 1677 A.D.) paryeit- . Cf. BHSD sub paryeati (1) seeks, searches for, strives after.
41 Tib. grol na bdag gis smre sngags don pa tshol bar mi gyur ro.
42 Ms. A: tad eva puna daurmanasi niyojand daurmanas bhavati; Ms. B: ...puna punar daurmmanasi

yojayana (sic)...; LVP 1897: 469, l. 4f.: tad eva punar manasi niyojand (n. 3: Paris, niyojayet; Camb. niyojana);
LVP 1913: 126, l.22f.: tad eva puna punar manasi niyojand daurmanasy bhavati.
43 Tib. yid mi bde ba can gyis kyang bar du gcod pas khrug pa can yin no. Cf. BHSD sub upysa irritation,

mental disturbance or perturbation.


44 The text is obscure and at the present stage of my research only a tentative translation is offered here; also

the critical apparatus is limited to a few notes. Some misreadings found in LVP 1897 have been corrected in LVP
146 MAREK MEJOR

5.1. An intermediate being (antarbhavasattva), having as its beginning ignorance


and the six bases of cognition as its end, abiding in one place only, is watching three
worlds. It sees a man and a woman being attached (anurakta) to each other. Then,
impelled by its actions committed in its former life (lives) it will be reborn in this life.
Having seen a man and a woman of its proper caste (tajjti)45 there originates in him
a great sensation (spara).
5.2. Now, if [the gandharva] is to become a man, he perceives himself in a male
form; he falls deeply in love with his mother-to-be, and becomes very much hostile
towards his father-to-be. Love and hatred (rgadveau) are the feelings of pleasure and
pain (sukhadukhavedane). Then, [the gandharva] is thinking: In which form shall
I make love with her? But due to the feelings of neither-pleasure-nor-pain he becomes
completely bewildered (vymugdha). Then, impelled by the wind of his former actions
(prvakarmavtaprerita), because of great desire, [the gandharva] resolves: I shall enjoy
carnally her (et rammi). Some other man would love my woman (kmayati),
having thus reflected, like a passage of a star (trsakramaa)46, having entered
through the head of his father-to-be, dwelling in his mind which is settled in semen
virile, [the gandharva] sees himself as loving his mother-to-be. He applies the cause of
joy (sukhakraam updadti), and then having become of equal taste with semen
virile (ukrea samarasbhya)47, because of affection caused by great desire, he comes
forth from the thunder (vajra) of his father through the avadhuti-vein (avadhti ni).
And through the vajra-dhatu-ishvari-vein (vajradhtvvar ni), which is placed in the
hollow of the lotus (padmasuirastha), he stays in mothers womb, in the rebirth-vein
(janmani). Then, departing drop by drop/by dripping (?) (karantaritavat)48 there
comes [new] existence (bhava). Gradually provided with the embryo stages of kalala,
arbuda, ghana, pe, and limbs, after nine or ten months he exits through the same way
by which he entered [the womb]; and thus he is born.
5.3. Or, in case [the intermediate being] would become a woman then it becomes
affected to its father-to-be, and hostile to its mother-to-be. Here, it perceives itself in
a form of a woman. Having entered through the head of its mother-to-be, and having
fallen down in the lotus (padme patitv), and having mixed [itself] with the semen
virile, it stays in her rebirth-vein. Then, as [described] before, it comes forth and is born.
5.4. Thus, due to ignorance and other [factors] the inhabitants of the world are born.
And the people are nothing but five aggregates (skandha). And these five aggregates are
painful, transmigratory. Those who are seeking liberation (mokrthin) have no business
with suffering (na ca dukhena kryam asti). Because of annihilation of ignorance there

1913. Cf. translation (from the Tibetan) in Stablein 1980: 212f.; (in part) Filliozat 1971: 164f.
45 Ms. B: tajjti; Tib. dei skye ba (ni).
46 Tib. sems skar mda bzhin du, like a thought-meteor. Cf. BHSD sub sakrama(a).
47 Tib. khu ba dang / ro cig tu gyur pas.
48 The passage is obscure to me; Tib. dzag pai sems bzhin du like trickling/dripping of the mind (?). (Mejor

2010: 153 l. 3: read dzag).


TANTRIC EXPOSITION OF THE DEPENDENT ORIGINATION ACCORDING TO... 147

is [also] a non-existence (abhva) of the five aggregates, emptiness (nyat)49, vanity


(tucchat)50. One who is seeking liberation has no business with vain things. Therefore
liberation (moka) is neither being (bhva) nor non-being (abhva).51

Bibliography

ABhK Abhidharmakoabhya
BHSD Edgerton, Franklin, Buddhist Hybrid Sanskrit Dictionary. New Haven 1953.
Dasgupta 1974 Dasgupta, Shashi Bhushan. 1974. An Introduction to Tantric Buddhism. 3rd edition.
University of Calcutta.
Eimer 1989 Eimer, Helmut. 1989. Der Tantra-Katalog des Bu ston im Vergleich mit der Abteilung
Tantra des tibetischen Kanjur. Studie, Textausgabe, Konkordanzen und Indices. Indica et Tibetica 17. Bonn.
Filliozat 1971 Filliozat, Jean. 1971. Le complexe dOedipe dans un tantra bouddhique, tudes
tibtaines ddies la mmoire de Marcelle Lalou. Paris, pp. 142148.
George 1974 George, C.S. 1974. The Caamahroaa-tantra, chapters IVIII. Sanskrit and Tibetan
texts with English translation. New Haven.
Harrison 1996 Harrison, Paul. 1996. A Brief History of the Tibetan bKa gyur. In: Tibetan
Literature. Studies in Genre. Edited by Jos I. Cabezn and Roger R. Jackson. Snow Lion Publications,
pp. 7094.
Isaacson 2008 Isaacson, Harunaga. 2008. Himalayan Encounter: The Teaching Lineage of the
Marmopadea, Manuscript Cultures, Newsletter 1, pp. 26.
Kritzer 2000 Kritzer, Robert. 2000. The Four Ways of Entering the Womb (garbhvakrnti),
Bukky Bunka (Buddhist Culture): Kysh Rykoku Tanki Daigaku 10, pp. 141.
LVP 1897 La Valle Poussin, Louis de. 1897. The Buddhist Wheel of Life from a New Source,
Journal of the Royal Asiatic Society, pp. 463470.
LVP 1913 La Valle Poussin, Louis de. 1913. Thorie des douze causes, Gand.
LVP, Koa La Valle Poussin, Louis de, LAbhidharmakoa de Vasubandhu. Traduction et annotations.
Nouvelle dition anastatique prsente par tienne Lamotte. Institute Belge des Hautes tudes Chinoises,
Bruxelles 1980. Mlanges chinois et bouddhiques, volume XVI. Tome I: Introduction, Chapitres 1 et 2;
Tome II: Chapitre 3; Tome III: Chapitre 4; Tome IV: Chapitres 5 et 6; Tome V: Chapitres 7, 8 et 9; Tome
VI: Fragments des kriks. Index. Additions.
Mejor 1991 Vasubandhus Abhidharmakoa and the Commentaries Preserved in the Tanjur. Alt- und
Neu-Indische Studien, hrsg. vom Institut fr Kultur und Geschichte Indiens und Tibets an der Universitt
Hamburg, 42. Franz Steiner Verlag, Stuttgart 1991.
Mejor 1994a Mejor, Marek. On the Formulation of the Prattyasamutpda: Some Observations
from Vasubandhus Prattyasamutpdavykhy, Proceedings of the International Seminar on Buddhist
Studies (Liw, 25 June 1994), Ed. by Agata Bareja-Starzyska and Marek Mejor, Studia Indologiczne 4,
1997, pp. 135149.
Mejor 1994b Mejor, Marek. On Vasubandhus Prattyasamutpdavykhy, Proceedings of the
International Seminar on Buddhist Studies (Liw, 25 June 1994), Ed. by Agata Bareja-Starzyska and
Marek Mejor, Studia Indologiczne 4, 1997, pp. 151161.
Mejor 2010 Mejor, Marek. 2010. Caamahroaatantra in: Mejor, Marek; Helman-Wany,
Agnieszka; Thupten Kunga Chashab, A Preliminary Report on the Wanli Kanjur Kept in the Jagiellonian
Library, Krakw. Studia Buddhica 1. Wydzia Orientalistyczny UW, Warszawa, pp. 151155.

49 Tib. stong pa nyid. Cf. BHSD s.v.


50 Tib. snying po med pa nyid. Cf. BHSD sub tucchaka.
51 The last part of the text is left here untranslated, the Tibetan differs from the Sanskrit.
148 MAREK MEJOR

Schayer 1934 Pre-Aryan Elements in Indian Buddhism, Bulletin International de lAcademie


Polonaise des Sciences et des Lettres, Krakw, Nos. 1-3: 55-65 [reprinted in: S. Schayer, O filozofowaniu
Hindusw. Artykuy zebrane. Ed. by M. Mejor. PWN, Warszawa 1988, pp. 459469].
Skorupski 1996 Skorupski, Tadeusz. 1996. The Canonical Tantras of the New Schools in: Tibetan
Literature. Studies in Genre. Edited by Jos I. Cabezn and Roger R. Jackson. Snow Lion Publications,
pp. 95110.
Snellgrove 2002 Snellgrove, David. 2002. Indo-Tibetan Buddhism: Indian Buddhists and their Tibetan
successors. Two vols. in one. Bangkok: Orchid Press (1st ed. London: Serindia, 1987).
Stablein 1980 Stablein, William. 1980. Medical Soteriology of Karma in the Buddhist Tantric
Tradition. In: Karma and Rebirth in Classical Indian Traditions. Edited by Wendy Doniger OFlaherty.
Univeristy of California Press, BerkeleyLos AngelesLondon, pp. 193216.
R O C Z N I K O R I E N T A L I S T Y C Z N Y, T. LXV, Z. 1, 2012, (s. 149157)

MIKOAJ MELANOWICZ

Some Aspects of the New History of Japanese Literature

Abstract

This article discusses the concept of a new textbook of the history of Japanese literature
commissioned by the Polish Scientific Publishers PWN in Warsaw, which was further
developed during the writing process. The purpose of the book had to be decided before
writing, as well as my thoughts on my earlier book on Japanese literature. To start, it was
necessary to decide on the division of the contents into periods, as well as the genres
and problems within a given epoch. It was also necessary to take into consideration the
scope of civilization information introduced into the history of literature, the degree of
dependence of Japanese literature on the literatures of neighboring countries, as well as
on European and American literatures. I call the aforementioned matters aspects points
of view on the ways these problems are dealt with in the contents of the publication, and
explain the contents comprised of the outlines of six epochs, with the onset of contacts
with China in the 6th century AD. Proceeding to the most complex period in the history
ofJapanese literature, i.e. the 19th and 20th century, I touch on problems of masculinity of
literature. In most of the textbooks authors and critics are men, hence I decided to examine
the role of female writers in the most important trends of literature in the 20th century.

Keywords: literature, Japanese, history of literature, textbook

The topic of this paper is the presentation of the concept of a new Polish textbook
for the study of the history of Japanese literature.1 To start with, it was necessary to

1 Melanowicz, M., Historia literatury japoskiej [The History of Japanese Literature]. The approximately 30

printed sheet book was commissioned by the Polish Scientific Publishers PWN in Warsaw, and published in May
2011.
150 MIKOAJ MELANOWICZ

decide on the division of the material into periods, as well as to specify the genres and
problems within a given epoch. The scope of the civilization information introduced into
the history of literature needed to be considered as well.2
Moreover, it was not possible to avoid presenting the degree of dependence of Japanese
literature on the literatures of neighbour countries, as well as the influence of such distant
literatures as European and American. I call the aforementioned matters aspects points
of view on the ways these problems are dealt with in the contents of a publication called
a textbook.
Historia literatury japoskiej (The History of Japanese Literature), which I have
recently written (commissioned by PWN Publishing House) is primarily addressed to
undergraduates of Japanese studies (Japanese philology, cultural and Oriental studies with
a Japanese specialization) attending classes on Japanese literature and culture. The book
is also a compendium for those who learn the Japanese language and who are interested
in Japanese culture in general.
It should be noted that Literatura japoska [Japanese Literature]3 was published in
the mid-1990s in three volumes. The first volume contained an outline of the historical
process, particularly the development of literary forms and genres up to the nineteenth
century. In the following volumes more focus was placed on the twentieth century: prose
was presented in the second volume, and poetry and theatre in the third. Hence, this was
not a coherent picture of the history of literature, as each of the three volumes was based
on a different pattern: the first one resembled the formula of a textbook for the study
of the history of literature from its beginning to the mid-nineteenth century, and more
precisely, to the moment when Tokugawas regime was overthrown. The second volume
contained presentations of novels and short stories from the beginning of the process of
civilization and cultural changes in Japan influenced by the West in the latter part of the
nineteenth century, up to the 1990s. The third volume serves as a kind of supplement
to the second. The picture of Japanese literature becoming modern, and seen through
the eyes of artistic prose, was enhanced with the history of the main trends in poetry
and drama, biographies of writers and their works, which were created and subjected
to identical social and political impulses as prose was at the end of the nineteenth and
throughout the twentieth centuries. Thus, they were part of the same, complex process.
Poems and theatrical plays of that time were often experimental and precursory to more
successful stylistic forms of prose expression. When previously writing about poetry and
drama, even though I did not intentionally ignore achievements in the field of novel,

2 I make a distinction between the two terms. I think of civilization, as the external sphere in the life of
communities, that is the institutions, legal, political and technological systems (according to It Shuntar); whereas
culture in this understanding is the internal sphere of human activity. Thus, it encompasses literature, philosophy,
religion as a belief, and other creations of the spirit.
3 Melanowicz, M., Literatura japoska Od VI do poowy XIX w. [Japanese Literature From the 6th to the

Mid-19th Cent.], Wydawnictwo Naukowe PWN, Warszawa 1994; Literatura japoska proza XX w. [Japanese
Literature 20th Century Prose], 1994; Literatura japoska poezja XX w., teatr XX w. [Japanese Literature 20th
Century Poetry and Theatre], 1996.
SOME ASPECTS OF THE NEW HISTORY OF JAPANESE LITERATURE 151

I did not make any attempts to simultaneously present all the genres and forms created
within one literary period or school. I rather made an artificial division into prose, poetry,
and drama (skipping criticism), despite the fact that many authors started their literary
careers writing poems, and subsequently concentrated on prose, and sometimes also
on drama.
In afterthought, I would say that this was not a coherent picture of the whole process
of literature creation. At that time I was not basing my thoughts on any of the already
classic (Aston, Florenz4), or later (Keene, Kat, Konishi5) textbooks. The layout of the
work and its particular implementation drew more from the experience I had gained during
my studies at Waseda University (19641966, 19711972) and from Japanese textbooks.6
At that time I was influenced most by a compendium entitled Nihon bungakushi no shid
to jissai (History of Japanese Literature guide and [didactic] practice)7, published by
Meiji Shoin, who specializes in, among other subjects, fundamental treatises on Japanese
literature.
The new History of Japanese Literature, on the other hand, was conceived as one
comprehensive text, comprised of the outlines of six epochs which make up the panorama
of Japanese literature, and takes into account the richness of forms and genres characteristic
of each epoch. In the presentation of the first epoch, i.e. antiquity (the Nara epoch,
8th cent.), poetry dominates, whereas in late antiquity (the Heian epoch, 9th12th cent.)
the main focus is on court tales. In the third part, devoted to the Middle Ages, emphasis
has been put on war epics and memoirs (the Kamakura epoch, 12th14th cent.), and in the
latter part, devoted to the late Middle Ages (the Muromachi epoch, 14th16th cent.) the
focus is on drama and the n theatre, as well as the popularization of these three genres
in modern times (the Edo epoch, 18th19th cent.). Owing to their rampant growth and
important educational and social roles in the twentieth century, novels and short stories
have been emphasized.
In the literary panorama of the early period most space has been devoted to
collective works, such as anthologies Manysh (Collection of Ten Thousand Leaves,
ca. 781A.D.), Kokinsh (The Anthology of [Japanese Poetry] Old and Contemporary, 905
A.D.), Shinkokinsh (The New Anthology of [Japanese Poetry] Old and Contemporary,

4 Aston W.G., A History of Japanese Literature, London 1899; Florenz K., Geschichte der japanischen Literatur,
2nd ed., Leipzig 1909.
5 Kat Shichi, A History of Japanese Literature, Paul Norbury Publications, 1979; Keene, D., Seeds in the

Heart. Japanese Literature from Earliest Times to the Sixteenth Century, New York 1993; ditto, World Within
Walls: Japanese Literature of the Pre-Modern Era, 16001867, London 1976; ditto, Dawn to the West. Japanese
Literature of Modern Era. Fiction, New York 1984; ditto, Dawn to the West. Japanese Literature of Modern Era.
Poetry, Drama, Criticism, Holt, Rinehart and Winston, New York 1984.
6 Hisamatsu Senichi, Yoshida Seiichi, Nihon bungakushi (History of Japanese Literature), Shibund, Tokyo

1977; Takagi Ichinosuke, Takeuchi Riz, Nihon bungaku no rekishi, Kadokawa Shoten, 1967 (received as a gift
from the film director shima Nagisa and his wife after their visit to Poland in 1968).
7 Authors of this non-standard textbook from 1965 (first edition in 1960) are: Asao Isoji, Ichiko Teiji, Gomi

Chiei, Hasegawa Izumi, Kodaka Toshir.


152 MIKOAJ MELANOWICZ

1205A.D.)8. Individual literary works have also been marked, such as Genji monogatari
(The Tale of Prince Genji, from 1008 A.D.) written by Lady Murasaki Shikibu, the
anonymous war story Heike monogatari (The Tale of the Heike Clan, ca. 1240 A.D.),
dramas of Zeami (14th15th cent.), novels of Ihara Saikaku (17th cent.) and the twentieth-
century novels of Natsume Sseki, Tanizaki Junichir, Abe Kb, e Kenzabur and
Murakami Haruki. Much space has also been given to the n theatre and drama, the
urban kabuki theatre and the puppet theatre bunraku, as well as to the haikai poetry
(haiku), as original symbolic forms, enriching the repertoire of the codes of Japanese
culture known in the world.
The factographic material has been put in order according to the chronology of
individual genres poetic, epic and dramaturgy. In this chronology like in the European
tradition the following epochs have been distinguished: antiquity, Middle Ages, and
modern times, as well as the contemporary period, that is the 20th century. These
universal terms, however, have local, (Japanese) meanings and conditioning. From the
Japanese perspective they do not belong to the same periods, and are not parallel to the
European ones. This means that the Japanese antiquity corresponds rather to the time
of the European Middle Ages. And so we can say that the Japanese have their own vision
of antiquity (kodai) and the Middle Ages (chsei), which they situate in different
times than the European epochs. That is why the description of Japanese literature and
culture has been preceded by a presentation of the principles according to which the
history of Japan has been divided.
Particularly serious problems arise when we try to use the names of intellectual and
literary trends and formations, like Renaissance, Baroque, Classicism, Enlightenment,
Romanticism, Realism or Positivism, which are also the names of cultural and literary
epochs in Europe. The use of each of those names is accompanied by a commentary which
usually limits the possibility of their free use in Japan. However, sometimes questions
are posed concerning the Japanese Renaissance or even Baroque. Such questions
and attempts at giving answers to them allow in a comparative perspective to bring
closer the possibility of comparing the names of epochs and periods, names which are
fundamental for putting the history of Japanese culture, including literature, into order.
It is important for understanding the dynamics of the development of literature to
pay attention to the distinctiveness of literary forms and figures which were born and
played an important part in shaping literary styles before the Europeanization of Japan.
Also important are the relations and comparisons with literatures of Japans geographic
neighbours, especially with China, which, for Japans early history, played a role similar to
that of Greece and Rome for the rest of Europe, although Japan has never been conquered
by China nor become its vassal. However, in the early period of shaping their statehood
the Japanese turned to China for their system of writing, architecture, arts, as well as the

8 In the translation of the titles of anthologies, the Japanese song (waka) has been replaced with the term

Japanese poetry, as more adequate, although in the Japanese tradition native poetry was called a song (uta,
waka).
SOME ASPECTS OF THE NEW HISTORY OF JAPANESE LITERATURE 153

continental religions and thought systems Buddhism, Taoism and Confucianism often
making fundamental changes in the process of their Japanization.
I have also tried to answer the questions concerning the relationship between
literature and philosophy, particularly because in ancient Japan there was no philosophical
discipline comparable to Greek philosophy. There is no doubt, however, that there existed
philosophical thought which originated in Japan, but it was contained mainly in Buddhist
and Confucian treatises, although belles-lettres were also a rich source for the expression
of the philosophy of life. This is because literature was an important tool of educating
the nation.

Japan in East Asia

The introductory reflections begin with the presentation of Japans geographical


position in East Asia, in order to explain that Japan had been shaping its statehood on
the Japanese Archipelago for several millennia. It had been developing in the vicinity
and in the sphere of influence of East Asian cultures, the centre of which was China,
or rather the countries known in ancient Japan under the names of dynasties, such as
Han, Sui, Tang, Sung (Song), Yuan (Mongolian dynasty), Ming, and Qing (Manchu
dynasty).
Due to the first contacts with China in the sixth century A.D., Japan began to embrace
the cultural and civilization values of the Han country (Kan in Sino-Japanese) sending
emissaries (kenzuishi) to Sui, and later to Tang (kentshi). The Japanese at that time
their country bore the name of Yamato were becoming acquainted with Chinese culture
through Korea, or rather the then kingdoms of Paekche and Silla, which they reached by
sea via the Tsushima islands. Emissaries, monks, students, and merchants were travelling
directly to China also from the ports of Hakata and Bnotsu on Kyushu. Trade contacts
were maintained with China also via the Ryukyu Kingdom.
The system of writing and Buddhism, the most important features of Chinese culture
(direct or indirect), reached Japan due to the early links between the Japanese and
the peoples of the Korean Peninsula. They enthusiastically adopted many of Chinese
achievements, however, they did not allow themselves to assimilate and quickly freed
themselves from direct Chinese influence.
In spite of the geographical proximity, the Japanese did not belong to the peoples of
Chinese origin. They also worshipped different gods. Apart from that, they had different
dwellings, clothes, and their own patterns of behaviour. They had been creating oral
literature for many centuries before adopting the writing system and Buddhism from Korea
and China. They also governed their country differently, which in the early historical
times occupied small territories around Naniwa (today Osaka), Nara, and Kyoto, and
later, in the Middle Ages, spread to almost the whole of the Japanese Archipelago. The
above mentioned differences between China and Japan existed not only in the distant
antiquity. They were also preserved in the period of assimilation of Chinese civilization
154 MIKOAJ MELANOWICZ

in the epochs of Yamato (Epoch of Burial Mounds), Nara and Heian, that is between the
sixth and the twelfth centuries, and later, especially from the eighteenth century.
Since the earliest times the Japanese have possessed the ability to discard or modify
foreign values. Such a theory can be posed only on the basis of the effects of cultural
development, since we do not have credible and sufficient sources to conduct significant
comparative studies of ancient Japanese-Chinese connections. However, proof of this theory
is found both in the development of Buddhism specific to Japan as well as in Japanese
literature, beginning with the poetic anthology Manysh (Collection of Ten Thousand
Leaves, 8th cent.) and the prose of the aristocrats of the tenth and eleventh centuries,
whose greatest masterpiece is Genji monogatari (The Tale of Prince Genji, 1008 A.D.).
In the following centuries the dominant trend of this literature began to reflect the
likes and the achievements of the ruling class, namely the military nobility, called bushi
(the samurai). This occurred after the victories of the Minamoto clan (12th cent.). The
epic testimony of those days became the Heike monogotari (The Tale of the Heike Clan,
?1240 A.D.). That time in Japanese historiography is also the beginning of the Middle
Ages, lasting till the end of the sixteenth century. Owing to the continuation of the
native aristocratic cultural patterns also in that epoch, as well as to the efforts aimed at
assimilating foreign patterns, the Japanese created original literature both in form and
in content, which became a great contribution to the treasury of world literature.

Division into periods and the outline of history

In this textbook the history of Japanese literature has been traditionally divided into
epochs whose names are identical with the names of geographic centres of politics and
culture. The epochs are as follows:
1 Yamato (6th7th cent.), during which we also distinguish the cultures of Asuka
(?552645 A.D.) and Hakuh (645710 A.D.), and in historiography this period is
also called the Epoch of Burial Mounds (precisely the archeological epoch kofun
jidai ca. 250538 A.D.);
2 Nara (8th cent.), during which we also distinguish the Tempy era (722748 A.D.);
3 Heian (9th12th cent.), the period of dominance of the Fujiwara clan;
4 Kamakura (13th14th cent.), the period of dominance of the Minamoto and Hj
clans;
5 Muromachi (14th16th cent.), the period of dominance of the Ashikaga dynasty shoguns;
6 Edo (17th19th cent.), the period of dominance of the Tokugawa clan shoguns;
7 The contemporary period, when the political and cultural centre is Tokyo. This period
is divided into eras of ruling emperors:
Meiji (18681912)
Taish (19121926)
Shwa (19261989)
Heisei (1989)
SOME ASPECTS OF THE NEW HISTORY OF JAPANESE LITERATURE 155

The division into periods following that of universal history has also been adopted:
1. antiquity (until the 12th cent.), Jap. kodai, jdai, chko;
2. Middle Ages (12th16th cent.), Jap. chsei;
3. modern times (16th19th cent.), Jap. kinsei;
4. contemporary period (since 1868) is divided into earlier (kindai, till the end of the
Second World War, 1945) and later part (gendai contemporary time, present
time or sengo post-war time).
In this outline I have included the majority of works previously highlighted in Japanese
textbooks and other compendiums of native literature. This means that I do not even
name all the works written on the Japanese Archipelago, which shows their great number.
Ipresent the evolution of genres and forms in the Japanese language, practically skipping
the literary output written in Chinese. In the initial part, the pre-writing era (prior to the
6th cent.), I discuss the oral epic forms: myths, legends, norito prayers, setsuwa tales, as
well as lyrical forms known as songs (uta, waka, kay). They have been preserved in
Kojiki, or The Book of Ancient Matters (Kojiki, 712 A.D.; Polish translation ed. 1986),
Nihonshoki, also known as Nihongi (Japanese Chronicle, 720 A.D.) and in Manysh
(Collection of Ten Thousand Leaves, after 759 A.D., around 781 A.D.).
I take into consideration some selected written monuments, but not literature, like codes
of criminal and administrative law (mi-ry, 689 A.D., and Taih-ry, 702 A.D.), and
the role played by emperors, as well as Korean and Chinese scribes, in the development
of literature. It was on the initiative of the ruling Empress Gemmei (708714 A.D.)
that the first chronicles and descriptions of land and customs were written down. Of
great significance to the development of literature was the construction of a permanent
capital city in the Yamato Valley and the move of the imperial court to the residence
in Heij-ky (lit. The Capital/City of Peace), in the city of Nara, which took place in
710 A.D. and began the era in which the courts and Buddhist culture flourished.
The official language in contacts with the continent was Chinese, however practical
use of this language was narrowed to the circle of officials and monks. In the beginning
the scribes were Korean and Chinese. Their Japanese disciples often distorted the rules
of the Chinese language and hence many sentences in the chronicles have features of
the courtiers native language. Out of many attempts at composing poems in Chinese,
a collection of 64 works entitled Kaifs (The Old Favourite Verses, 751 A.D.) from
the end of the 7th and beginning of the 8th century has been preserved which shows the
imitation of Chinese poets, rather than the authors invention.
Courtiers and literary men were educated according to Chinese models in several
types of schools, e.g. in the Department of Education of the Ministry of Ceremonies
in the specializations of history, literature, Confucianism, law and mathematics. The
Ommy Department (Department of Yin and Yang) was responsible for the education
of astrologers, fortune-tellers and calendar makers. Medical schools educated natural
medicine practitioners, magicians, herbalists and acupuncture specialists. Many monks
and officials went to study in China.
156 MIKOAJ MELANOWICZ

The works preserved from that period show the existence of a rich literature transmitted
orally, for example, by the storytellers (kataribe) on the Japanese Archipelago. The oldest
texts were written down with the use of Chinese characters in Japanese (Kojiki) and
Chinese (Nihongi). Also the native poetry was written down with Chinese characters in
Japanese. In antiquity (we may also say in the Japanese antiquity, in the time of the
Nara epoch) there was the development of poetry and chronicle prose. And in the Heian
epoch, it was the artistic prose that flourished short stories, tales and novels to an
extent not encountered anywhere else in the world.

Masculinity of literature

Proceeding to the most complex period in the history of Japanese literature, namely,
the end of the nineteenth and the twentieth century, I will mention the need to introduce
several innovations in the layout of material and placement of accents. First of all, in the
previous compendiums of Japanese literature men were given the dominant place. In the
twentieth century, we see men as representatives of naturalists (Tayama Katai, Shimazaki
Tson, Masamune Hakuch etc.), idealists (Arishma Takeo, Shiga Naoya, Mushanokji
Saneatsu), aesthetes (Tanizaki Junichir, Nagai Kaf, Sat Haruo) etc. I would like to
point out here that at the end of the nineteenth century (Ozaki Ky, Kda Rohan) there
is one female writer mentioned (Higuchi Ichiy) among pseudo-classics. This is proof
of the theory of masculinity of the most outstanding works of literature almost until
the end of the twentieth century.
Moreover, when we look through the authors of textbooks, we also very rarely
see names of women (apart from specialist works devoted to feminine literature).
Thus, we may also claim that it is men who decide about the contents of textbooks.
Of course, we cannot draw the simple conclusion that this is the reason for the definite
majority of men among the honoured artists. However, a person who is writing a new
textbook must pay more attention to this problem than it has been done before, and at
least should not multiply the textbook stereotypes, but make an effort to examine the
role of female writers in the most important trends of literature in the twentieth century.
I made such an attempt and concluded that there are gaps in this matter. That is why
Iincluded achapter entitled The Awakening of Women, in which I discuss the visible
activity of female writers at the end of the Meiji era and the beginning of Taish, i.e.
in the years 19101916. At that time a great variety of women writers appeared, their
names remaining in the shadow of men who were writing at the same time. Many of
these women do deserve to be remembered by the critics, particularly because it was
the time of debut for many women writers who became famous in mid-century and
who encouraged women to express their feelings artistically. Owing to a revolution in
education after the Second World War, many prominent female writers emerged, often
called the successors of Lady Murasaki, the author of the most famous Japanese story
entitled Genji monogatari (The Tale of Genji, 1008 A.D.).
SOME ASPECTS OF THE NEW HISTORY OF JAPANESE LITERATURE 157

Certainly, among aspects worth mentioning, we can include many other specialist
problems, such as the way poetry, drama and criticism have been presented, in order
to balance and appreciate in the textbook the significance of achievements of poets,
dramatists and critics. Worth considering is the explanation of the role played by the
Koreans and other national minorities writing in Japanese. I will, however, leave these
aspects for later examination.
R O C Z N I K O R I E N T A L I S T Y C Z N Y, T. LXV, Z. 1, 2012, (s. 158170)

NINA PAWLAK

Morphological Coding of Verb-Object Agreement


in African Languages

Abstract

The aim of the paper is to demonstrate the structures of some African languages that
mark agreement between the verb and the object. The means of coding agreement are rich
and differentiated and include pronominal affixes on verbs for objects but not for subjects,
verb endings and tonal contrast as well as some other modifications of verb when used
with object. Morphological devices tend to code for the grammatical properties of the
object, such as gender and number, or in languages with class systems the noun class
attributed to the object. Some other languages mark the agreement with either nominal
or pronominal form of the object. The languages may also denote a semantic distinction
between single and many. With reference to marking properties of semantic Patient
rather than syntactic object, it is postulated to relate the presented system of marking
verb-object agreement as traces of an ergative concord system in African languages.

Keywords: linguistics, African languages, morphology, grammar

Introduction

In typological picture of worlds languages, African languages are distinguished


mainly for their unique sound systems and morphophonological processes involving
tones.1 Linguistic descriptions of particular languages present these features as inter-
systemic devices, but also in their cross-linguistic perspective. Less attention has been
paid to syntactic patterns that organize the structure of the sentence and to their coding
means that express the relations between the parts of the sentence. The paper deals with
1 Cf. Heine, Nurse 2000; Clements, Rialland 2008.
MORPHOLOGICAL CODING OF VERB-OBJECT AGREEMENT IN AFRICAN LANGUAGES 159

rules of clause composition that have a form of cross-reference between different parts
of asentence and affect verb. Termed as agreement, in many languages they are related
to marking subject with the verb. In languages where the verb is inflected, subject-verb
agreement is manifested by verb affixes indicating the person, gender and/or number
of the subject. European languages commonly mark agreement of this kind with verb.
English shows distinctive marking only in the third person singular, present tense form
of verbs. The agreement is marked by adding -s.

English: The student copies the notes


The students copy the notes

Polish has some distinctive agreement markers also in the other persons and in other
tenses.

Polish: Wiatr powali drzewo/drzewa


The wind struck down the tree(s)

Wichura powalia drzewo/drzewa


The gale struck down the tree(s)

Wichury uszkodziy most/mosty


The gales damaged the bridge(s)

Hungarian, another European language, has a double agreement-marking.2 Unlike


English, the markers in Hungarian fully distinguish all persons and all numbers in all
tenses to mark agreement between the subject and the verb. Moreover, the same set of
verbal suffixes indicate definiteness of the object. In this language, verbal suffixes are
used to mark agreement with two different noun phrases.3
The systemic patterns are often violated as some verbs are not regular in marking
agreement and are not regular in making reference of this agreement. In English, the
verb to be which varies according to person and number of its primary argument
(i.e. subject), in some structures may be ambiguously interpreted in terms of agreement
with the primary or the secondary argument, e.g.: five plus five is/are ten. In French,
the reference to object is determined by the structure of a sentence. When verbs are
conjugated with avoir in the compound tenses, they do not agree with their subjects.
Avoir verbs in these predicates require agreement with their direct objects or direct object
pronouns but only when these precede the verb, therefore Il a vu Marie / Il la vue He
saw Marie / He saw her.

2 I would like to thank Prof. Jerzy Baczerowski for his valuable comments and extension of my previous

knowledge related to Hungarian agreement-marking.


3 MacWhinney, Plh 1997: 68.
160 NINA PAWLAK

The above examples indicate that similarly to the relation with the subject, verbs
mark the agreement with the object and indicate its grammatical features such as person,
gender and number. However, the typological overview of structures with verb-object
agreement reveals the asymmetry in coding the features of subjects and objects. According
to Hopper and Thompson4, languages that mark subjectverb agreement tend to code for
the person and number of the subject, whereas languages that mark object-verb agreement
tend to code for the definiteness and animacy of the object.
The aim of the present paper is to demonstrate the structures of some African languages
that mark agreement between the verb and the object. They will be analyzed in terms
of their coding features.

1. Grammatical correlation between verb and object in African languages

In many African languages, the modifications of verb are related to verb complements.
Among the main devices there are verb suffixes known as verbal extensions, which change
the meaning of the verb, create causatives, benefactives, reciprocals and the like. They
correspond to the object and code the information about their function. In Swahili, the
extension (morpheme) ia of the applicative form indicates that the action is applied on
behalf of, towards or with regard to some objects, e.g.:5

Swahili: a. Mama a-li-pika chakula


/mother 3Sg-PAST-cook food/
Mother cooked food

b. Mama a-li-pik-ia-a watoto chakula


/mother 3Sg-PAST-cook-Ext-Infl children food/
Mother cooked food for children

Similarly in Fulfulde, the information about the grammatical function of the following
object (case) is placed with verb, e.g.:6

Fulfulde: a. o-wolw-ii cf. o-wolw-an-ii moodibbo


/3Sg-speak-PAST/ /3Sg-speak-Ext-PAST teacher/
he spoke he spoke to the teacher

b. o-tay-ii kusel cf. o-tay-ir-ii kusel lai


/3Sg cut-PAST meat/ /3Sg-cut-Ext-PAST meat knife/
he cut (the) meat he cut (the) meat with a knife
4 Hopper, Thompson 1984.
5 More explanations available in Lodhi 2002: 4-26 and in Pawlak 2010.
6 Following Arnott 1970: 338, 353.
MORPHOLOGICAL CODING OF VERB-OBJECT AGREEMENT IN AFRICAN LANGUAGES 161

These morphemes that mark correlation between verb and object are not considered
as being instances of agreement. They are classified among verbal derivation that changes
the valency of verb. But apart from verbal modifications of this kind, there are also some
other morphological changes strictly related to the object and its grammatical categories.
They will be presented here as a manifestation of an agreement between the verb and
its object.
Particular examples illustrating this feature represent languages of three different
language families, namely Chadic (being a part of the Afro-Asiatic phylum), Nilo-Saharan
and Niger-Congo. It will be shown that the agreement manifests itself in a number of
different structures in which syntactic information about the categories of an object is
shifted to the verb. Following the data, verbs agree with their objects in grammatical
features (nominal class, person, number), but also in some discourse features (definiteness,
presence/absence of an object) that also include animacy and some other semantic attributes
of the object.

2. The functions of concordance system in Swahili

The concordance (congruence) in Swahili is based on class system that marks


agreement between the clause constituents. It operates both within nominal and verbal
phrase. As for verbal marking, there are sets of verbal morphemes that are referred to
the subject and the ones referred to the object. In the latter case, transitive verbs have a
pronominal recapitulative marker in a verbal complex, even though its nominal equivalent
is also present in the sentence, e.g.7:

Swahili: a. Wazee walimpenda mtoto (class 1)


/parents 3Pl-PAST-Obj-love child/
The parents loved (the) child

b. Mtoto alikinunua kitabu (class 7)


/child 3Sg-PAST-Obj-buy book/
The child bought the book

The form of the pronominal marker in Swahili has a reference to the class to which
the noun object is attributed8 but its use of this marker within a verbal complex is
determined by many factors, including semantic features of the object. The justification
for the use of an agreement marker -m- in the first example is found in the fact that
the object (mtoto child) is animate9. With this feature, the marker is obligatory. In
7 Ohly, Kraska-Szlenk, Podobiska 1998: 78.
8 The system of classes allows coding for number (plural or singular).
9 The noun mtoto represents the class denoting human beings, therefore the feature +Human is more adequate

in this case.
162 NINA PAWLAK

the second example, -ki- marks the definiteness of the object (kitabu book), therefore
the clause Mtoto alinunua kitabu the child bought a book is also acceptable. Object
marking in Swahili is to some extent defined as optional which also means a big
variety of considerations that determine the choice of either a phrase with a marker
or without it.10 Investigations into object marking have noticed many conditions for
object marking in Swahili, such as stress on the object, its specificity, distinctiveness,
and the discourse-oriented factors that code for the status of information accessible for
hearer.11

3. Other strategies of incorporating object marker into the verb

3.1. Pronominal marker of the object

In some African languages segmental markers are used to indicate that the verbal
complement is an object. In Chadic languages, such a marker is in fact a pronoun that
carries the grammatical categories of object (gender, number).12 For example in Musgu,
object morpheme is indicated as a pronoun that copies the following object. The feature
is interpreted as a structural borrowing from Niger-Congo, e.g.:13

Musgu: a. ym-t dlyo he caught (her) the hyena


/he catch.PAST-her hyena/

b. m tl- Mnjky I saw (them) Musgu people


/I see.PAST-them Musgu/

Some other Chadic languages share the feature of pronoun copying the object. The
pronoun may also indicate the indirect object, like in Daffo14 and Gude15:

Daffo: gofs ref NyroNi he showed (them) the Nyorong masks to the women
/3sg.PAST show them women Nyorong/

Gude: k ka-n-paa Musa buura Musa set down the bag for him
/COMPL set-him-down Musa bag/

10 For example, proper nouns have pronominal marking, even if they are inanimate.
11 Seidl, Dimitriadis 1997: 385.
12 In Chadic (similarly to other Afroasiatic) gender distinction (masculine/feminine) is relevant only for singular.
13 Meyer-Bahlburg 1972: 174.
14 Seibert 1998: 96.
15 Hoskinson 1983: 110.
MORPHOLOGICAL CODING OF VERB-OBJECT AGREEMENT IN AFRICAN LANGUAGES 163

The justification for the use of a marker that corresponds to the object is found in
marking focus. In Kru of the Niger-Congo, here illustrated by the Wob language, there
is a vocalic marker (-E) preceding the object, e.g.:16

Wob: di-E ko s He ate rice today


/he ate-E rice today/

cf. S mE di ko it is TODAY he ate rice


/today FOC he ate rice/

3.2. Objective forms in conjugation patterns

In some other African languages, the verbs are object-sensitive, but the structure
of verb phrase is not so clear in manifesting the sequence of its constituents. However,
the development of present-day forms from the earlier segments is recognizable. In
Kanuri, a Nilo-Saharan language, morphological changes of the verb are related to the
pronominal object. The pattern of verbal morphology is referred to as object conjugation
or objective forms17. The verb root w look at has therefore the following forms of
its verbal complex18:

Kanuri: a. wnmin you are looking at19


wusmin you are looking at me

b. wnwi you (Pl.) are looking at


wswi you (Pl.) are looking at us

Morphological segmentation of the verbal complex allows distinguishing morphemes


representing personal pronouns; it is -s- for 1st person singular, and -sa- for 1st person
plural.20 In diachronic interpretation, the structure is similar to that of many other African
languages in which object pronoun is incorporated into the verbal complex. Synchronically,
they are part of the conjugation system. Similarly in Lamang, which is a Chadic language,
a conjugation paradigm contains a pronominal object (hardly distinguishable in the linear
structure nowadays) that takes the position between the immediate postverbal extensions
and other extension suffixes21, e.g.:

16 Marchese 1983: 119.


17 The label objective conjugation was proposed by J. Lukas in his description of the language (Lukas 1937).
18 Lukas 1937: 36, 112.
19 The constituents of the verb in its finite form follow the order VOS, i.e. verb-object-subject sequence.
20 Cyffer 1978: 306.
21 Wolff 1983: 228.
164 NINA PAWLAK

L -N -n b la -i
/go -CAUS -him -out PERF 1Sg/
[lNnbl] I have brought him out

Conjugation patterns that incorporate pronominal marker manifest a special instance


of verb-object agreement. In these structures, object pronouns are used for coding verbal
argument and they have the status of constitutive morphemes within the verbal complex.
As it had happened with subject pronouns in many worlds languages, they became
integrated into the verbal complex.

4. Agreement for syntactic features of the object

The complex structure of verb phrase in African languages includes the division of
verbs into the classes, according to their morphologic and syntactic properties. In Fyer,
a Chadic language, segmental markers are used to indicate that the verbal complement
is an object. In this way, object morphemes distinguished in the verbal phrase allow
differentiating the nominal and pronominal object. However, the class of verbs determines
the use of a marker. In one class, the nominal object is marked by -a and the pronominal
object is introduced immediately after the verb, e.g.:22

Fyer: Yi mun sh I love you


Yi mun-a ar I love the child

In the other class, the nominal object is unmarked whereas the pronominal object
is marked by -t-, e.g.:

Fyer: M lf-t-is he slaughtered it


M lf rdoN he slaughtered an animal

In Hausa, which is also a Chadic language, morphological class of verbs termed


as grade II verbs23 or changing verbs24 has been distinguished. The morphological
changes of these verbs are determined by marking the distinction between nominal and
pronominal form of the following object. The final ending of the verb is -i when the
object is a noun and -e /-ee/ when the object is used in pronominal form, e.g.:

22 Jungraithmayr 1970: 73.


23 Parsons 1960.
24 Abraham 1962.
MORPHOLOGICAL CODING OF VERB-OBJECT AGREEMENT IN AFRICAN LANGUAGES 165

Hausa: a. Yr y syi littf


/boy 3Sg/PAST buy book/
The boy bought a book

b. yr y sy sh
/boy 3Sg/PAST buy it/
The boy bought it

The morphological changes of verb endings are apparently connected with the relation
to the following object. When the object constituent is transformed into a syntactically
independent item, the verb occurs with an underlying -a /-aa/ ending, e.g.:

Hausa: bn d yr ya sy littf n
/thing which boy 3Sg/PAST buy book COP)
What the boy bought was a book

Also in Yoruba, a Niger-Congo language, the form of some verbs is determined by


the following object. In this language, it is the tonal contrast that carries the information
about either nominal or pronominal form of the object. It is one particular group of
monosyllabic verbs termed as class II verbs25 that receive a low tone before an object
in pronominal form, e.g.:

Yoruba: Lo as2o konn wear the same type of dress


L use it (wear it)

The tonal feature functions as an exponent of grammatical agreement between the


verb and the object. The nominal and pronominal object is differentiated in this way
and the information about it is shifted to the verb. Unlike in a phonologically defined
feature of tonal polarity which makes the tone of the clitic pronoun contrasted with
the tone of the verb, here it is the verb which changes its tone when followed by
a pronoun.
Tonal structure of the verb may be sensitive for marking presence or absence of an
object. In Yoruba, a formal contrast can be observed between a structure in which there
is a following complement and one in which there is none. In monosyllabic verbs the
tone differentiates such two contextual forms:26

25 Bamgboe 1966: 83.


26 Bamgboe 1966: 81.
166 NINA PAWLAK

Verb without following Verb with following


complement complement
t ta sell
m mo know
r ra buy
l lo use
w wo look

Here are examples in which w and wo are differentiated according to their syntactic
context:

Yoruba: a. w8n m8 8n w
/they know it look/
They know how to examine it

b. wo two8n ynb
/look that Europeans/
Look at that of Europeans

Identification of presence or absence of the object in the clausal structure is a systemic


device characteristic of head-marking languages27. The way in which morphology and
syntax interact, however, may be also an indication of agreement. In the languages
presented above, the criterion for differentiating the form of the verb is nominal or
pronominal form of the object.

5. Congruity conditioned by semantics

The verb may take morphological exponents to mark the semantic features of the object.
Three kinds of verbal forms that manifest it are to be mentioned, namely pluractional
verbs, verbs modified by extensions and suppletive verb stems. They can be met in the
languages of all three families.

5.1. Pluractional verbs

Verbal plurality is motivated by either syntactic or semantic agreement with the


arguments. Number agreement with the subject and (or) object is known from Chadic
languages. Pluractional verbs28 reflect a conjugational concord system in few languages,
27 On typology of morphological marking of grammatical relations in African languages see Heine, Nurse
2000: 187.
28 The label itself was coined by Paul Newman (Newman 1990). It replaced the former intensive verbs known

from the descriptions of Chadic languages.


MORPHOLOGICAL CODING OF VERB-OBJECT AGREEMENT IN AFRICAN LANGUAGES 167

and a derivational category in many other languages. Plural verb stems denote semantic
plurality that is a result of an action either performed by many subjects or affecting many
objects, or simply repeated many times. Regarding the nature of verb-object agreement
marked in this way, it is interesting to refer to P. Newmans statement29 that the use
of pluractional stem often depends upon semantic congruity between the verb and the
patient. In the Hausa examples the verbs gin and gggin build are differentiated
following the agreement with the singular form of the noun object (i.e. single object)
and plural form (many objects) respectively:

Hausa: a. An gin makarantaa The school was built


/Non-pers.PAST build school)

b. An gggin mkrntuu The schools were built


/Non-pers.PAST build.PLUR schools/

In Beria, a Nilo-Saharan language, there are pairs of verb stems which remain in
complementary distribution, if the noun object is allocated. The suppletive forms allow
distinguishing singular and plural object, e.g.:30

Beria: /d-/ bring something (one piece)


/tE-/ bring something (many pieces)

In Ibibio, a Niger-Congo language, the difference between inflectional modification and


derivationally motivated concord is not clearly indicated. The alternative verb forms are
used to distinguish between singular and plural noun object31 following the verb, e.g.:32

Ibibio: wk tear one thing wi tear many things


sak mention one thing si mention many things
tak hit one person tai hit many persons
b receive one thing bi receive many things

Syntactic behaviour of the above verbal forms reflects the lexical variety of verbs and
their phrasal compositions. The agreement, however, is recognized in partial modification
of the verbs that are similar to verbal extensions. In this case, the agreement is referred
to the semantic features of the object and allows marking distinction between the notion
single and many, usually with an additional specification related to distribution or
other semantic properties related to the plurality of action that refers to many objects.

29 Newman 1990: 83.


30 Fadoul Khidir 1996: 80.
31 Since the nouns very often have no morphological marking for plural, it is a semantic, not grammatical

category of plurality.
32 Essien 1990: 40.
168 NINA PAWLAK

6. Typology of object-sensitive verbs

Regarding the criteria by which the object-sensitive verbs are used in particular
languages of three genetically unrelated languages, three types of morphological devices
may be distinguished.
Firstly, the verbs with object morphemes usually represent one specific class of verbs.
The morphemes, which mark an agreement, are either segmental or tonal. They may be
regarded as instances of inflectional morphology. Structures that have such a device to
mark verb-object agreement tend to code for nominal or pronominal form of the object.
Differences between verbs followed by the object are mostly recognised for Chadic
languages.
Secondly, there are affixal object pronouns that indicate an object within a verbal
complex. This type of object conjugation with the pronouns incorporated into the verbal
complex makes the verb copying the structure of the whole phrase. The pronoun keeps
the agreement for gender and number, if systemically relevant. The feature is well known
from Niger-Congo languages where the class systems function and both pronominal and
nominal forms of the object are present in one clause. In this case, the pronoun marks
class representation of the object. Very often, the structures with such a recapitulative
pronoun referring to noun object represent the focus marking system.
Thirdly, there are modifications of verb (either morphological or lexical) motivated
by semantic features. With these modifications, the plurality of objects in the functional
relation between verb and noun object is coded on verb. Though basically related to the
category of number, this coding is not regarded as grammatical agreement, as the features
are not marked obligatorily and may be omitted or defined by the context.

Conclusions

Regarding typological criteria, languages which mark agreement between the verb
and the object are rather less common than languages that mark agreement between the
verb and the subject33. African languages share the properties of many worlds languages
in this respect, but at the same time the list of coding means for agreement between the
verb and the object is relatively rich, at least in comparison with European languages.
In terms of grammatical relations, the most distinctive feature of the structures presented
above is that inflectional or derivational morphology of verbs marks the congruity with
the object, not with the subject. If we admit that in semantic relations it is often a Patient
the information of which is placed with a verb, we may state that this kind of agreement
meets the requirements of an ergative concord system. African languages of three different
families seem to manifest the traces of such a system in their syntactic structures.

33 MacWhinney, Plh 1997: 68.


MORPHOLOGICAL CODING OF VERB-OBJECT AGREEMENT IN AFRICAN LANGUAGES 169

Abbreviations

-CAUS causative (extension)


COMPL completive marker
COP copula
Ext (verbal) extension
FOC focus marker
Non-pers non-personal pronoun
Obj object marker
PAST past (tense)
(3) Pl (third person) plural pronoun
PLUR pluractional (stem)
PERF perfective
(3) Sg (third person) singular pronoun

References

Abraham, R.C. 1962. Dictionary of the Hausa Language, London.


Arnott, D.W. 1970. The Nominal and Verbal Systems of Fula, London: Oxford University Press.
Bamgboe, Ay. 1966. A Grammar of Yoruba. Cambridge, Ibadan: University Press Ltd.
Cyffer, Norbert. 1978. Die Verbalstruktur im Kanuri, Afrika und Ubersee 56, 3/4, pp. 294311.
Clements, G.N.; Rialland, Annie. 2008. Africa as a phonological area. In: Heine, B.; Nurse, D.
(eds.), A Linguistic Geography of Africa, Cambridge: Cambridge University Press, pp. 3685.
Essien, Okon E. 1990. A Grammar of the Ibibio Language, Ibadan: University Press Ltd.
Fadoul Khidir, Zakaria. 1996. Quelques caracteristiques des verbes du beria, Afrikanistische
Arbeitspapiere 45, pp. 7781.
Heine, Bernd; Nurse, Derek. 2000. African Languages. An Introduction, Cambridge: University Press.
Hopper, P.J.,; Thompson, S.A. 1984 The discourse basis for lexical categories in universal grammar,
Language 60, pp. 703752.
Hoskinson, James Tylor. 1983. A Grammar and Dictionary of the Gude Language, Ph.D. Dissertation.
Ohio State University.
Jungraithmayr, Herrmann. 1970. Die Ron Sprachen, Tschadohamitische Studien in Nordnigerien.
Glckstadt, I.I. Augustin.
Lodhi, Abdulaziz Y. 2002. Verbal extensions in Bantu (the case of Swahili and Nyamwezi, Africa
& Asia, 2, pp. 426.
Lukas, Johannes. 1937. A Study of the Kanuri Language, London: Oxford University Press.
MacWhinney, Brian; Plh, Csaba. 1997. Double Agreement: Role Identification in Hungarian,
Language and Cognitive Processes 12 (1), pp. 67102.
Marchese, Lynell. 1983. On Assertive Focus and the Inherent Focus Nature of Negatives and
Imperatives: Evidence from Kru, Journal of African Languages and Linguistics 5, pp. 115129.
Meyer-Bahlburg, Hilke. 1972. Studien zur Morphologie und Syntax der Musgu. Hamburg, Buske Verlag.
Newman, Paul. 1990. Nominal and Verbal Plurality in Chadic, Dordrecht, Foris Publications.
Ohly, Rajmund; Kraska-Szlenk, Iwona; Podobiska, Zofia. 1998. Jzyk suahili, Warszawa, Dialog.
Parsons, F.W. 1960. The Verbal System in Hausa, Afrika und bersee, vol. 44, pp. 136.
Pawlak, Nina. 2010. Jzyki afrykaskie, Warszawa: Wydawnictwa Uniwersytetu Warszawskiego.
170 NINA PAWLAK

Seidl, Amanda; Dimitriadis, Alexis. 1997. The Discourse Function of Object Marking in Swahili,
Proceedings of CLS 33: The Main Session, pp. 373389.
Seibert, Uwe. 1998. Das Ron von Daffo (Jos-Plateau, Zentralnigeria), Frankfurt am Main, Peter Lang.
Wolff, Ekkehard. 1983. A Grammar of the Lamang Language, Glckstadt, I.I. Augustin.
R O C Z N I K O R I E N T A L I S T Y C Z N Y, T. LXV, Z. 1, 2012, (s. 171179)

KLAUS SAGASTER

Bible Terminology in Mongolian Translation

Abstract

The paper deals with the difficulties which translators of the Bible had to face when
rendering Christian terms in Mongolian. Which problems did they have to solve in order
to convey ideas to people whose cultural-religious background was very different from
the the Jewish-Christian setting of the Bible? These problems and the decisions on their
solution are shown with the help of six key terms, occuring in three sentences of the
Gospel of St. John: beginning, Word, saviour, world, resurrection and life.
The study is based on six Mongolian, one Buryat and two Kalmuck translations.

Keywords: Bible, Mongolian, translation, Tibetan, Buryat, Kalmuck

How did the translators of the Bible solve the problem of terminology?1 This is,
of course, an age-old and universal question, but I want to illustrate it with the help of
three examples which I took from the Gospel of St. John in six Mongolian, one Buryat
and two Kalmuck translations. I also consulted two Tibetan translations.

My three examples are the following:


1. In the beginning was the Word (Jo 1:1).

1 Other studies of Christian terminology in Mongolian are: Klaus Sagaster, Johann Jhrigs mongolische

bersetzung von Kirchensegen und Vaterunser. In: Zentralasiatische Studien 38 (2009), pp. 283311; idem,
Mongolische und tibetische bersetzungen des Vaterunsers. In: Rocznik Orientalistyczny LXIII/1: Altaica et
Tibetica. Anniversary Volume dedicated to Stanisaw Godziski on His Seventieth Birthday. Edited by Agata Bareja-
-Starzyska, Filip Majkowski, Jan Rogola, Jerzy Tulisow. Warszawa 2010, pp. 198214; idem, Brot und Wein.
Matthus 26, 26-30 in mongolischer bersetzung. In: Denise Aigle et al.: Miscellanea Asiatica. Mlanges en
lhonneur de Franoise Aubin / Festschrift in Honour of Franoise Aubin, Sankt Augustin 2010, pp. 499515.
172 KLAUS SAGASTER

2. This is indeed the Christ, the Saviour of the world (Jo 4:42).
3. I am the resurrection, and the life (Jo 11:25).

Here we find six basic terms: beginning, word, saviour, world, resurrection
and life. How were they rendered in Mongolian and Tibetan? How did the translators
succeed in conveying the ideas represented by these terms to people who had a cultural
background very different from the Jewish-Christian setting of the Bible?
The older Mongolian translations are from 1846 and 1952. The newer translations
are from 1990, 1993, 2003 and 2004. The Buryat translation was published in 2010.
The two Kalmuck translations are from 1887 and 2002. The two Tibetan translations
are from 1933 and 1968.

1. Let us take our first example:

In the beginning was the word,


in Latin In principio erat verbum,
in Greek (John 1:1).

What means in the beginning?


Seven of the eleven translations three Mongol, the Buryat, one Kalmuck and the
two Tibetan use an exact equivalent of in the beginning (Mo ekin-dr / chnd,
Ka erte ekin-d, Bu anchanhaa naaa; Ti thog-mar).2 Four translators three Mongol
and one Kalmuck decided to explain what in the beginning means. They did not
translate literally, but by way of interpretation. The shortest Mongol interpretation is the
one of 1993. It says: All began, and the Word was (Br chld Zarlig bui blg).
We may understand this formulation in the sense that Christ already existed when all
began, and all obviously means the world. This is corroborated by the three other
translations. The Kalmuck translation of 2002 writes: The Word was before the world
originated (or began) orl dchs urd g b). The Mongol translation of 2003 is
quite similar: Before the world became established, the Word already existed (sansar
totanin btk-ee urid yosun nigente orosiu abai). The most detailed explanation is
given by the interpretative translation of 1900: This man who mediated between the Lord
of the World (= God) and mankind existed before the world came into being (Ertncijn
zn ba chn trlchtnijg choorond zuuil gsn tr chn orlon dlchij bij bolochoos
mn orin baj). This man means, of course, Christ. The Buryat version of 2010
translates In the beginning (anchanhaa naaa), adding by way of parenthesis before
the world became established (yurtms dlchjn zochon butgdchh trn).
From the four non-literal translations and from the Buryat translation it becomes
clear that in the beginning means before the world existed. It is surprising that the
translators did not say before the world was created. After all, the first sentence of the

2 Sigla: Bu = Buryat; Ka = Kalmuck; Mo = Mongolian; Ti = Tibetan.


BIBLE TERMINOLOGY IN MONGOLIAN TRANSLATION 173

Bible is In the beginning God created the heaven and the earth (Genesis 1:1). However
this may be, the following verse 3 of the Gospel of St. John clearly states that all things,
that means the world, were created by the Word or that they came into being through
the Word. The Greek ... all ... originated and the Latin omnia ... facta
sunt all ... was made is rendered in the following ways: all things were produced
by him (qamu yaumad anu tegn-iyer bolbasurauladabai) (Mo 1846); all things
were created by him (qamu yaum-a anu tegn-iyer btgegdebei / chamag yum
Tgr btgd) (Mo 1952 / Mo 1993); The Lord of the World created through
him the world and all things which exist in it (Ertncijn zn tgr orlon ertnc ba
tnd orich bch yumyg btlgsn yum) (Mo 1990); God created all through him
(burqan anu tegn-i tsiglen tmen bkn-i egdn btgebei) (Mo 2003); All things
came into being through the Word (Bch yum Tgr bij bolson) (Mo 2004); All
things were created through Him (Buchy yuumn gr l bthdhn bajna) (Bu
2010); All things originated through him (xamuq yuuman inu tgr boluqsan mn,
(Ka 1887); All originated through the Word (Cugn gr dam dgd (Ka 2002).
The Tibetan translations say The Word made all whatever there is (ci-yang yod-pa
thams-cad bka-des mdzad-pa yin) (Ti 1933); The Word made all things (dngos-po
thams-cad bka-des mdzad) (Ti 1968).
It remains the secret of the translators why they chose the expressions to originate,
to become established, to come into being in their interpretation of verse 1 of the
Gospel of St. John instead of using the word to create.

Let us now proceed to the term Word.


The Mongolian translations offer four different renderings of Word. The translation
of 1846 simply uses the Greek word loos (). This did not mean anything for
the Mongolian addressee, but the term Word in the context of the statement In the
beginning was the Word is so complicated that it needs explanation in any case, no
matter in what language.
The Kalmuck translation of 1887 and the Mongolian translations of 1954 and 1993
render Word by arliq / arli / zarlig. This is an ideal choice, since arli means
decree, command, order and therefore also word of a superior, word with reference
to a king or a god. Above all, arli is the word of the Buddha. Since in Mongolian
the words for Buddha and God, the Christian God, are the same burqan , arli is
the proper translation for Word which, as the Bible says in John 1:1, is identical with
God: In the beginning was the Word, and the Word was with God, and the Word was
God. The two Tibetan translations use the exact equivalent of arli, bka. The Ganur,
bKa-gyur, is the translation of the words of the Buddha.
The Mongolian translations of 2002 and 2004 as well as the Buryat translation of
2010 use another expression for Word: Mo ge / g, Bu g3. This is also correct. ge

3 The Buryat translation explains the reason why the Greek term Logos is rendered by g. See n

chlsn, p. 209.
174 KLAUS SAGASTER

is the general name for word, but it can also mean word of a superior. For example,
many letters of the Ilkhans of Persia begin with Our (the Khans) Word (manu ge).4
This, of course, was long ago, and it is doubtful whether for the Mongols of today the
term ge has the same meaning as the term arli. The word ge is also the name for a
traditional genre of Mongolian folk literature. The ge are tales in which the rules of right
behaviour are explained.5 The characters of the tales, however, are animals. Therefore,
this kind of ge hardly evokes the idea of Word of God or of the Word which is Christ.
A surprising rendering of Word is offered by the Mongolian translation of 2003.
It is the word yosun. This term reflects one of the most important, but at the same time
most complex ethical and social concepts of the Mongols. Yosun is the right order of
things. In this sense, it can mean rule, principle, custom, doctrine, system, behaviour,
manner, etiquette. The Two Orders (qoyar yosun) are the right orders of the spiritual and
secular matters. All what is without yosun yosun gei is wrong, bad. The sphere of
the term yosun also comprises the idea of human rights and even human dignity. I think
that yosun is an ideal term to convey to the Mongols the meaning of Johns difficult
concept of Word. The Word is the right order of things, represented and guaranteed
by God and Christ.
Another explanation is given by the Mongolian interpretative translation of 1990:
As we already know, here the Word is Christ who mediates between God and mankind.

2. Now we proceed to our second example:

This is indeed the Christ, the Saviour of the world,


in Latin hic est vere Salvator mundi,
in Greek: (John 4:42).

Let me begin with a philological remark. The English King James translation of the
Bible, which I quote, has the wording Christ, the Saviour of the world. Christ is also
added in two oldest translations, the Mongolian of 1846 and the Kalmuck of 1887. This
addition is due to a text different from that used by the other translations, which do not
have the word Christ.
The concepts saviour and saviour of the world are not very difficult to convey
to people who belong to a Buddhist culture. Saviour (Mo aburai, Ti skyob-pa)
is an epithet of the Buddha.6 The Tibetans and the Mongols also call the Dalai Lama
and the Panchen Lama saviour (helper; protector; deliverer [skyabs-mgon]).7 For the

4 See D. Tumurtogoo With the Collaboration of G. Cecegdari, Mongolian Monuments in Uighur-Mongolian


Script (XIIIXVI Centuries), Taipei, Taiwan 2006, pp. 150156.
5 Walther Heissig, Zur berlieferung der ge-Dichtung. 1. Vier unbekannte ge. In: Zentralasiatische Studien

1 (1967), pp. 163235.


6 Yumiko Ishihama and Yoichi Fukuda, A New Critical Edition of the Mahvyutpatti. Sanskrit-Tibetan-Mongolian

Dictionary of Buddhist Terminology. (Tokyo) 1989, p. -1-, no. 15.


7 Sarat Chandra Das, A Tibetan-English Dictionary with Sanskrit Synonyms. Revised and Edited by Graham

Sandberg and A. William Heyde. Calcutta 1902, p. 99.


BIBLE TERMINOLOGY IN MONGOLIAN TRANSLATION 175

Mongols even Chinggis Khan is a saviour (aburai). He is, for example, addressed
by this term in the ritual texts, which are recited in the Chinggis sanctuary in Ordos in
Southwest Inner Mongolia.8
How is the term saviour rendered in Mongolian? The translations use two words:
aburai (avrag / avra) protector, provider of refuge and tonilai deliverer.
Both terms are basically identical, but there is a slight semantic differentiation: aburai
is one who gives protection, refuge (abural), and tonilai is one who delivers, for
example from suffering. The Buryat translation has avraga, the Kalmuck translations
have tonilaqi. The Tibetan translations render saviour by the binomial skyabs-mgon
protector. This term is composed of skyabs protection, refuge and mgon protector,
master, lord. It is the exact equivalent of Mongolian aburai protector.9 In Buddhism
the term skyabs-mgon has a very important meaning, particularly the first element, skyabs
protection, refuge. In this paper, of course, I cannot go into details.
We may state that both protector and deliverer are adequate equivalents of
saviour. Both protection and deliverance are means of saving the world, mankind, from
suffering, from suffering which is the existential condition of both Christians and Buddhists.
We still have to ask how the translators render the word world. They use the
following five terms, all of which have different semantics.
1. Mongolian and Kalmuck yirtin: This word corresponds to jig-rten in the Tibetan
translations. Yirtin.means the world in Buddhist cosmography. Since there exist
innumerable world systems, yirtin can also be rendered by cosmos. In the
Mongolian translation of 1990, the name of God is rendered as Lord of yirtin /
(ertnc) (ertncijn zn), that means the Lord of the World or the Cosmos.
2. Mongolian orilang, Kalmuck orlo: orilang/orlo is the Mongolian equivalent
of Sanskrit sasra (Mo sansar in John 1:1), the Buddhist cycle of existences.
3. Mongolian delekei, Buryat dlchj is theearth, in particular surface of the earth.
4. Mongolian orilang delekei. Here the word orilang is combined with delekei.
5. The Buryat translation clearly states what world means the people of the world
(dlchjn zon), i.e. mankind. (In John 1,1 the Buryat version uses the binomial jurtms
dlchj.)
3. Our last example is

I am the resurrection, and the life,


in Latin Ego sum resurrectio, et vita,
in Greek (John 11:25).

For both the Mongols and the Tibetans the Christian term resurrection is not easy
to understand. For them, a new life comes about through reincarnation, which is a typical
8
Klaus Sagaster, Ein Ritual aus dem heutigen inggis-Heiligtum in Ordos. In: Zentralasiatische Studien 23
(1992/1993), pp. 145151 (p. 146).
9 Bod-sog-rgya gsum-gyi shan-sbyar tshig-mdzod chen-mo / Tbed mongol kitad qariaulusan yeke toli /

Zang-meng-han duizhao da cidian, Liaoning 2001, p. 73.


176 KLAUS SAGASTER

Buddhist concept: It means continuous change of existences. Resurrection, however, is


something else. Resurrection is a unique event: Christ resurrected only once, and human
beings will also resurrect only once. In order to make this clear, the Christian translators
deliberately did not choose a Buddhist term, like to change, to take another form, to be
reincarnated. They solved the problem in different ways:
The Tibetan translations render the term resurrection almost literally by rising (Tib.
lang-ba). The Tibetan translation of 1968 writes I am the rising and the life (lang-ba
dang gson-pa-nyid nga-yin). The translation of 1933 additionally explains that rising
means rising from the dead: It writes I am the rising from the dead, and the life
(nga ni shi-nas lang-ba-nyid dang gson-pa-nyid yin-te). In the non-Tibetan translations,
however, we do not find, with one exception, the words rising or to rise, but another
term, which obviously is more comprehensible for the Mongols, the Buryats and the
Kalmucks. It is the term revival (amilal, Mo 1993; amilalt, Mo 2004; amidyralga, Bu
2010; mdrllhn, Ka 2002) or, as an interpretation of what revival means, the term the
one who revives (amiduraului, Mo 1846, Mo 1952; dakin amiduraului, Mo2003).
Therefore, Christ says, according to these translations, either I am the revival, and the
life (bi bol amilal ba am mn [Mo 1993]; bi bol amilalt ba am mn [Mo 2004]; Bi
chadaa Amidyralga gb, Ami nahan gb [Bu 2010]; Bi mdrllhn boln irhl
[Ka 2002]) or I am the one who revives, and (I am) the life (bi ber amiduraului
kiged amin mn [Mo 1846]; .bi ber amiduraulu i kiged amin mn bui [Mo 1952]; bi
bolbasu darui dakin amiduraului mn. bi bolbasu darui ami nasun mn [Mo 2003]).
Only the Kalmuck translation of 1887 preserves the idea of rising and says: I am
the one who, by reviving, causes to rise, and (I am) the life (bi amiduruulun bosxoqi
bold amin igi mn). A quite particular rendering is given by the Mongol translation
of 1990. It reads: I am able to revive all the dead men and to give (them) life (Bi nas
barsan chnijg chrtl dachin amiluul, amdral g adna).
The Mongolian translation of resurrection as revival points to the close relation
between resurrection and life. The Mongol terms for revival (Mo amilal, amilalt; Bu
amidyralga; Ka mdrllhn), and for to revive (Mo amiduraulqu, dakin amiduraulqu;
Ka amiduruulxu) are derived from the word amin life. The reasoning is that reviving
leads to resurrection, and resurrection leads to life, to the real, the eternal life.
The word life (Mo. amin, am; Bu ami nahan; Ka amin, irhl; Tib. gson-pa) did
not create any problem for the translators. As we could see, this was not the case with
many other terms. To translate the Bible is not an easy task.

Christian Terms

1. John 1:1:
In the beginning was the word
Latin: In principio erat verbum,
Greek. .
BIBLE TERMINOLOGY IN MONGOLIAN TRANSLATION 177

In the beginning:
in the beginning: Mo ekin-dr / chnd; Ti thog-mar (Mo 1846, Mo 1952, Mo2004;
Ti 1933, Ti 1968)
in the beginning: Bu anchanhaa naaa (Bu 2010)
in the beginning: Ka erte ekin-d (Ka 1887)
all began, and: Mo br chld (Mo 1993)
before the world originated (or began): Ka orl dchs urd (Ka 2002)
before the world became established: Mo sansar totanin btk-ee urid (Mo2003)
before the world came into being: Mo orlon dlchij bij bolochoos mn (Mo1990)
Word:
lgos: Mo los (Mo 1846)
word (of a superior), decree, command, order: Ka arliq (Ka 1887); Mo arli/
zarlig (Mo 1952, Mo 1993); Ti bka (Ti 1933, Ti 1968)
word: Ka g (Ka 2002); Mo g (Mo 2004); g (Bu 2010)
order, principle, rule, custom: Mo yosun (Mo 2003)

2. John 4:42:
This is indeed the Christ, the Savour of the world,
Latin: Hic est vere Salvator mundi,
Greek: .

saviour of the world


protector of the world: yirtin-yin aburai (Mo 1952)
protector of the world: ertncijn avrag (Mo 2004)
protector of the world: jig-rten-gyi skyabs-mgon (Ti 1933, Ti 1968)
protector of mankind: chn trlchtnij avrag (Mo 1990)
protector of the people of the world: dlchjn zoniie Abaraga (Bu 2010)
protector of the world: orlogin avra (Ka 2002)
protector of the world: dlchijn avrag (Mo 1993)
deliverer of the world: orilang delekei-yin aburai (Mo 2003)
Christ and deliverer of the world: kristos bolun yirtin-yin tonilai (Mo 1846)
deliverer of the world, Christ: yirtinyigi tonilaqi kiristos (Ka 1889)
saviour:
protector; provider of refuge: Mo aburai, avrag (Mo 1952, Mo 2003, Mo1993,
Mo 2004); Bu abaraga (Bu 2010); Ka avra (Ka 2002); Ti skyabs-mgon
(Ti 1933, Ti1968)
deliverer: Mo tonilai (Mo 1846); Ka tonilaqi (Ka 1887)
world:
world, cosmos: Mo yirtin, ertnc (Mo 1846; Mo 1952, Mo 2004); Ka yirtin
(Ka 1887); Ti jig-rten (Ti 1933, Ti 1968). God: Lord of the World / Cosmos
(ertncijn zn) (Mo 1990)
178 KLAUS SAGASTER

world, cycle of existences, Samsara: Ka orlo (Ka 2002) (= Mo. orilang; Mo


sansar in John 1:1)
world, surface of the earth: Mo delekei, dlchij (Mo 1993)
world, cycle of existences surface of the earth: Mo orilang delekei (Mo 2003)
mankind: Mo chn trlchtn (Mo 1990)
people of the world, mankind: dlchjn zon (Bu 2010)

3. John 11:25:
I am the resurrection, and the life,
Latin: Ego sum resurrectio, et vita,
Greek: .

resurrection:
rising: Ti lang-ba (Ti 1968)
rising from the dead: Ti shi-nas lang-ba-nyid (Ti 1933)
revival: Mo amilal (Mo 1993), Mo amilalt (2004); Bu amidyralga (Bu 2010); Ka
mdrllhn (Ka 2002)
the one who revives: Mo amiduraului (Mo 1846, Mo 1952); dakin amiduraului
(Mo 2003)
the one who, by reviving, causes to rise: Ka amiduruulun bosxoqi (Ka 1887)
to be able to revive all dead men: Mo nas barsan chnijg chrtel dachin amiluul
.. ad(ach) (Mo 1990)
life:
Life: Mo amin / am (Mo 1846, Mo 1952, Mo 1993, Mo 2004); Ka amin (Ka1887);
Mo ami nasun; (Mo 2003); Bu ami nahan (Bu 2010); Mo amdral (Mo 1990);
Ka irhl (Ka 2002); Ti gson-pa (Ti 1933, Ti 1968)

Bible Translations

a) Mongolian
1. Mo 1846
Bidan-u een ba tonilai iisus keristos-un sine tistamint kemegi nom anu orosibai:
Wiliyam Swan [William Swan]. Idward Stalibras [Edward Stallybrass] qoyar anu
egn-i griyeg eke-ee mongol kelen-dr oriulbai:: Anggliy-a oron-u London
qotan-a. 1846 on-a darudabai
New edition:
The New Testament of our Lord and Saviour Jesus Christ. Translated out
of the original Greek into the Mongolian Language, by Edward Stallybrass
und William Swan. St. Petersburg: British and Foreign Bible Society 1880
2. Mo 1952
The New Covenant commonly called The New Testament of Our Lord and Saviour
Jesus Christ in Mongolian. Revised Version. Hongkong: Hongkong Bible House 1952
BIBLE TERMINOLOGY IN MONGOLIAN TRANSLATION 179

3. Mo 1990
in gr. Gongkong: Olon ulsyn Biblijn ngsn nijgmlg 1990
4. Mo 1993
in grs. The New Testament. Recovery Version. Mongol-angli bibli. Ulaanbaatar:
Mongol Bayaryn Md 1993
5. Mo 2003
Ibegelt nom (Kkeqota 2003)
6. Mo 2004
Ariun Bibli. Chuuin Gr. in Gr. (Ulaanbaatar:) Ariun Bis Biblijn Nijgmlg
2004. (Previous editions: 1995, 1996, 1997, 2000.)

b) Buryat
7. Bu 2010
n Chlsn. Moskva: Nangin Bg oruulgyn institut 2010

c) Kalmuck
8. Ka 1887
Novyj zavet gospoda i spasa neego Iisusa Christa. Tom pervyj. S greeskago
podlinnika na kalmyckij jazyk perevel Aleksej Pozdneev. Izdano idiveniem
Velikobritanskago i inostrannago biblejskago obestva. Sanktpeterburg: Depo
Velikobritanskago i inostrannago biblejskago obestva. 1887
9. Ka 2002
in boocan. Moskva 2002

d) Tibetan
10. Ti 1933
Dam-pai gsung-rab es-bya-ba bzhugs-so. Zhal-chad gsar-bai mdo-rnams ni / Tibetan
New Testament, R.V., Ed. 2933. British & Foreign Bible Society, Shanghai 1933
11. Ti 1968
Zhal-chad snga-phyi gnyis-kyi mdo bzhugs-so / The Holy Bible In Tibetan. (Noplace
indicated): United Bible Societies 1983. (This Edition comprises a reprint of the
1948 Old Testament and the 1968 New Testament.)

e) Greek/Latin
Nestle-Aland, Novum Testamentum Graece et Latine. 26th edition, 7th impression.
Stuttgart 1984

f) English
The Holy Bible. Authorised King James Version. Oxford / London / New York /
Toronto. Undated
R O C Z N I K O R I E N T A L I S T Y C Z N Y, T. LXV, Z. 1, 2012, (s. 180194)

TADEUSZ SKORUPSKI

The Buddhas Stpa and Image.


The Icons of his Immanence and Transcendence

Abstract

This paper broadly explores the innate or doctrinal permutations of the Buddhas
stpa and image, and attempts to establish their mutual relationship. In the first instance
the Buddhas stpa was considered as the symbol or epitome of his enlightenment and
parinirva, although as such the stpa was considered to be empty, in the sense that
the Buddha was utterly absent. However, eventually the stpa received a whole gamut
of doctrinal interpretations, which viewed it as an epitome or repository of the absolute
Buddha-body (dharmakya). In the case of the Buddhas image, the pattern of doctrinal
developments is similar to that of the stpa. Since the Buddha was absent from this
world, initially he was not represented in iconic forms but only through symbols. Then
again, eventually the Buddha was depicted in anthropomorphic forms, which did not
reflect his mundane appearance. Instead his images were cast in conformity with doctrinal
permutations, which propounded an idealised person or state of Buddhahood completely
divested of the Buddhas historicity and mundane appearance. The Buddhas stpa and
image do have their respective interpretations and functions. But when they are combined
and considered together, they may be viewed as the Buddhas transcendent and immanent
presence in this world.

Keywords: Buddhism, stupa, Buddha image, cult, relics

The overall focus of this paper is not specifically or exclusively on the stpa
architecture and on the iconography of the Buddhas image, but predominantly on the
historical and doctrinal considerations, which disclose their contextual and ideological
background, and underpin their innate and mystic permutations. As monuments and icons,
THE BUDDHAS STPA AND IMAGE. THE ICONS OF HIS IMMANENCE AND TRANSCENDENCE 181

stpas and images do have their respective and distinct function and significance, but
they also have certain ideological interconnections. Casting a broad vision, allegorically
and philosophically, we begin our intellectual journey with emptiness and with empty
cosmic space. We then advance to fill that empty space with the Buddhas stpas and
images. Finally we conclude our journey by dissolving all those stpas and images in
cosmic space and emptiness.

Historical and contextual scenario

In one of his instructive discourses the Buddha makes the following statement, which
he addresses to his disciples:

The body of the Tathgata continues to exist even though the roots
productive of a new existence have been eradicated. As long as his body
survives, so long the gods and men can see him. But, once his body is
dissipated and his life is terminated, the gods and men will see him no
more.1

In another discourse Upasva, one of the notable disciples, asks the Buddha whether
there would be any consciousness in the person endowed with all-encompassing vision
and liberated from sasra, after being emancipated and becoming cold.
In reply the Buddha states that just as a flame agitated by the force of the wind
dwindles and vanishes, so the Sage, once released from his psychophysical individuality
(nma-rpa), becomes appeased and vanishes. No person can measure him or speak of
him. When all phenomena are dissolved, all ways of speaking are also removed.2
Then again upon being asked by the mendicant Vacchagotta whether the Tathgata
exists after death or does not exist, the Buddha declines to give an affirmative or negative
answer. Instead he explains to Vacchagotta that it is a speculative question, and that it
is not conducive to spiritual progress and the attainment of liberation.3 In other words,
once the flame of a lamp is extinguished, it is futile and inappropriate to ask or speculate
about the vanished flame.
Taking into consideration the above three discourses, it is poignantly apparent that once
the Buddha passes away, his Sagha and lay followers would be doomed to experience
and endure in their hearts his utter absence, and to gaze in their minds into the vacuum
of cosmic space.
Essentially, prior to his passing into parinirva, the Buddha firmly signals the
traceless dissolution of his human body and consciousness in conformity with his

1 Dghanikya, I.46.
2 Suttanipta, verses 10741076.
3 Majjhimanikya, I.484-86.
182 TADEUSZ SKORUPSKI

fundamental teaching: all conditioned phenomena have the common characteristics of


impermanence, suffering, and impersonality. Ultimately there is no such thing as a being
(sattva), person (purua) or a self (tman), but merely a transitory stream of mental and
physical phenomena or elements (dharmas). The world of living beings is no more than
a constantly changing succession of illusory appearances. In this context the Buddhas
human form had no greater reality or durability than the bodily forms of all other living
beings. However, there is one tremendous difference, namely the Buddha has attained the
supreme and perfect enlightenment (bodhi), and the final and utter emancipation from the
bonds of sasra. As an enlightened being, the Buddha is beyond all forms of existence,
and transcends this phenomenal world of illusory experiences and appearances. In his
absolute and perfected state, he also transcends and rejects the philosophical concepts of
being and non-being. Thus bearing in mind the above scenarios, the Buddha as a human
being, and as an enlightened being, is about to pass into parinirva, understood here
as a complete dissolution of the Buddhas existence in sasra.

The Buddhas legacy

The Buddha clearly signals his demise and renounces the remainder of his life, but
he does not utterly abandon his followers. Having reached Kuinagara, the place of his
parinirva, the Buddha instructs nanda that his bodily remains should be disposed in
the same way as the remains of a universal monarch (cakravartin). He explains to nanda
that his bodily remains should be wrapped in five hundred pieces of linen-cloth, placed
on a funeral pyre and cremated. After the cremation a stpa should be erected at the
crossroads. Whoever places wreaths or perfumes before the stpa with devout thoughts,
will gain much benefit and happiness.4 In another conversation with nanda, the Buddha
identifies four places the sight of which should evoke noble emotions in his followers,
namely the place of his birth (Lumbin), the place of his enlightenment (Bodhgay), the
place where he set in motion the Wheel of Dharma (Srnth, Vras), and the place
of his parinirva (Kuinagara).5 Finally, after the Buddhas death and cremation, his
flesh completely vanishes without leaving any ashes, and lo and behold, the only thing
that remains are the Buddhas stainless relics (arra).6
It is apparent from the above scenarios that although he passed away, the Buddha left
behind him a visionary legacy and his relics. He instructed his disciples to build a stpa
and to venerate it with acts of devotion and offerings. He also encouraged his devotees
to go on pilgrimages to the four places of his principal life events. Thus the Buddhas
instructions inspired a very unique and dynamic pattern of Buddhist life and practice
with the primary focus on the worship of the stpa containing his relics.

4 Dghanikya, II.141-42.
5 Dghanikya, II.140.
6 Dghanikya, II.164.
THE BUDDHAS STPA AND IMAGE. THE ICONS OF HIS IMMANENCE AND TRANSCENDENCE 183

The stpa story and architecture

After the Buddhas cremation, his relics were divided into eight shares and distributed
among eight rulers, who built stpas over the received shares of the relics. In addition
two more stpas were erected, one over the vessel or vase (droa) used for dividing the
relics, and one over the embers.7 Thus, shortly after the Buddhas cremation, altogether
ten stpas were built, eight containing proper relics, one containing the droa vessel,
and one containing the embers. The subsequent history and conditions of these stpas
are not well known or documented. According to the available records, during his rule
the emperor Aoka (268232 BC)8 significantly enhanced and spread the cult of the
stpa. He is said to have extracted the Buddhas relics from the above eight stpas, and
enshrined them in a symbolic number of eighty-four thousand stpas, which he erected
in various parts of his empire.9 This was the second and perhaps the largest redistribution
of the Buddhas relics. During the subsequent centuries, and up to the present time, the
stpas continued to be constructed in India and eventually in other Buddhist lands of
Asia. As the Buddhist history progressed and new doctrines were formulated, the stpa
architecture was modified and enriched with various elements, which reflected the doctrinal
and ethnic innovations.
The original shape or architecture of the stpa is largely a matter of academic
speculations, although it is generally believed that the ancient stpas were simple mounds
of piled up earth. In one late Buddhist narrative, the visionary invention of the stpa
prototype, or perhaps of the stpas physical appearance, is attributed to the Buddha
himself. In this particular narrative the Buddha concretely demonstrates the basic stpa
design to Trapua and Bhallika, the two merchants who visited him at Bodhgay soon
after his enlightenment. He folds his three monastic robes into squares and piles them
up on the ground. Next he takes his alms bowl and places it upside down on the top
of the robes. Finally he takes his mendicant staff, and positions it vertically over the
alms bowl. It is in conformity with this allegorical model that Trapua and Bhallika
constructed a stpa after returning to their native Bactria.10 In fact the above stpa model
does not represent the first stpa prototype, but it is a retrospective attribution, which
reflects the already well established tripartite structure of the stpa: foundation, dome,
and superstructure.
The foundations (ved, nem, medh) of the surviving stpas are circular or square
in shape, and some texts also speak of octagonal and other shapes.11 Eventually these
foundations, in particular square foundations, were increased to three, four or more
concentric or pyramidal terraces or platforms. The dome (kumbha or aa) arises above
the foundation, and it can have several shapes. Some texts specify four primary shapes:
7 Dghanikya, II.166-67.
8 The dates as given in E. Lamotte, 1988, 88.
9 E. Lamotte, 1988, 240241; J.S. Strong, 1983, 109119.
10 Siyuki, I, 4748; M. Benisti, 1960, 4243. See also E. Lamotte, 1988, page 66 and note 155.
11 A. Bareau, 1962, 235; T. Skorupski, 2002, 165.
184 TADEUSZ SKORUPSKI

heap of grain (dhnya-kra), inverted alms bowl (ptra), oval or birds egg (khaga),
and vase (kalaa).12 The Buddhas relics, inserted inside an urn or a reliquary, are deposited
in the bosom of the dome. Thus the dome constitutes the essential component of the
stpa. The third stpa element, the superstructure, includes a square pavilion (harmik)
surmounted by a tapered pole or spire (yai). The significance of the pavilion is not
conclusively determined. In some sources, the pavilion serves as a dwelling of the four
kings and guardians of the cardinal directions (lokapla). The eyes painted on some
Nepalese stpas are suggestively interpreted as the Buddhas all-seeing eyes, but more
probably they betoken the four guardian deities. The tapered spire can be simple or
elaborated. Depending on the stpa, the spire supports one or a series of disc-like umbrellas
(chatrval).13 On some Tibetan and Nepalese stpas, the spire is sheltered by a large
umbrella-like rain canopy, and surmounted by the sun and moon emblems.
In the course of time, the three principal elements of the stpa were ingeniously and
imaginatively enriched with new decorative and architectural additions, and their shapes
and proportions were manipulated in a variety of ways, something that is amply reflected
in the architectural variations of the existing stpas in the Buddhist lands of Asia.
Some ancient stpas, such as the one at Sc, were also furnished with balustrades
and gates positioned towards the cardinal directions. Some of such balustrades and gates
were decorated with carved jtaka stories, and the Buddhas symbolic representations. At
some stage after the institution of the Buddhas anthropomorphic representations, niches
or chapels were excavated or constructed on the sides of stpas to shelter images. Thus,
as discussed later on, the Buddhas stpa and image came to be merged or moulded
together into a combined icon expressive and symbolic of the Buddhas imperceptible
and perceptible configurations.

The categories of the Buddhas relics

As we have seen above the Buddhas body was cremated, and his relics were divided
and deposited in eight stpas. Thus it would seem that initially there were no other
relics. However, as the worship of the stpa spread and became popular, a variety of
narratives and legends were composed about a number of the Buddhas relics other than
those deposited in the eight stpas. When these narratives and legends are collated, they
speak of seven major post-mortem relics: four canine teeth, two collar-bones (clavicles),
and the cranial bulge (ua).14

12 T. Skorupski, 2002, 165. Some texts also mention the domes shaped like a lotus bud (padma), water bubble

(budbuda), bell (gha), and gooseberry (malaka). H. Ruelius, 1980, 269.


13 The ancient stpas, for instance at Sc, had one or more umbrellas fixed above the harmik. The umbrella

is the emblem of royalty. The Buddha said that stpas of the Tathgata should have thirteen umbrellas, stpas of
arhats four umbrellas, of non-returners three, of once-returners two, and of stream-winners one umbrella. G. Roth,
1980, 184; A. Bareau, 1962, 236.
14 Dghanikya, II.167-68; K. Trainor, 1997, 121; J. Strong, 2004, 185.
THE BUDDHAS STPA AND IMAGE. THE ICONS OF HIS IMMANENCE AND TRANSCENDENCE 185

In addition to the above seven post-mortem relics, there also exist records about several
other relics of the Buddha, such as his alms-bowl, robe, shadow, hair, and footprints.15
During the Tang period in China, elaborate festivals were conducted in honour of the
Buddhas finger-bone relic.16 Chapter twenty-eight of the Mahvasa, considered to
be a rather late composition, contains one of the most comprehensive accounts of the
distribution of the Buddhas relics. According to this source, in addition to the relics
distributed after the Buddhas cremation at Kuinagara, the Buddhas forty teeth, the
hair of his head and body, and all his personal articles, including his alms bowl, robes,
mendicant staff, and razor, were redistributed among the gods and men.
The different relics allegedly left by the Buddha were eventually classified into three
or more categories. Certain sources appertaining to the Pli tradition divide the Buddhas
relics into three categories: 1. bodily relics (arra-dhtu) left immediately after the
cremation, 2. contact or personally used articles (pribhogika-dhtu), such as his alms
bowl and mendicant staff, and 3. indicative or commemorative relics (uddesika-dhtu)
such as images.17
The Thai tradition distinguishes four types of cetiyas on the basis of the articles that
are deposited inside them: 1. cetiyas containing bodily relics (dhtu-cetiya); 2. cetiyas
containing contact or personally used articles (pribhogika-), including the bodhi tree;
3. cetiyas containing the teachings (dhamma-) in the form of the formula of dependent
origination, or canonical texts inscribed on palm-leaves; and 4. cetiyas containing indicative
objects (uddesika-), such as images, paintings and footprints.18
The Tibetan classifications of relics include the relics of the Buddha and of holy
or enlightened lamas and teachers. Out of several classifications we present here only
one classification, which distinguishes four categories of relics: 1. Bodily remains (sku
gdung gyi ring bsrel), namely the bones of Buddhas, Bodhisattvas, rvakas, and holy
lamas. 2. Mustard-seed-like relics (yungs bru lta bui ring bsrel), which are the Buddhas
multiplying relics (phel gdung), and the relics (ring bsrel) of accomplished luminaries
(mkhas grub). 3. Indicative relics (sku bal gyi ring bsrel) which include the hair, nails
and other similar objects. 4. Relics of the absolute body (chos skui ring bsrel) which
have the form of dhras (gzung).
The Sanskrit terms dhtu and arra19 primarily denote the mustard-seed-like relics,
which emerged after the Buddhas cremation, but in a broader sense they also denote all
other relics as discussed above. These relics can also emerge during the lifetime of holy
persons, from their heads or clothes, or again from other relics, images or stpas. They
are described as being brilliant white in colour and approximately the size of mustard
seeds or rice grains. Some sources say that they can also appear in different colours

15 Fa-hsien, 3437, 39, 88, 109. See also Siyuki, I, 45, 67, 9899.
16 K. Chen, 1964, 27880; B. Faure, 2003, 1141.
17 Jtaka, IV.228-29; K. Trainor, 1997, 89; Y. Bentor, 1994, 23.
18 Y. Bentor, 1994, 23.
19 Transcribed in Chinese as shililuo or sheli, Japanese shari. Translated in Sino-Japanese as reikotsu (sacred

bone), and in Tibetan as ring bsrel (lasting object). B. Faure, 2003, 1128.
186 TADEUSZ SKORUPSKI

and dimensions.20 In his commentary on the Parinibbna-sutta, Buddhaghosa says that


the Buddhas remains resemble jasmine buds (sumana-makula), pearls (mutt), and gold
(suvaa).21 In one account the Buddha tells nanda that he decided to fracture his
diamond body (vajrakya) into granules resembling mustard seeds, and to disperse them
in the world so that his followers, without seeing him, may worship his relics and gain
merit, due to which they may become reborn in happy places, or again so that they may
gain the path of deliverance.22 The relics are also believed to have the power and capacity
to move between places, or to abandon the places where they receive no veneration, and
to travel to the places where they would be worshipped.23
The identity of the Buddhas relics is rather controversial and doctrinally problematic.
As already discussed above, in conformity with the early Buddhist teachings, having
passed into parinirva, the Buddha had completely disappeared from sasra. From the
perspective of this doctrinal position, the Buddha is absent and consequently his relics
and stpas are empty and powerless. On the other hand there evolved convictions and
doctrines recorded in texts and inscriptions, which firmly assert that the Buddha is present
in his relics, and that they are endowed with his attributes and powers. We indicate below
a few sources, which affirm the Buddhas infinite lifespan and presence in his relics.
In the Mahparinirva-stra, the Buddha states that the Tathgatas body is a diamond
body (vajrakya), which is indestructible and immutable for countless aeons. It is not the
body of a man or god; it is not subject to fear; it is not mixed with food; it is not a body;
it does not involve origination (utpda) or cessation (nirodha); it is unlimited, infinite,
traceless, formless, pure, immovable, without sensation, and unconditioned (asaskta).24
In one of his commentaries Buddhaghosa affirms the identity of the Buddha while he
was alive with his bodily relics after death. In another place he also states that a monks
failure to worship at a cetiya is equivalent to negligence to attend upon the Buddha.25
The inscription on the Shinkot reliquary states that the Buddhas relics are a living
being endowed with breath (prasameta).26 Similarly the inscriptions at Ngrjunikoa
indicate, in Schopens words, that their redactor did not think of the dhtu or relic as
a piece or a part of the Buddha. He seems, in fact, to have thought of it as something
that contained or enclosed the Buddha himself, something in which the Buddha was
wholly present.27
The available accounts of the Buddhas three parinirvas also reinforce the belief in
the Buddhas tangible presence in his relics and stpas. According to the Theravda sources,
the Buddhas first parinirva occurred at Bodhgay, where he became enlightened and

20 J.S. Strong, 2004, 10, 46.


21 K. Trainor, 1997, 120.
22 P. Skilling, 2005, 289.
23 K. Trainor, 1997, 130; J.S. Strong, 2004, 224.
24 Hbgirin, 178.
25 K. Trainor, 1997, 9293.
26 E. Lamotte, 1988, 431; G. Schopen, 1997, 126.
27 G. Schopen, 1997, 158.
THE BUDDHAS STPA AND IMAGE. THE ICONS OF HIS IMMANENCE AND TRANSCENDENCE 187

liberated from the bonds of defilements and ignorance. The Buddhas second parinirva
took place at Kuinagara, and consisted in the dissolution of his psychophysical aggregates.
The third and final parinirva is due to take place at the end of the cosmic dissolution
of the world, when the Buddhas relics will completely disappear. At that time all his
relics will reassemble at Bodhgay, assume the form of his golden body, and then burst
into flames and utterly disappear. The overall idea is that so long as the Buddhas relics
remain in the world, his parinirva is incomplete, and thus as much merit can be gained
from worshipping the Buddhas relics as from worshipping the Buddha himself.28
We have discussed above the categories and nature of the Buddhas relics, because
they are the kernel or heart of stpas and the rationale for their construction. As structures
the stpas are symbolic of the Buddhas parinirva, but it is their content, namely the
relics, that endows them with spiritual symbolism and mystic potency.

The stpa symbolism

As such the stpa is not considered to be a burial site, but it is seen as the epitome of
the Buddhas liberated mind and parinirva. The Buddhas presence in the stpa is also
controversial and problematic, like his presence in the relics. The traditional Theravda
maintains that the Buddha is not present in any physical form,29 and the Abhidharmakoa
postulates that the stpa is empty in the sense that the Buddha is absent.30 By contrast
some Mahyna sources assert that the Buddhas life is timeless and that his dharmakya
is timelessly present. The Buddhas presence is contained in the stpa, and although the
stpa enshrines relics, the devotees perceive it as the Buddha.31
A number of the relevant texts superimpose doctrinal and symbolic interpretations
on the architectural and other elements of the stpa. We give here only one indicative
interpretation of the stpas primary elements. The width of the base-foundation epitomises
the purity of the thirty-two major marks of a great being (mahpurua), and its height
denotes the twelve links of dependent origination (prattyasamutpda). The dome
epitomises the Buddhas absolute body (dharmakya), and its encompassing garland
symbolises the six kinds of intuitive knowledge (abhij). The pavilion represents the
purity of the three vehicles leading to enlightenment (triyna), and its girth is expressive
of the four noble truths. The spire of thirteen umbrella-discs epitomises the thirteen
stages leading to the attainment of buddhahood.32 In some texts the stpa is invested
with practically the entire array of Buddhist teachings and the Buddhas attributes: the

28 D.W. Chapell, 1980, 130; G.P. Malalasekera, 1960, I, 804; II, 303; K. Trainor, 1997, 13.
29 K. Trainor, 1997, 166.
30 Koa, IV, 747-48.
31 Saddharmapuarka, H. Kerns translation, 298310, particularly pages 299303, 307. See also Hirakawa,

19631964, 88.
32 For further details see G. Roth, 1980, 187193; T. Skorupski, 2002, 166168.
188 TADEUSZ SKORUPSKI

four noble truths, the thirty-seven limbs of enlightenment, emptiness and compassion,
and the Buddhas personal attributes and powers.
The Buddhas presence in his stpas is a recurrent subject of Buddhist deliberations
and debates, but essentially all such deliberations form an integral part of a much wider
spectrum of Buddhist speculations and theories, which are concerned with the Buddhas
identity, presence and activities in this world after his parinirva. The ultimate goal of
true Buddhists is to eliminate the cycles of rebirth, and to abandon the realm and toils
of sasra, and the Buddha was the first one to have attained the final emancipation
from the clutches of sasra. After his demise, his followers explored various ways of
explaining the Buddhas nature, and also the identity of his relics enshrined in stpas,
which they treated with veneration. As we have seen some strands of early Buddhism
maintained that the Buddha ceased to be present in this world after his parinirva, and
some believed in his tangible presence in the relics. The Mahyna adepts produced
a fairly coherent and tenable theory of the three buddha-bodies, which enabled them to
retain the Buddhas presence in the world on mundane and supramundane levels. In their
form-bodies (rpakya) the countless Buddhas are active in countless world systems. All
Buddhas jointly abide in the absolute body (dharmakya), which constitutes their common
ground. Essentially this body epitomises the quintessence of buddhahood; it is formless,
and beyond human comprehension and verbal expressions. In addition according to the
Mahyna position, the dharmakya as the buddha-potentiality is present in all sentient
beings. Thus although the Buddha has passed away, the nature of buddhahood continues
to operate in the world on different levels and in different forms. This being the case,
it can be said that the relic stpa does clearly epitomise the Buddhas mystic presence
in this world.
As already stated above, the stpa is not a grave or burial site, but it is the
monument expressive of the Buddhas enlightened mind and parinirva. Then again
it is a depersonalised architectural icon of the Buddhas immanent and transcendent
abiding among his followers. Finally, we note that some texts also suggest that the
stpas architectural configuration resembles or epitomises the Buddhas seated
position.

The Buddhas imageless icons

Since the Buddhas human body and personality were utterly dissolved, the immediate
generations of his followers did not produce his iconic or anthropomorphic representations
for a considerable period of time. It is generally agreed among the Buddhist scholars that
the first anthropomorphic images of the Buddha were invented and produced around the
beginning of the Common Era.
Initially in the early sculptures, the Buddha is not represented in iconic or
anthropomorphic forms, but his presence and deeds are represented by an empty space,
which becomes transmuted and makes itself visible in the form of symbols. These uniconic
THE BUDDHAS STPA AND IMAGE. THE ICONS OF HIS IMMANENCE AND TRANSCENDENCE 189

symbols epitomise the highest truth realised by the Buddha, and they are expressive
of his true nature as being identical with the supreme truth, the Dharma. It is not the
Buddhas human body or personality that preoccupied the minds of his immediate and
later followers, but it was his enlightened identity with the Dharma. The Buddha said:
the person who sees the Dharma, sees the Buddha, and again the person who sees the
Buddha, sees the Dharma.33
The uniconic or symbolic depictions of the Buddha mostly epitomise the specific events
or deeds, which he experienced or performed during his lifetime. The Buddhas nativity
is symbolised by the goddess Lakm, or by his mother, the queen My, leaning against
a tree in the Lumbin grove. The attainment of enlightenment at Bodhgay is articulated
by the bodhi tree and the empty diamond seat (vajrsana) beneath the tree. The first
sermon is indicated by a wheel (cakra), the emblem of a great being or universal monarch
(cakravartin), and the parinirva is symbolised by the stpa. Finally the Buddhas
presence is also marked or indicated by footprints. Thus all these symbolic representations
serve as the abstract or indicative icons of the Buddhas life events and activities. In
other words they do not epitomise the Buddha as a person, but they are expressive of
the truths and deeds, which he realised and performed.
None of the above symbols, which represent the Buddhas principal life events, can
be considered or treated as a single or unique and universal symbol of the Buddhas
person or his activities. Considered individually, each symbol encapsulates and depicts
one specific event or deed. Considered together, they do denote the Buddhas doctrinal or
perhaps mystic image, but they do not project the Buddhas image as a person, because
his body and personality were dissolved at the time of his parinirva.34
As all or most of the symbolic representations are linked with the Buddhas specific
buddha-events, and the places where he performed his buddha-deeds, they denote the
Buddhas mystic dimensions, without taking into account his involvement in the sphere
of sasra. The Buddha did live and teach in this world and for the sake of this world,
but his early and later followers disregarded the Buddhas mundane humanity, and
focused on his transcendent and universal character. Essentially the Buddhas followers
had no interest in the mundane personality of the Buddha, but they consistently and
persistently endeavoured to formulate, expand and enrich, and to depict the buddha-ideal
beyond and above history, time, and embodied reality in any historical and biographical
sense.

33 Sayutta, III.120. Some texts assert that the Tathgatas body is the dharmakya born of morality

(la), meditation (samdhi), wisdom (praj), liberation (vimukti), and knowledge and vision of liberation
(vimuktijnadarana). These are the five pure (ansrava) or transcendental (lokuttara) skandhas. Dgha, III.279;
Sayutta, I.99; Aguttara, I.162; stra, III.1349, 1384.
34 For an insightful interpretation of the above symbols, and indeed for the origin and iconographic developments

of the Buddha image in India and other Buddhist lands of Asia, see D.L. Snellgrove, ed., The Image of the Buddha.
See also Dietrich Seckel, 1989, 180: The Symbolizing Principle.
190 TADEUSZ SKORUPSKI

The Buddhas anthropomorphic images

As already mentioned, it is broadly agreed among Buddhist scholars and art historians
that the first anthropomorphic images of the Buddha appeared in sculptures during the
initial centuries of the Common Era. Gandhra and Mathur are considered as the
two principal regions in which the first Buddha images were intellectually conceived
and concretely sculptured in stone. Stylistically, the Buddha images produced in these
two regions, are differentiated as respectively the Gandhra and Mathur styles, or as
appertaining to the Gandhra and Mathur schools.
Once again, the Buddhas followers and the artists whom they commissioned did not
seek to depict the Buddha in anthropomorphic or human forms, which would in any way
approximate to the appearance he may have had during his lifetime. The Gandhra and
Mathur schools showed no interest in the mundane appearance of the Buddha, but instead
they strove to depict the idealised or superhuman Buddha, drawing on two distinctly
different sources of inspiration. We cannot discuss here the details of these sources of
inspiration, but we only sketch their principal theoretical and cultural configurations.
The Buddhas sculptures produced in Gandhra are characterised as being inspired or
influenced by the Hellenistic perceptions and visions of the Olympian gods, and perhaps
also the Greek emperors. In a nutshell, in the Gandhra sculpture, the Buddha is not
represented as a bald-headed mendicant, but as a royal and divine figure deprived of his
human attributes. The Buddhas body is enveloped in a plentiful monastic cloak, which
markedly resembles the himation of the ancient Greeks, or the toga of the ancient Romans.
His hair is arranged in wavy strands knotted together on the top of the head. This rich
mound of hair is reminiscent of the hairstyle of Greek gods. He often wears masterly
groomed moustache and beard. The overall appearance is realistic, but at the same time
it projects itself as a superhuman and super-divine figure worthy of respect and devotion.
The Mathur school largely drew its inspiration from the literary sources about the
Buddha as the Lord (bhagavn), universal monarch (cakravartin), and great being or
superman (mahpurua). According to the relevant texts, the great being is described as
having on his body thirty-two major marks (lakaa) and eighty minor marks (anuvyajana)
of distinction.35 In the Mathur sculptures, the Buddhas head is covered with clusters of
short hair curling to the right. He has elongated ear-lobes, almond-like eyes, and perfect
facial features. He is cloaked in his buddha-robe, and when seated, he sits on a lion
throne (sihsana). Once again in a nutshell, the overall appearance projects a majestic
and superhuman figure, exuding tranquillity, supremacy, and transcendental aloofness.

35 For the lists of the thirty-two lakaas and eighty anuvyajanas, see stra, I.271-79; Lakkhaasuttanta,
Dghanikya, III.143-44; Mahvyutpatti, 236267, 269349. There exist legendary accounts, which allege that the
first images of the Buddha were produced during the Buddhas life, and under his authority. For a discussion of
these legends see C. Wickramagamage, 1984, 249255.
THE BUDDHAS STPA AND IMAGE. THE ICONS OF HIS IMMANENCE AND TRANSCENDENCE 191

While the Gandhra style images eventually discontinued to be produced, the Mathur
style continued and inspired the subsequent generations of buddha-images in India and
beyond.
Just to reiterate, the anthropomorphic images of the Buddha do not reflect the image
of the historical Buddha as a human being, but they projects the iconic image of idealised
and abstract buddhahood.

Is there the ultimate Buddha-icon?

Just as in the case of aniconic or symbolic representations, essentially there is no


single image or icon of the Buddha, which can be treated as the ultimate icon. There is
a whole series of images, which epitomise the Buddhas life events and his activities:
birth, enlightenment, first sermon, and so on.
Then again, as Buddhism asserts the existence of countless Buddhas, in addition
to the images of kyamuni Buddha, we have images of other Buddhas who preceded
him such as the Buddha Dpakara, and of the Buddhas described in Mahyna texts,
such as Amitbha, Akobhya, Bhaiajyaguru, and so on. We also have multiple images
of the Buddhas of the ten directions, and of the so-called thousand Buddhas. Finally in
the tantras, we have the fivefold manifestation or icon of buddhahood, as represented by
the five cosmic Buddhas, Vairocana and the other four Buddhas, who epitomise the five
aspects of the buddha-omniscience and activity. In the tantras, we also have the collective
icons or manifestations of buddhahood arranged in maalas. As diagrams the maalas
epitomise an ideal and perfected universe, which is replete with Buddhist deities, who
collectively encapsulate the totality of buddhahood. Individually all those deities symbolise
particular aspects of the buddha-qualities and activities, or of the Buddhist teachings.
Thus, in Buddhism there are single and multiple icons, but it is difficult to identify
and indicate any individual and specific icon as the ultimate icon or image expressive
of the all-encompassing silhouette of kyamuni Buddha.

Relationship between the stpa and the image

In relation to buddha-images, the stpa has always commanded, and continues to


command, the position and status of primacy and pre-eminence. As already indicated
above, the stpas superiority derives mainly from the fact that it contains the Buddhas
relics. Right from the very beginnings, the Buddhist way of life and devotion were
centred on the worship of the relics, tangibly reminiscent of the Buddha as the ideal of
human perfection, although significantly removed into the realm of idealised abstraction.
The images played, and continue to play, the important role in the worship of the
Buddha as tangible expressions and the locus of devotional focus directed towards the
idealised state of buddhahood. But they never dislodged the centrality of the stpa.
192 TADEUSZ SKORUPSKI

The stpas with images inserted on their sides, as it were combine the distinctive roles
and functions of the Buddhas stpa and image. Some of the most powerful images, like
the stpas, contain the Buddhas relics. Thus the Buddhas relics unite and underpin the
mystic sacredness and potency of the Buddhas stpa and image. In this configuration,
it could be postulated that the stpa and the image, as a combined icon, epitomise the
Buddhas mundane and transcendent permutations. However, if we must reduce the things
to one or zero, then perhaps the relics should be identified and envisaged as the ultimate
icon of the Buddha in this world.

Concluding reflection

According to the perfection of wisdom (prajpramit),36 conventionally or from the


perspective of human truths, there are beings, mendicants, holy people, arhats, Bodhisattva
and Buddhas. However, from the perspective of the ultimate truth, there are no living
beings, no arhats, no Bodhisattvas, and no Buddhas. Thus ultimately, all mundane and
supramundane images of any kind arise from emptiness, but in order to gain their true
apperceptions one must dissolve them in emptiness, the ultimate imageless reality of all
imagined phenomena, beings and icons.

Bibliography

BEFEO Bulletin de lEcole Fraicaise dExtreme Orient


CHJ Ceylon Historical Journal
EFEO Ecole Fraicaise dExtreme Orient
HR History of Religion
IHQ Indian Historical Quarterly
JAOS Journal of the American Oriental Society
JBRS Journal of the Bihar Research Society
JIABS Journal of the International Association of Buddhist Studies
JICPBS Journal of the International College for Postgraduate Buddhist Studies, The International College
for Postgraduate Buddhist Studies, Tokyo
MASC Memoirs of the Archaeological Survey of Ceylon
MCB Melanges Chinois and Bouddhiques
MRDTB Memoires of the Research Department of the Ty Bunko
PUP Princeton University Press
TJ The Tibet Journal

Bareau, A. 1962. La construction et le culte des stpa daprs les Vinayapiaka, BEFEO, 50,
pp. 229274.
Benisti, M. 1960. Etude sur le Stpa dans lInde Ancienne, BEFEO, pp. 37115.

36 Pacaviati, 261.8-13.
THE BUDDHAS STPA AND IMAGE. THE ICONS OF HIS IMMANENCE AND TRANSCENDENCE 193

Bentor, Y. 1995. On the Indian Origins of the Tibetan Practice of depositing Relics and Dhras in
Stpas and Images, JAOS, pp. 248261.
Bentor, Y. 2003. The Content of Stpas and Images and the Indo-Tibetan Concept of Relics,
TJ, 28.1-2, pp. 2146.
Bentor, Y. 1994. Tibetan Relic Classifications. In: Kvrne, P. (ed.), Tibetan Studies, Proceedings
of the 6th Seminar of the International Association for Tibetan Studies, Fagernes 1992, vol. 1, Oslo: The
Institute for Comparative Research in Human Culture, pp. 1630.
Cayton, L.J. 1996. The Buddhist Stupa: Its History, Dimensions and Symbolism According to Tibetan
Sources, PhD dissertation, University of Wisconsin.
Chen, K., Buddhism in China, A Historical Survey, Princeton: PUP, 1964.
Chapell, D.W. 1980. Early Foreboding of the Death of Buddhism, Numen, 27, pp. 122154.
Dallapiccola, A.L.; Zingel-Av Lallemant, S. (eds). 1980. The Stpa, Its Religious, Historical and
Architectural Significance, Wiesbaden: Franz Steiner Verlag.
Dikshitar, V.R.R. 1938. Origin and Early History of Caityas, IHQ, 14, p. 441.
Divyvadna, ed. E.B. Cowell and R.A. Neil, Cambridge, 1886.
Fa-Hsien. 1965. A Record of Buddhist Kingdoms, translated by J. Legg, New York: Dover Publications.
Faure, B., Dato, Hbgirin, Paris/Tokyo, 2003, vol. 8, pp. 11271158.
Foucher, A. LArt Grco-bouddhique du Gandhra, tome I, Paris: Imprimerie Nationale, 1905; tomeII,
Paris: Imprimerie Nationale, 1922.
Foucher, A. 1949. La Vie du Bouddha daprs les Textes et les Monuments de lInde, Paris.
Fussman, G. 1986. Symbolism of the Buddhist Stpa, JIABS, 9.2, pp. 3753.
Hinuber, Oskar von. 2009. Cremated like a King: The Funeral of the Buddha within the Ancient
Indian Cultural Context, JICPBS, pp. 3366.
Hirakawa, A. 19631964. The Rise of Mahyna Buddhism and Its Relationship to the Worship of
Stpas, MRDTB, 22, pp. 57106.
Kajiyama, Y. 1985. Stpa, the Mother of Buddhas, and Dharma-body. In: Warder,A.K. New Paths
in Buddhist Research, Durham: The Acorn Press, pp. 916.
Koa, Abhidharmakoa and Bhya of Vasubandhu, with Sphurth of Yaomitra, edited by S.D.Shastri,
Varanasi, 19701972.
Lamotte, E. 1988. History of Indian Buddhism, Louvain-La-Neuve.
Malalasekera, G.P. 1960. Dictionary of Pli Proper Names, volumes III, London.
Martin, D. 1989. Crystals and Images from Bodies, Hearts and Tongues from Fire: Points of Relic
Controversy from Tibetan History, Tibetan Studies, Narita, pp. 183191.
Martin, D. 1994. Pearls from Bones: Relics, Chortens, Tertons and the Signs of Saintly Death in
Tibet, Numen, 41, pp. 273-324.
Nidnakath (of the Jtakahakath), translated by N.A. Jayawickrama, The Story of Gotama Buddha,
Oxford: PTS, 1990.
Paranavitana, S. 1946. The Stpa in Ceylon, MASC, vol. V, Colombo.
Roth, G. 1968. Remarks on the Stpa-Lakaa-Krik-Vivecanam, JBRS, Professor S.H. Askari
Felicitation Volume, Patna, pp. 3146.
Roth, G. 1980. Symbolism of the Buddhist Stpa. In: Dallapiccola, A.L.; Zingel-Av Lallemant,S.,
pp. 183210.
Ruelius, H. 1980. The Stpa in the ilpastras and the Rituals of the Sinhalese. In: Dallapiccola,A.L.;
Zingel-Av Lallemant, S., pp. 267276.
stra = Mahprajpramit-stra, E. Lamotte, tr., La Trait de la Grand Vertu de Sagesse,
Mahprajpramitstra, vol. I, 1944; vol. II, 1949; vol. III, 1970.
Schopen, G. 1997. Bones, Stones, and Buddhist Monks, Honolulu: UHP.
Seckel, D. 1989. The Buddhist Art of East Asia, Bellingham: Western Washington University Press.
Siyuki, Buddhist Records of the Western World, vols. III, tr. by S. Beal, London 1884, reprinted
Delhi 1969.
194 TADEUSZ SKORUPSKI

Skilling, P. 2005. Cutting Across Categories: The Ideology of Relics in Buddhism, Annual Report
of the International Research Institute for Advanced Buddhology at Soka University for the Academic Year
2004, Tokyo: Soka University, pp. 269319.
Skorupski, T. 2001. Two Eulogies of the Eight Great Caityas. In: The Buddhist Forum, VolumeVI,
Tring, pp. 3755.
Skorupski, T. 2002. Kriysagraha, Compendium of Buddhist Rituals, Tring.
Snellgrove, D.L. (ed.). 1978. The Image of the Buddha, Paris: UNESCO.
Strong, J.S. 2004. Relics of the Buddha, Princeton and Oxford: PUP.
Strong, J.S. 1983. The Legend of King Aoka, Princeton: PUP.
Trainor, K. 1997. Relics, Ritual, and Representation in Buddhism, Rematerializing the Sri Lankan
Theravda Tradition, Cambridge: CUP.
Wickramagamage, C. 1984. The Origin of the Buddha Image. In: Buddhist Studies in Honour of
Hammalava Saddhtissa, ed. by G. Dhammapala, R. Gombrich & K.R. Norman, Sri Lanka: University
of Sri Jayewardenepura Nugegoda, pp. 249255.
R O C Z N I K O R I E N T A L I S T Y C Z N Y, T. LXV, Z. 1, 2012, (s. 195211)

ELLIOT SPERLING

Pho-lha-nas, Khang-chen-nas,
and the Last Era of Mongol Domination in Tibet

Abstract

The lives of Pho-lha-nas and Khang-chen-nas are well known, most importantly from
Luciano Petechs seminal study of early 18th-century Tibet. Since the publication of that
study, the appearance of several previously inaccessible sources has allowed us to form
a fuller image of the situation in Tibet before and during their lifetimes, a period that
coincided with the last era of Mongol dominion in Tibet. The effects of this domination
and the concomitant integration of Tibetans and Mongols in military, political and other
spheres of Tibetan life were observable decades after the end of Pho-lha-nass rule. As
a result, while the Dga-ldan pho-brang government of the Dalai Lamas is a useful lens
through which to view Tibets history at this time, it is equally useful to construct a view
of Tibet as simultaneously a Mongol realm, a Qanate of Tibet.

Keywords: T
 ibet, Mongolia, Chinese, archival materials, 18th century, political relations,
Jungar

[T]he Grand Secretariat officials [Tib. krung-thang < Ch. zhongtang ] had
never seen lay Tibetans and so asked us What custom is this, to wear earrings
of two different sorts on the left and the right? As I was in the place of first
rank I answered by explaining that the wearing of a turquoise was an old Tibetan
custom and the wearing of a pearl a custom that spread when those of the royal
lineage of the Koko Nor king Guri Qan [Tib. Mtsho-sngon-gyi Rgyal-po Gau-shri
Khang], were kings of Tibet.1

1 Elliot Sperling, Awe and Submission: A Tibetan Aristocrat at the Court of Qianlong, International Review

of History, vol. 20, 1998, p. 329.


196 ELLIOT SPERLING

This passage from Rdo-ring Bstan-dzin dpal-byors account of his sojourn in Beijing
in 17921793 a century and a half after they heyday of Guri Qan first struck me
when I encountered it in the Rdo-ring paitai rnam-thar several years ago for what it
indicated regarding the impact of the Mongol presence in Tibet that had effectively begun
with the campaigns of Guri Qan in the 1640s to establish the power and authority of
Dga-ldan pho-brang, the government of the 5th Dalai Lama, over Tibet. The long lasting
consequences of this particular Mongol episode in Tibetan history, just in terms of its
effect on Tibetan society, were further underlined by an incident (which I have described
elsewhere) in which Bstan-dzin dpal-byor, whose career was almost exclusively within
the bounds of Tibet and the Nepalese borderlands, came face to face with the Qianlong
Emperor and had little difficulty conversing with him in Mongol.2
The general impression that many people have of this Mongol era in Tibetan history
is (to put it rather simplistically) of a period in which the government of the Dalai Lamas
came to dominance over Tibet through an effective alliance with the Qoot Mongols
under Guri Qan, the latter providing the military force that brought victory to Dga-ldan
phobrang. Then, after a few years, that force faded into the background until Lajang
Qan reasserted the authority that had previously belonged to the Qoot. A Junar invasion
followed, only to be defeated by Tibetan forces led by Mi-dang Pho-lha Bsod-nams
stobs-rgyas (i.e., Pho-lha-nas) and Khang-chen-nas in alliance with the Qing. Thereupon
Mongol influence in Central Tibet came to an end, albeit with a continued Mongol
presence and Mongol power in parts of A-mdo.
While this general outline retains some currency, the texture of Tibetan history as we
experience it in relevant texts of the time gives a distinctly different feeling. Indeed, the
Mongol element is much stronger than the outline actually suggests. I would propose,
therefore, that we may do well to imagine this era somewhat differently: to see Central
Tibet within the context of Mongol dominion and military power for a longer period
and in a manner that was more pervasive in Tibetan society than previously imagined.
Let me put this somewhat differently. There is nothing controversial in stating that one
can write a valid history of India from the mid 18th to the mid 19th centuries from the
standpoint of Fort William and Calcutta, while maintaining that one may also write
a valid history of India during the same period from the standpoint of Agra and Delhi.
Both histories would be valid in their own ways; they would be different, of course, but
nevertheless compelling and useful.
So too, one can write a history of Tibet from the mid-17th to the mid-18th centuries
from the standpoint of the Potala that would be valid; and compose another one from
the standpoint of the Dam encampments of the Qoot that would be equally valid. One
would be a history of Dga-ldan pho-brang, the other a history of the Qanate of Tibet.3

2 Ibid., pp. 331332.


3 I would like to acknowledge the input of my colleague, Prof. Christopher Atwood, who suggested this
useful term.
PHO-LHA-NAS, KHANG-CHEN-NAS, AND THE LAST ERA OF MONGOL DOMINATION IN TIBET 197

This essay, focusing on aspects of the careers and family histories of Pho-lha-nas and
Khang-chen-nas is written with the latter notion specifically in mind.
The penetration of Central Tibetan society by elements of Mongol rule is best
acknowledged by understanding that the dominant role of the Qoot brought with it
processes of integration between Tibetans and Mongols. There was no hard and fast
divide between the two at many levels. In this regard the extent of Indian participation
in the machinery of British rule over India provides a useful point of comparison. The
relatively few Mongols in Tibet, like the relatively few British in India, were highly
dependent on personnel from the peoples whom they dominated in order to maintain
a working state. The integration of Tibetans into the forces of Guri Qan, forces that
are generally described simply as Mongol, is obvious from the material on Mi-dbang
Pho-lhas family found in his biography, the famous Mi-dbang rtogs-brjod by Mdo-khar
zhabs-drung Tshe-ring dbang-rgyal. Therein we find that the ties between the Pho-lha
clan and the Mongol forces of Guri Qan date from before Mi-dbang Pho-lhas birth, to
the time of his grandfather:

[12] Then, at that time, in order to destroy the powerful Karma


bstanskyong in the area of Nyang in Gtsang, who was made arrogant
by pride, an innumerable body of troops of Guri [Tib. Gu-shr] Bstan-
dzin chos-kyi rgyal-po, the bodhisattva and marvelous manifestation of
the wisdom of the mahsattva Vajrapi, [13] set out for Bsam-grub-rtse
at Gzhis-ka-rtse [Tib. Gzhis-ka Bsam-don grub-pai rtse] via the region
of Gling-dkar. And when they did so, the courageous Dar-rgyas [i.e.,
Pho-lhas great-uncle] calmly took up the duty, riding in the front rank
before the troops. In the place called Mang-ra lug-gdong they started
to cross weapons with the sentinels of the Gtsang forces. At that time,
although a few who comported themselves with cowardice fled, Dar-
rgyas, the ancestor, retained possession of himself and without hesitation
roused his courage and attacked, thereby inflicting defeat on the other
side, a thing which greatly pleased the Mongol lord. As the first mark of
his perfect gratitude he bestowed on him a garment most fitting to wear,
which was like something made from the essence of a molten moon.
Not long afterwards the drumbeat of his triumphs in battle resounded
victoriously to the utmost heights and he was rewarded and praised and
by order he was made the commander of eighty armed [i.e., go-mtshan
= go-mtshon] infantry.4

4 Zhabs-drung Tshe-ring dbang-rgyal, Mi-dbang rtogs-brjod (Chengdu, 1981), pp. 1213: dei tshe dei dus

na yul Gtsang Nyang gi ljongs su stobs kyi dbang phyug Karma bstan skyong dbang po dregs pas mngon par
gying ba tshar gcod pai slad du byang chub sems dpa sems [13] dpa chen po Lag na rdo rjei ye shes kyi sgyu
phrul gu shr Bstan dzin chos kyi rgyal poi dpung gi tshogs bgrang du med pa yul Gling dkar nas brgyud de
gzhis ka Bsam don grub pai rtse mor chas pai tshe snying stobs dang ldan pa Dar rgyas kyis dpung gi sngon
du gro bai bya bai khur bzod par byas pa las / Mang ra lug gdong zhes bya bai sai char yul Gtsang ljongs
198 ELLIOT SPERLING

This integration of Tibetans into the Mongol military establishment is very


much a background element in the general history of the Pho-lha clan. Pho-lha-nass
greatgrandfather, Orgyan, had two sons, Dar-rgyas, just mentioned, and Gsum-dga. And
Gsum-dga (more commonly referred to as A-gsum), served the Qoot too, in his case
taking up an official post on the fringes of the victorious Shangs region.5 During this
period Dar-rgyas continued his rise:

[13] At this time in Nags-rong in the east the Bka-brgyud-pa rulership


[Tib. Dkar-srid] in Dwags[-po] and Kong[-po] was crooked and as aresult
no respect was shown to the lord [Guri Qan]. Then, at the time in which,
in the manner of the unsullied Ljon-shing tree, what had been [previously]
neglected and which then began to be provided for, the sprouts [i.e. the
younger members of the line] of the lineage of the Mongol lord, [14]
along with the one known as Star-sdong-pa Gu-ru who was active near
the mountains on the south shore of the flowing Skyid-chu, went before
the ruler. At the same time, Dar-rgyas, the family elder, went there with
his attendants, and afterwards his deeds in battle were supremely brave,
courageous and skilled. In particular, at a terrifyingly narrow and difficult
part of the route, in a deep defile, he joined with the enemy in battle.
And at that time, bestirred by the divine and incorruptible lineage of his
ancestors, filling the space between his ears [and the bow] with wild and
glorious arrows he shot into them again and again, and sent some ten of
the wild and aroused [enemy] on the path to impermanence. Thus, with
marvelous bravery and heroics he won the admiration of the King of
Tibet, the Earthly Brahma, and ornate presents were given and a gracious
edict promulgated that as a dbang under the assembly he was at no time
to be afflicted with taxes and was to dwell in comfort!6

kyi dpung gi mel tshe ba dag dang mtshon chai las rtsom pa dres par gyur te / dei tshe sdar mai brtul zhugs
dzin pa ga zhig ni byer bar gyur mod / mes po Dar rgyas dpa brtul dang ldan pa des ni tsham tshom bral bar
mthu rtsal bskyed de rgol bas pha rol de dag pham par byas pa las Mong gol gyi rje bo yang rab tu dgyes pa
bskyed par gyur te gtang rag legs soi dge mtshan dang por zla ba zhun mai khams las grub pa lta bui gyon
pa mnabs su rung ba yang bstsal / dus de nas ring min gyi char g.yul las rnam par rgyal bai rgyal rnga srid
rtsei bar du snyan par sgrogs pa na di nyid la mngon par gzengs bstod de dpung bu chung go mtshan can
bcu phrag brgyad kyi kha lo sgyur bai byed por lung gis mngags / /. N.b., Luciano Petech, China and Tibet in
the Early XVIIIth Century (Leiden, 1972), p. 283 refers to Dar-rgyas as A-sum, most likely a confusion with his
brother, Gsum-dga, i.e., A-gsum. On Bsam-grub-rtse as a former name for Gzhis-ka-rtse, see Alfonsa Ferrari,
Mkyen Brtses Guide to the Holy Places of Central Tibet (Rome, 1958), p. 144.
5 Zhabs-drung Tshe-ring dbang-rgyal, op. cit., p. 13: Shang phyogs rnam par rgyal bai gling gi sne mo. On

the Shangs area, which is marked by the course of the Shangs-chu that flows into the Gtsang-po north-west of
Gzhis-ka-rtse, see Ferrari, op. cit., p. 159.
6 Zhabs-drung Tshe-ring dbang-rgyal, op. cit., pp. 1314: dus der yul shar phyogs Nags rong na spyod pa

Dwags Kong dkar srid dang bcas pas gya gyui sbyor bas rje bo la gus par mi blta bar / grib ma med pai ljon
shing bzhin du yal bar dor bai du god rtsom pa dei tshe Mong gol gyi rje bo dam pa dei rigs kyi myu [14]
gu dag dang / dal bab Skyid chui gram ngogs lho phyogs kyi ri boi gam na spyod pa Star sdong pa Gu rur
PHO-LHA-NAS, KHANG-CHEN-NAS, AND THE LAST ERA OF MONGOL DOMINATION IN TIBET 199

Indicative of the confidence that the family was gaining, A-gsum and others were
covertly placed in charge of several strategic areas within the area of Nyang where, not
surprisingly, those previously in authority were little trusted by the Qoot regime that
had defeated the local rulers.7 Sometime later, A-gsum came to the attention of Guri
Qans sixth son, Dalai Qongtaiji,8 who dominated the Kokonor region and played an
important role in Tibetan affairs in the late 17th century.9 Thus, we read further in the
Mi-dbang rtogs-brjod:

Afterwards, when the lord of the Mongol lands [i.e., the Kokonor lands
of the Qoot], known as Dalai Qongtaiji [Tib. Da-las Hor-(sic) thai-ji],
made exceptional efforts in working to increase the accumulated merit and
wisdom that lay within the Land of Snows, he commanded [A-gsum] to
shoulder the burden of a friend who would show him which customs were
to be retained and which were to be rejected, [so that] then there would
be nothing misguided in the stages of any actions that he might wish to
undertake. The opportunity thus opened made for a complete meeting
of the minds with the lord, and he and the Mongol lord behaved in the
particular ways of warm friendship, as if they were the closest of relatives,
holding to the wisdom of Brahma in the round of activities. Thus, the
regard of the All-Seeing Lord of the Jina [the 5th Dalai Lama] for the great,
fearful earth-rending one [Dalai Qongtaiji] grew tremendously. And then
in gratitude [A-gsum] was granted the house of Pho-lha with its bound
subjects, like a fine granule born from the womb of the earthly meeting
place of the nga lords, to have and use as his own.10 And a full edict
was promulgated saying If you become tired of this place and tension
arises, just say so and youll be granted a home and subjects to the extent

grags pa dbyings sa gong mai mdun na don dang chabs cig tu mes po Dar rgyas kyis kyang zham ring pai
tshul du bskyod nas g.yul gyi bya ba la mchog tu dpa bai snying stobs rngo thogs pa dang / lhag par lam gyi
phrang nyam nga ba bgrod dka ba gram ngogs na gcong rong zab pa zhig gi sai mdun du pha rol gyi rgol
ba dang g.yul bsres pai tshe / mes kyi gnam ru sra ba rgyud kyis bkug pa bdungs pa las nyag phran gyi mtshon
cha rtsub mo dpal dang ldan pa rna bai bar du bkang ste yang dang yang du nye bar bsnun pa las / gdug can
mi srun pa bcu phrag tu nye ba mi rtag pai lam du bkod / de lta bu dpa zhing rtul phod pai ngo mtshar gyis
Bod rje sai Tshangs pai thugs phrogs te tshong dus og gi dbang rigs kyi cha las dus dang dus ma yin par
dpya dang khral gyis nye bar gtser pa med pa ci bder gnas bcar yod pai bka yig phyag rtags kyis rnam par
spud pa legs soi skyes kyi bka drin du stsal / /.
7 Ibid., pp. 1415. Nyang refers to the region marked by the course of the Nyang-chu, which flows through

Rgyal-rtse and on towards Gzhis-ka-rtse. See Ferrari, op. cit., pp. 59 and 142143.
8 For his place among the sons of Guri Qan, see Dung-dkar Blo-bzang phrin-las, Dung-dkar tshig-mdzod

chen-mo (Beijing, 2002), pp. 507508 and Sum-pa Ye-shes dpal-byor, Chos-byung dpag-bsam ljon-bzang (Lanzhou,
1992), p. 1000.
9 See Zahiruddin Ahmad, Sino-Tibetan Relations in the Seventeenth Century (Rome, 1970), pp. 66, 148,

208209, 225, 241, and 251.


10 On the location of Pho-lha, west of Rgyal-rtse and west of the Nyang-chu, see Petech, op. cit., p. 27 and

Turrell V. Wylie, The Geography of Tibet According to the Dzam-gling rgyas-bshad (Rome, 1962), p. 143.
200 ELLIOT SPERLING

that you desire in Dbus, the land of your birth, and a seal ornamented
with a beautiful picture was given to him. And he visibly gasped.11

And so it is that the installation of the family as the Pho-lha lineage, with the grant
of lands and subjects, ultimately derives from its place within the Mongol presence in
Tibet in the wake of Guri Qans campaigns. As noted, the general outline of these
events is not unknown, but there are nuances that make for a better understanding of
the significance of the Mongol presence in Tibet during this period. Thus, we can note
that Luciano Petech mentions that when Mi-dbang Pho-lha later gathered his forces in
Western Tibet in the wake of Khang-chen-nass murder in 1727, he could rely on his local
popularity there because since his birth he had been regarded as the incarnation of the
Mongol lama dGa-ldan tse-dba who had conquered ma-ris for Tibet in 16781683.12
This association with Dga-ldan tshe-dbang (or better Galdan Tsewang) needs to
be understood a bit more fully. Mi-dbang Pho-lha was not simply rumored to be an
incarnation of Galdan Tsewang, he asserted as much himself.13 Indeed, this connection
is given prominence in the section of his biography describing his birth and first years:

Regardless of what was related previously as to how thoughts arose as to


whether he was another [rebirth of] Galdan Tsewang Paljang [Dga-ldan
Tshe-dbang dpal-bzang-po],14 who stood above others, not only were the
sacred instructions that he was that very lord of men true and evident,
from the time of his birth he had memories of [his former] courage and
daring He was more courageous, more articulate and stronger than
other children and particularly attached to the sort of people who were
ordained and were Mongols. He was especially fond of weapons15

11 Zhabs-drung Tshe-ring dbang-rgyal, op. cit., p. 16: dei phyi nas yul Mong gol gyi rje bo Da las Hor thai

jir grags pa gangs can gyi ljongs na bsod nams dang ye shes kyi tshogs spel bai las mngon par du byed par
lhags pa na / de la yul lugs kyi blang dor ston pai grogs su bka bsgos pa khur du bzod par byas nas / ci dang
ci dod pai bya bai rim pa ma khrul zhing / go skabs phyed pa rje boi bsam pa dang mthun pa legs par tshar
phyin pa bgyis pa las Mong gol gyi rje bo dang lhag par yid gcugs pa mdza bshes smon drin lta bur spyod
par gyur pa sogs bya ba las kyi khor lo la Tshangs pai blo gros dang byed pa chang bas Thams cad gzigs pa
Rgyal bai dbang po dang / sai byer jigs chen poi gzigs pa je cher gyur nas gdengs can dbang po bor rgyas
kyi dun sa nor dzin mai lhums nas skye bar ldan pa dang mtshungs pai brum Pho lha khang bzang mnga ris
dang bcas pa rang gir spyod pai bka drin bstsal zhing / gzhan yang di skad ces / gal te khyod nyid yul khor
de nyid la yid gdung bar gyur pa dang / bag mi phebs pai rnam pa shar na smros shig dang / slar yang btsas
bai yul Dbu rui phyogs nas khyim dang bangs ji dga ba dod pa jo bar byao zhes bkai yi ge gtsang ma
phyag rgyai re kh bkra bas mtshan pa byin nas mngon par dbugs dbyung ngo / /.
12 Petech, op. cit., p. 120.
13 Zhabs-drung Tshe-ring dbang-rgyal, op. cit., p. 87.
14 See his biography in Dung-dkar, op. cit., pp. 597600.
15 Zhabs-drung Tshe-ring dbang-rgyal, op. cit., p. 66: de yang gong du ji ltar bsnyed pai khyu mchog Dga

ldan tshe dbang dpal bzang po de nyid gzhan zhig yin nam snyam na de ltar mi blta ste / dam pai lung gis mii
dbang po di nyid yin par bsnyon du med cing gsal bar bstan par ma zad kyi / bltams ma thag pa nas dpa zhing
brtul phod pai dran pa bsnyes byis pa gzhan las brtul zhugs che ba / smra mkhas pa / drag shul che ba / rab
PHO-LHA-NAS, KHANG-CHEN-NAS, AND THE LAST ERA OF MONGOL DOMINATION IN TIBET 201

Then talk arose among those who were ordained and were monks: E-ma!
This child is undoubtedly the rebirth of the infinitely articulate Galdan16

The question arises as to why such attention is paid to the establishment of an


incarnation link between Mi-dbang Pho-lha and this particular Mongol figure. In part,
we can deduce a good deal of reasoning from Pho-lha-nass fathers relationship with
this same Galdan Tsewang, something that in itself is indicative of the way Tibetans and
Mongols were enmeshed together during this period of the Qanate of Tibet. But there is
one element that adds considerably to the significance of the link. It is not just the fact
that Galdan Tsewang was both a fighter and a monk (though not concurrently, contrary
to what Petech implies).17 In his role as the former he was specifically linked with
Gtsang, the region wherein the Pho-lha lineage had been established by Dalai Qongtaiji,
and particularly with Mnga-ris. In his role as the latter, we can note that he was in fact
ordained by none other than the Panchen Lama. Indeed, the account of his ordination
hints at conflicting personal factors (perhaps reflected in the literary description of the
young Mi-dbang Pho-lhas being drawn to both monks and Mongols):

At that time, the one who stood above others, with the wisdom of Brahma,
the qualities of vara, the skills of Vishnu; the great courageous one
skilled at strategy against a fully-enclosed enemy encirclement, the great
hero Galdan Tsewang Paljang, removed himself from the throne, spread
out a long, pure white kha-btags and secretly said I ask the gods to
look down on me. Unlike the many scholars who are naturally born in
Dbus, I am a Mongol from the far borders. But former, deep karmic
traces [of understanding] have grown in my heart to the effect that perfect
worldly wealth cannot be relied on, is impermanent, and is misery by
its nature. And at that he was given vows by the omniscient [Panchen
Lama] Blobzang chos-kyi rgyal-mtshan [15701662].18

tu byung ba dang / yul Mong gol gyi skye boi rigs la lhag par chags par byed cing / mtshon chai bye brag la
gces par dzin
16 Ibid, p. 72: dei tshe rab tu byung ba dang / dge sbyong dag phan tshun gtam gleng ba / e ma bus pa di

ni rab byams smra ba Dga ldan gyi skye ba slar ongs par gdon mi za ba zhig ste
17 See Dungkar, op. cit., p. 598, who notes that when appointed to lead the forces against Ladakh he gave back

his so-thar vows.


18 Zhabs-drung Tshe-ring dbang-rgyal, op. cit., p. 49: dei tshe mii khyu mchog Tshang pai blo gros / Dbang

phyug gi yon tan / Sred med kyi bui rgyu rtsal / khor loi mu khyud lta bui dgra thabs la mkhas pai dpa bo
chen po Dga ldan tshe dbang dpal bzang po rang nyid khri las babs te lha rdzas rab dkar dri ma med pa kha
tshar can srid du ring ba mngon par bstar nas gsung gsang mthon pos di skad ces / lha gzigs su gsol / yul Dbus
mkhas pa byung bai rang bzhin can du ma skyes kyi / bdag ni mtha khob Mong gol gyi yul duo / on kyang
sngon gi bag chags zab mo snying la brtas pai srid-pai dpal byor phun sum tshogs pa ni yid brtan mi rung
ba / rtags pa ma yin pa / sdug bsngal gyi rang bzhin can du shes nas thams cad mkhyen pa Blo bzang chos kyi
rgyal mtshan las rab tu byung /. On this particular Panchen Lama, see Dungkar, op. cit., p. 1258.
202 ELLIOT SPERLING

Galdan Tsewang was unquestionably a crucial figure in the making of an authoritative


Mongol regime in Western Tibet. He was also a crucial figure in the continuing rise of
the Pho-lha clan. His role in the Ladakh campaign was mentioned above. But here as
in many other campaigns the Tibetan participation was significant:

[36] At that time the Mongol forces were skilled at subduing enemies
by reliance on mounted [warfare]. But they were not skilled at moving
on foot to enter and bring down fortresses. Thus, there were sent from
the region of Lhasa those who could apply broad intelligence [to the
task], first and foremost Ngam-ru-ba, with the strong armor of courage;
[37]from the region of Nags-rong, Kong-po A-bo bkra-shis; the father of
the lord of men [i.e., Mi-dbang Pho-lha], Padma rgyal-po; Dgon Bla-ma
phrin-las; and the one known as the rje-drung [noble scion] of the
ruling lineage of Bye-ri Stag rnam-par-rtse. And they approached Mnga-
ris in four divisions of almost 5,000.19

Ultimately the Tibetan and Mongol army inflicted a severe defeat on the Ladakhi
forces, with Padma rgyal-po, Mi-dbang Pho-lhas father, serving as one of the victorious
commanders. Out of gratitude, Galdan Tsewang rewarded him and his minister Bu-chung
liberally, with items from the royal stores of Ladakh.20 From that point on Padma rgyal-po
became one of Galdan Tsewangs close followers:

Although the lord of men Galdan had many subordinates whose practical
application of intellect was broad and whose robust youth matched their
skills, among them the one who was the object of immense and particular
regard was Padma rgyal-po, the father of the excellent lord of men. He was
naturally insightful, his mind was close to that of the lord, and because

19 Zhabs-drung Tshe-ring dbang-rgyal, op. cit., pp. 3637: dei tshe yul Mong gol gyi dpung rnams ni bzhon pa
la brten te dgra bo gzhom par mkhas pa yin gyi / rang nyid kyi rkang pas song nas mkhar rdzong du zhugs pa
bebs pa la mi mkhas pa yin pas / slar yang yul Lha ldan gyi phyogs nas blo gros kyi jug pa yangs pa / snying
stobs kyi go cha btsan [37] po ngam rub ba / yul Nags rong gi phyogs nas Kong po A bo bkra shis / mii bdag
po gang dii yab gcig Padma rgyal po / Dgon bla ma Phrin las / Bye ri Stag rnam par rtse bai sa skyong bai
rigs las rje drung du grags pa rnams gtso bor mngags te yan lag bzhi pai dpung gi tshogs brgya phrag lnga bcur
nye ba yang yul Mnga ris kyi phyogs der lhags so /. Nags-rong may be in the area of Kong-po, as alluded to in
the name Kong-po A-bo bkra-shis, but it might also be a reference to Nags-shod, north of Lhasa and southeast
of Nag-chu; concerning which see Wylie, op. cit., p. 103. As for Bye-ri Stag rnam-par-rtse, cf. Petech, op. cit.,
p. 52, and the reference to sTag-rtse in Bye-ri, which is then identified as Taktse-dsong on the right bank of
the sKyid-cu to the east of Lhasa. Following Petech further, the person in question is likely the zhabs-drung
(= rje-drung) Rdo-rje rnam-rgyal of Stag-rtse.
20 Zhabs-drung Tshe-ring dbang-rgyal, op. cit., pp. 3839.
PHO-LHA-NAS, KHANG-CHEN-NAS, AND THE LAST ERA OF MONGOL DOMINATION IN TIBET 203

he accomplished whatever he was ordered to do, he impressed [Galdan


Tsewang] who cherished him like a brother, loving him like kin21

This description mirrors, not surprisingly, the relationship between Dalai Qongtaiji
and A-gsum. It is worth emphasizing here that the nature of Tibets history at this point
was an intertwined history of Tibetans and Mongols in many spheres, and not simply
a sort of service phase in which the Mongol role was simply to provide support for
Dga-ldan pho-brang. The Mongols predominated, but they also worked with Tibetans
within the order that was being established through Mongol domination. As indicated
by Mdo-mkhar zhabs-drungs comments about the need for Tibetan military skills in
dealing with fortresses, this was by no means an era in which Tibetan power was wholly
passive. Tibetan power stood at the service of a Mongol order, not unlike the massive
Indian power that was once at the service of a British order.
As an indicator of the extent of Mongol-Tibetan integration within the Tibetan
structures, we should turn again to the lineage of Mi-dbang Pho-lha. We have already
noted that his grandfather and great uncle served in the military campaigns waged in
Tibet under Guri Qan and his son Dalai Qongtaiji and that the elevation of the family
as the ennobled family of Pho-lha was born out of his grandfathers service to Dalai
Qongtaiji. But this goes one step further when we look at Mi-dbang Pho-lhas renown
as the incarnation of Galdan Tsewang. This notion did not simply connect Pho-lha to
the martial heritage of the family, it connected him to the lineage of Guri Qan himself,
for Galdan Tsewang, military commander and monk, was the grandson of Guri Qan,22
and Mi-dbang Pho-lha was his incarnation.

***

We find other hints at the extent to which the Mongol Qanate functioned in Tibetan life
and Tibetan affairs from the mid-17th to the mid-18th centuries. In this regard we should
turn now to Mi-dbang Pho-lhas close ally, Khang-chen-nas (or Khang-chen-pa) Bsod-nams
rgyal-po. This well-known figure, who appears in Mi-dbang Pho-lhas biography almost
exclusively with the Mongol title he bore, Daiing Btur, was also tied to Mongol forces.
In his case he was famously appointed to administer Mnga-ris and later, after playing
an important part in the Junar defeat, was made the highest ranking of the bka-blon.23
These facts are well known, as is the story of his assassination, which sparked the civil

21 Ibid., p. 51: de ltar mi rje Dga ldan de nyid la zham ring du spyod pa blo gros kyi jug pa yangs pa dang

sgyu rtsal gyi mkho lag rdzogs pa mang du mchis mod kyi / de dag gi dbus na mchog tu gzigs pa yangs pa ni
mi rje mchog gi yab gcig Padma rgyal po de nyid yin te / de yang rang bzhin gyis yid gzhungs pa / rje bo la blo
ba nye ba / ji ltar bsgo ba de bzhin du sgrub pai ngang tshul gyis yid phrogs nas phu nu lta bur brtse zhing
dungs pa / mdza bshes / smos drin lta bur mnyes gshin par gyur...
22 Sum-pa mkhan-po, op. cit., p. 1000.
23 Dung-dkar, op. cit., p. 295; Petech, op. cit., pp. 61 and 78.
204 ELLIOT SPERLING

war of 17271728 out of which Mi-dbang Pho-lha emerged victorious and supreme.
What is of further interest, however, are some of the Chinese-language materials relating
to the Qing courts involvement in the aftermath of the murder and what they say of
Khang-chen-nass Mongol connections.
The life of Khang-chen-nas is well known to students of Tibetan history, most
importantly from Luciano Petechs study of early 18th-century Tibet. Since the publication
of Petechs monograph on the subject several previously inaccessible sources have become
available. These include Chinese archival materials, now in published form, and two
versions of the Rdo-ring paitai rnam thar. A preliminary study of all of these, combined
with information from already accessible materials (most significantly, the Mi-dbang
rtogs-brjod), allows us to form a fuller image of Khang-chen-nas and of the political
circumstances in Tibet during his lifetime.
The basic outline of Khang-chen-nass life was sketched by Petech.24 Khang-chen-
nas was a member of the Dga-bzhi clan, whose history is a part of the content of the
Rdo-ring paitai rnam-thar, by Rdo-ring Bstan-dzin dpal-byor.25 He was appointed
sgar-dpon of Mnga-ris skor-gsum by Lajang Qan (Lha-bzang) Khang26 and it was in
that post that he attained renown when, in 1719, he attacked and destroyed a Junar force
that was carrying hostages and sacred images taken in the Junar occupation of Lhasa.
This is what the Rdo-ring paitai rnam-thar particularly emphasizes:

[25] As for the one known as Daiing Btur, during the time of His
Holiness the Great Sixth, Rin-chen tshangs-dbyangs rgya-mtsho, he
gradually and incrementally served in government positions such as
rdzong- and estate-level grain and revenue collector. Then, during the
time of the Tibetan king Lajang Qan, when he was serving as sgar-
dpon, the leader of Mnga-ris skor-gsum in the west, the troops of the
Junar Tshe-ring don-grub poured into Tibet and killed Lajang Qan.
Although the Junars held Dbus-Gtsang, in Mnga-ris [Khang-chen-nas]
held his own and did not surrender to them. Afterwards, when the Junars
were unbearably frightened and overcome with awe at the arrival of the
divine troops of the Great Emperor [e.g., Kangxi], fierce sorts, such as
the Junar taiji and jaisang, retreated via Mnga-ris, taking several rare
government treasures, [26] most prominently jeweled objects, as well as

24 See Petech, op. cit., particularly pp. 6165.


25 The Rdo-ring Paitai rnam-thar, is an extremely important source for Tibets history in the 18th century.
Although much of it deals with the career of its author, Bstan-dzin dpal-byor, who played an important role in
the disastrous Tibetan diplomacy connected to the conflicts with the Gurkha state in Nepal between 1787 and
1792, the book is essentially a history of his family, the Dga-bzhi or Rdo-ring clan. Two versions of the text are
available in modern printings; see Dan Martin, Tibetan Histories (London, 1997), p. 150. A third version was in
the possession of Rtsis-dpan Zhwa-sgab-pa Dbang-phyug bde-ldan and utilized by him in writing about the Gurkha
War in his Bod-kyi srid-don rgyal-rabs (Kalimpong, 1976).
26 In late 1715 or early 1716, according to Petech, op. cit., p. 61.
PHO-LHA-NAS, KHANG-CHEN-NAS, AND THE LAST ERA OF MONGOL DOMINATION IN TIBET 205

some religious supports, such as the Arya Lokevara from the Potala and
others.27 At that time the sgar-dpon Khang-chen-pa deceptively invited
them into his tent for a feast and then sent the tent crashing down. Thus,
it is said, a few of the servants who were not in the tent fled while most
of the principle leaders were neutralized. Once more the precious Arya
image as well as things belonging to the government that the Junars
were carrying, most prominently jeweled objects, were handed over to
their owner. And thus the Emperor [and the Dalai Lama], priest and
patron, gave to Daiing Btur the rank of beise and appointed him to
the position of chief among the bka-blon. Later, however, because the
bka-blon from Dbus, Nga[-phod-pa], Lum[-pa-nas], and Sbya[r]-ra-ba,
were unsympathetic towards him they attacked and killed him in the
Khams-gsum sitting room in Lhasa.28

This truncated version of Khang-chen-nass career does demonstrate the significance


of the events in Mnga-ris in his rise. And those events also show that Khang-chen-nas
was able to deploy his own forces within a Mongol-dominated Tibet. Thus, Tibetan
events were not simply being dictated by the fortunes of the Qoot and the Junars;
rather, their power predominated in the era that saw the rise of Khang-chen-nas and his
patron Pho-lha-nas both Tibetan to positions of dominance. In the latters case, for
a good portion of the first half of the 18th century.
The story of Khang-chen-nass strike against the Junars in Mnga-ris does not go
wholly unchallenged. Given that the Rdo-ring paitai rnam-thar comes from the pen of

27 This is the famous sandalwood image of Avalokitevara in the Arya Chapel (Phags-pa lha-khang) in the
Potala, said to have been the tutelary deity of Srong-btsan sgam-po; see Bod rang-skyong-ljongs rig-dngos do-dam
U-yon lhan-khang, ed., Pho-brang Po-ta-lai lo-rgyus phyogs-bsgrigs (Lhasa, 1994), pp. 3637. A photograph of
the image can be found in Phuntsok Namgyal, ed., The Splendor of Tibet: The Potala Palace (Beijing 2002), p.83.
28 Bstan-dzin dpal-byor, Rdo-ring Paitai rnam-thar (Chengdu, 1986), pp. 2526: Rdai ching Sba dur du

grags pa de ni gong sa drug-pa chen po Rin chen tshangs dbyang rgya mtshoi dus nas gzhung gi rdzong gzhis
bkar yong sogs las tshan rim pai zhabs degs sgrub cing / de rjes Bod kyi rgyal po Lha bzang Khang gi dus stod
Mnga ris skor gsum gyi go byed sgar dpon las thog mdzad skabs Jun sgar Tshe ring don grub gyi dmag Bod
du lhags nas Lha bzang Khang dkrongs / Dbus Gtsang Jun sgar pas bdag bzung byed skabs kyang Mnga ris su
rang tsho zin nas Jun sgar par mgo ma gtad / de rjes gong ma bdag po chen poi lha dmag byor bai jigs zil
ma bzod par Jun sgar gyi thai ji dang / jai sang sogs drag rigs nas gzhung gi [26] rin chen rgyan chas gtsos sku
chas rtsa che ga zhig dang rtsei Phags pa Lo ki shwa ra sogs nang rten kha shas khyer te Mnga ris brgyud
phyir log byed skabs sgar dpon Khang chen pas g.yo skor gyis gur nang du mgron bod ston gshoms thog gur
rdib btang bar brten / gur nang du med pai g.yog rigs re gnyis bros thar skad las / gtso drag phal cher dmigs
med du btang mthar / slar yang Phags pa rin po chei sku dang rin chen rgyan chas gtsos gzhung gi sku chas
Jun sgar pas khyer ba rnams nor bdag poi lag tu rtsis phul bar brten / gong ma mchod yon nas / Rdai ching
sba dur la pas sei cho lo dang bka blon gyi gtso boi las khur du bsko gzhag btsal kyang / rjes sor Nga Lum
Sbya ra ba sogs Dbus pai bka blon rnams dang thugs nang ma gshin par brten Lha ldan gzim chung khams
gsum gyi nang du brkus bsad kyis dkrongs shing /. Concerning the assassination of Khang-chen-nas, see Petech,
op. cit., pp. 113121.
206 ELLIOT SPERLING

a family member Khang-chen-nas was the authors great-uncle29 it is not too surprising
that a modern history of Mnga-ris, the Mnga-ris skor-gsum sngon-byung lo-rgyus by
Gangs-ri-ba Chos-dbyings rdo-rje, gives somewhat less credit to Khang-chen-nas for the
operation and rather more to local figures:

[110] Some of the sgar-dpon led the people of Mnga-ris, protecting the
frontiers of the great motherland and contributing to the reinforcement of
its unity. For example, during the time of the sgar-dpon Khang-chen-pa or
Daiing Btur Bsod-nams rgyal-po, Junar Mongol troops reached Tibet
and Lajang Qan was assassinated. Taking sacred supports and treasures of
the Tibetan Government, they fled in the direction of Li-yul [Khotan;
i.e., modern Xinjiang]. At that time, however, Mnga-ris [leaders] held
their own and [111] met to discuss standing up to the Junar troops.
In accord with that, with the Gtsod-tsho dpon and the Brong-pa dpon
leading them, the various officials, acted as auxiliaries to the righteous
and upright sgar-dpon. With various stratagems they set out a grand feast
within a tent and when the Junars were distracted the tent was brought
down on them. Thus, most of the Junar leaders were caught unawares
and the lesser, ordinary ones, were also captured and ultimately handed
over. Once more all of the treasures of the government, first and foremost
the image of the precious Arya [Avalokitevara], along with precious
ornaments, were given back to their owner. As a result, the sgar-dpon
Khang chen-pa Bsod-nams rgyal-po was appointed bka-blon. The officials
who had been his comrades-in-arms received substantive edicts.30

The discrepancy in ascribing credit for the defeat of the Junars in Mnga-ris may
indicate a certain degree of friction between local power and the sgar-dpon appointed

29 See Bstan-dzin dpal-byor, op. cit., p. 25.


30 Gangs-ri-ba Chos-dbyings rdo-rje, Mnga ris skor gsum gyi sngon byung lo rgyus (Lhasa, 1996), pp. 110111:
Sgar dpon ga res Mnga ris mi dmangs kyi go khrid de rlabs chen mes rgyal gyi mtha tshams srung skyobs
dang gong bu gcig gyur la shugs bsnon byed pai bya bzhag nang byas rjes kyang jog gnang mdzad yod / dper
na / sgar dpon Khang chen paam / Dai Ching sba rdur Bsod nams rgyal poi skabs Sog po Jun sgar gyi dmag
Bod du byor nas Lha bzang Khng bkrongs te Bod gzhung gi sku rten gces nor mang po khyer te Li yul phyogs
su bros skabs kyang Mnga ris rang tsho bzung nas Jun sgar dmag [111] la kha gtad gcog rgyui gros gleng
mdzad don ltar Gtsod tsho dpon dang Brong pa dpon gyis gtsos dpon khag so sos lhags bsam zol med kyi sgar
dpon la ram degs byas te thabs shes sna mang gi sgo nas dbu gur nang du mgron tshul gyi ston mo gzab rgyas
bshams te mgo g.yengs bai skabs su gur rdib btang bar brten / Jun sgar bai gtso drag phal cher dmigs med
du btang mthar lhag lus dkyus ma rnams kyang dzin bzung byas te mthar sprod btang slar yang Phags pa rin
po chei sku dang rin po chei rgyan chas gtsos gzhung rgyu gces nor thams cad nor bdag gi phyag tu rtsis bul
zhus bar brten sgar dpon Khang chen pa Bsod nams rgyal po bka blon du bsko bzhag gnang ba dang / dpung
rogs dpon rigs rnams la bka gtan rgyab snon gnang ba red /. On the six primary Mnga ris dpon, see pp. 108
and 110; and also Mnga ris srid gros rig gnas lo rgyus bsdu rub u yon lhan khang, Bod ljongs Stod Mnga ris
skor gsum nye rabs chab srid kyi lo rgyus dang dgon sde khag zhig gsos grub pai gnas tshul / spyi tshogs gsar
pai phel shugs sogs rgyas par brjod pai bel gtam rin chen gter gyi phreng ba, pp. 205224.
PHO-LHA-NAS, KHANG-CHEN-NAS, AND THE LAST ERA OF MONGOL DOMINATION IN TIBET 207

from Lhasa. The Dga-bzhi clan the clan of Khang-chen-nas remained important in
the region for a brief period afterwards in the person Dga-bzhi-ba Tshe-brtan bkra-shis
who served as sgar-dpon after Khang-che-nass elevation to the bka-shag.31 Nevertheless,
there are other documentary sources that also reduce the degree of prominence traditionally
accorded Khang-chen-nas for the strategy that brought disaster to the retreating Junar
forces.32
Even if one accepts a decisive role for local leaders in these events, Khang-chen-nas
was hardly bereft of power. He had his own forces at his disposal; that much is clear.
And his domination of the situation was owed to the milieu of a mixing of Mongols and
Tibetans in the Qanate of Tibet under the Qoot Qan. Perhaps a glimpse of this comes
through in the aftermath of Khang-chen-nass assassination in 1727, less than ten years
after the events just described. Khang-chen-nas as bka-blon in Lhasa was not simply
an official obedient to the whims of bureaucracy. He retained his own forces and these
clearly appear to have been largely Mongol. The Manchu officer Mala (), who was
already in Lhasa serving as amban at the time of the assassination,33 submitted amemorial
on October 14, 1727, detailing some of what hed been told by Nga-phod, one of the
anti-Khang-chen-nas conspiritors:

Khang-chen-nas [Kangqinai ] each year paid [his] Oirat soldiers in


contravention to the numbers established by Dalai Qan []. From
the treasury he took more than 200,000 liang []. The officials who

31 Ibid, pp. 111112. Dga-bzhi-ba Tshe-brtan bkra-shis was Khang-chen-nass elder brother; see Luciano Petech,

Aristocracy and Government in Tibet. 17281959 (Rome, 1973), pp. 5152. He died fighting as an ally of Mi-dbang
Pho-lha, in 1727.
32 The Amnye Machen Institute also has in its possession documents that seem to fit the description of what

our source terms substantive edicts granted by the Tibetan government to the Mnga-ris officials (Tib. dpon),
including one, dated to February 17, 1720, given to the Stod (Mnga-ris) sgar-dpon, the Chu-tig dpon residing in
Rtsa-hrang and other officials, that takes particular note of Khang-chen-nass stand against the Junars (lhag tu
mi dbang Rdai ching bh thur nas Sbyong mgar sar dgra gdos gnang). This is elaborated as well in Gu-ge
Tshe-ring rgyal-po, Mnga-ris chos-byung gangs-ljongs mdzes-rgyan (Lhasa, 2006), p. 70. Therein mention is also
made of the role of the Gtsod-tsho dpon, already described as prominent in the attack on the Junars. Another
document held by the Amnye Machen Institute, seemingly dated to January 24, 1820 (it post-dates the era of the
8th Dalai Lama but no rab-byung cycle is specified) also alludes to the efforts of the Gtsod-tsho dpon in the fight.
Gu-ge Tshe-ring rgyal-po states that because the Gtsod-tsho dpon put forth first-rate strategems (Tib. jus) and
intelligence (Tib. rig-pa) he was subsequently known as Jus-rig Tshe-ring Btur (Tib. Sba-dur). He is no doubt
the Ju-rig-thu Bhthur mentioned in the first line of the 1720 document and this appellation is surely behind the
grant of the related title Huriqtu Taiji to the newly-appointed Mnga-ris sgar-dpon mentioned by Petech op. cit.,
1972, p. 226. I am again indebted to Tashi Tsering for providing access to the Mnga-ris documents.
The superficial lexicographical logic of Gu-ge Tshe-ring rgyal-pos etymology for Jus-rig [-thu] Btur belies its
insubstantial nature: it is little better than a folk etymology for what is at heart an appropriate Mongol title: Horitu
btur (i.e. baatur) or Brave Hero. I am grateful to Dr. Agata Bareja-Starzyska for kindly pointing this out to me.
33 On Mala, see Wu Fengpei and Zeng Guoqing , Qingdai zhuzang dachen zhuanle

(Lhasa, 1988), pp. 25.


208 ELLIOT SPERLING

came last year were close to him and didnt report it. How should one
describe this?34

Other such reports make it clear that Khang-chen-nass authority indeed rested in
good part on the significant number of loyal Mongols under his command referred to
in this memorial. (This is, of course, another sign of the persistence of elements of the
Mongol order in Tibet.) Khang-chen-nass Mongol forces are mentioned in other reports
as well. A memorial that can be dated to October 6, 1727, from Yue Zhongqi [],
Governor-General of Shaanxi and a close confidant of the emperor, describes the
information gathered from enquiries made in Dam, the region to which those Mongol
forces appear to have retreated after Khang-chennass death: According to what the
mass of Mongols were saying, Following the murder of our leader, Khang-chen-nas,
we would prefer to put ourselves under the Dalai Lama.35 Another memorial from Yue
Zhongqi submitted on April 12, 1728, indicates the ultimate disposition of the Mongols
in question: looking into the many Oirats who had been under Khang-chen-nas, today
some have come under the control of Bsod-nams dar-rgyas [the Dalai Lamas father]
and some have gravitated to Pho-lha-nas.36
The intense interaction between Tibetans and Mongols, their mixing within fighting
units during the period of Lajang Qans rule, is captured in an animated passage from
the Mi-dbang rtogs-brjod describing a lively scene during the heyday of Lajang Qans:

... At that time, the ruler of men [Lajang Qan] and his entourage, riding
excellent mounts, went to the great, happy groves of the Dam region.
And after getting there they passed the time enjoying all manner of games
and amusements. Some, atop horses possessed of the force of clouds, fired
arrows and firearms; some, as a band of horsemen, fully surrounded both
predators and animals of prey, caught them by their necks and tails, and
firing arrows, slew them. Some, groups of Mongols and groups whose
lineage was that of the imperial rulers of Tibet [Pur-rgyal Bod-rigs],
competed against each other as they played at games of infinite levels of
skill. And there they dwelled, engaging in their competitions.37

34 Zhongguo Zangxue yanjiu zhongxin , et al., eds., Yuan yilai Xizang difang yu zhongyang

zhengfu guanxi dangan shiliao huibian (Beijing, 1994), vol. 2,


p. 389: , , .
, , ?
35 Ibid., p. 386: , , .
36 Ibid., p. 428: , , .
37 Zhabs-drung Tshe-ring dbang-rgyal, op. cit., p. 221: dei tshe mii bdag pokhor dang bcas pa gzhon pa

bzang po la bcibs te Dam ljongs kyi dga bai tshal chen por gshegs so / gshegs nas kyang dus dus su rtsed jo
rnam pa sna tshogs rtse bar byed / res ga ni rta bzang po sprin gyi shugs dang ldan pai steng nas mda dang
/ mei khrul khor phang par byed / res ga ni gcan gzan dang ri dwags mang po rta pai tshogs kyis yongs su
bskor ba las gnya ba dang / mjug ma nas dzin cing / nyag phran phangs te srog kyang phrogs par byed / res
PHO-LHA-NAS, KHANG-CHEN-NAS, AND THE LAST ERA OF MONGOL DOMINATION IN TIBET 209

It is not surprising, given this atmosphere, that Khang-chen-nas, in the far west of
Tibet at the time of the Junar attack and ultimate retreat, seems to have availed himself
of the Qoot who had lost their leader with the death of Lajang Qan. Again, the Mi-dbang
rtog-brjod, this time recounting the crucial encounter of Khang-chen-nas with the Junars:

Shortly thereafter, some of the Junars with remnants of Lajang Qans


Mongols were about to move back to the Junar lands through Mnga-ris.
At that time, Khang-chen-pa, the one in charge, thought to himself: Ai!
The ruler of men, Lajang Qan, who showed great kindness to me, has
suffered grievous harm at the hands of the Junars. I can see that those
of his Mongols who are now but a remnant have been expelled to the
far reaches. If lesser beings [i.e., the Junars] can in no way accept this
absolute victory of ours, we will have a meeting of the minds with the
people of [Mnga-ris] bskor-gsum and wipe out these Junars. And we
will free these remnants of Lajang Qans forces. Thus he pledged this in
his heart and the word spread that ultimately, acting as he had pledged,
he gathered up his courage and wiped out the evil ones.38

I would posit from this, and from the later Qing report concerning the Oirats attached
to Khang-chen-nas, that his power in the ensuing period (i.e., the time from the ouster
of the Junars up until his own death), rested upon a continuing Mongol presence in
Tibet and a Tibetan and Mongol governing force within society. In essence then, the
intermingled milieu that the Mi-dbang rtogs-brjod describes and the comment of Rdo-ring
Bstan-dzin dpal-byor at the beginning of this paper are telling. This conflicts with what

ga ni Mong gol gyi skyes bui tshogs dang / Pur rgyal Bod rigs kyi skyes bui tshogs gran zlar byas nas rgyu
rtsal gyi rim pa mtha klas pa la rtse bar byed / gran par byed cing gnas so / /
38 Zhabs-drung Tshe-ring dbang-rgyal, op. cit., p. 371: de nas mi ring ba na Jun gar pai skye bo ga zhig

dang / Lha bzang Khang gi zham ring pa Mong gol dag yul Mnga ris kyi sai cha las brgyud de Jun gar gyi
yul grur gtong bar gzas pa la / dus der gtso bo Khang chen pa de nyid di snyam du bsam par gyur te / kye ma
/ mii bdag po kho bo la drin chen po dang ldan pa Lha bzang Khang de yang Jun gar pai tshogs di dag gis
nyam nyes par byas / dei zham ring du gyur pai Mong gol lhag ma di rnams kyang mthar spyugs par byed
pa la ltos / skye bo ngan pa rnams kyi blo la ni rang nyid rgyal zhing phun sum tshogs pai mtha di tsam mo
zhes chog shes nam yang ma mchis pa yin na / Bskor gsum gyi skyes boi tshogs dang blo gros mthun par byas
te / Jun gar pai skye bo di dag tshar bcad / Lha bzang Khang gi zham ring pa di dag ci bder btang bar byao
snyam pai dam bca snying la bkod nas / mthar dam bcas pa dang mtshungs pai bya ba la zhugs zhing snying
stobs bskyed de ma rungs pa rnams tshar bcad do zhes pai gtam kyang grags par gyur to / /.
The general impression given by much Tibetological literature implies that Lajang Qan, through actions that
included the removal of the 6th Dalai Lama and the execution of sde-srid Sangs-rgyas rgya-mtsho, ultimately lost
whatever significant support he had enjoyed within Tibet, a view most recently expressed by Paul Kocot Nietupski,
Labrang Monastery: A Tibetan Buddhist Community on the Inner Asian Borderlands, 17091958 (Lanham, MD,
2010), p. 8: the Zunghar Lazang Khan had alienated the Tibetan monasteries and nobles so much that he
instigated his own downfall Lazangs popular support in central Tibet collapsed. However, as the passage just
cited from the Mi-dbang rtogs-brjod shows, hostility to Lajang Qan was by no means universal. The regime of
Mi-dbang Pho-lha can even be characterized as something of a loyalist regime: loyal to the murdered Lajang Qan.
210 ELLIOT SPERLING

has often been the traditional account of the period, i.e., that the Qoot conquered Tibet
for the 5th Dalai Lama, then offered him rule over the land and contented themselves
with life in theDam region until Lajang Qan became eager for power and took it from
the Dalai Lamas government. Although a certain portion of this is true Lajang Qan
did seek to expand the scope of his power the idea that the Mongols were largely
in the background from the mid-17th century or that Tibetans and Mongols inhabited
largely separate spaces is a simplification that does not benefit our understanding of the
situation. The place of the Mongols in Tibet was not as inconsequential or as invisible
as some modern accounts, out of silence, might imply. The fact that Khang-chen-nass
great-nephew, Rdo-ring Bstan-dzin dpal-byor, was capable of conversing in Mongol is,
as Ive noted, indicative of the enduring influence and effect of Tibets second Mongol
Century, which we may broadly place from the mid-1600s to the mid-1700s. Just as
significantly, Rdo-ring Bstan-dzin dpal-byor, under questioning by Qing officials in
Beijing noted, without any hint of displeasure at the fact, that his personal adornments
in part reflected Tibets history as a kingdom of Guri Qan and the Koko Nor Mongols.

***

As a final note, we might point out that this also says something about Tibets place
within the process of Mongol assimilation into the Qing imperium and the concomitant
transformation or perhaps subversion of the Mongol world view. The process did
entail the fostering of a Lhasa-centered faith among the Mongols, as detailed by Johan
Elverskog in his important analysis of the nexus between traditional Mongol beliefs,
Manchu state building and Buddhism. But what was at work in this was not just a simple
imperial desire to tie the Mongols to a Tibetan Buddhism that was removed in language,
sacred sites, etc., from what a localized Mongol Buddhism would have wrought.39 As
it is, there were already some very strong ties between Lhasa and the Mongols of the
Koko Nor at least, many of whom were quite conversant with the ways of Lhasa and
the faith manifested there. Indeed, the sandalwood image of Phags-pa Lokevara, the
recovery of which was so significant in the rise of Khang-chen-nas, was particularly
valued by the Koko Nor rulers: it was ultimately the Mongol noblewoman Dalai Kun-ci
rgyal-mo, who arranged for the image, held to have been the tutelary deity of Srong-btsan
sgam-po, to be returned to Lhasa and the Potala in 1645 after a long period of custody
outside Central Tibet (largely a result of the hostilities that preceded the final triumph
of the Dge-lugs-pa and the Qoot), including a time in which it was kept by the Tmed

39 Johan Elverskog, Our Great Qing: the Mongols, Buddhism and the State in late Late Imperial China (Honolulu,

2006), pp. 117118: [T]he courts political support of the Lhasa-based Gelukpa justified the Qing ideologically
since it was this development the institutionalization of Gelukpa orthodoxy that mandated the Mongols to
become Lhasa-oriented Qing Buddhists. As a result, the incorporation of the Mongols into the Buddhist Qing
entailed the thorough transformation of their culture to a Tibetan-style Gelukpa one and the suppression of locally
produced Buddhisms, which potentially created competing centers of authority.
PHO-LHA-NAS, KHANG-CHEN-NAS, AND THE LAST ERA OF MONGOL DOMINATION IN TIBET 211

in the Koko Nor region.40 It is worth noting that the image like the adornments worn
by Rdo-ring Bstan-dzin dpal-byor can be perceived as a link between the imperial
Tibetan dynasty and the lineage of Guri Qan.
All in all, given the Mongol presence in Lhasa to the point of Mongol remaining
as a language of certain members of the Tibetan elite (and quite possibly others), one
ought to temper the admittedly valid notion of the Qing production of Lhasa-oriented
Qing Buddhists among the Mongols with a recognition of the often close relations
military, religious and otherwise that existed between Tibetans and Mongols in the
heart of Central Tibet during the course of Tibets Mongol Century.
And with this I return to the idea voiced at the outset. I am not stating that Tibet was
simply Mongol. Far from it. But I am saying there is a Mongol aspect to Tibets history;
an angle from which it would be worth reframing some of our perceptions. Again: if
valid histories of India between the mid-18th and the mid-19th centuries can be written
from the standpoint of Fort William and also from the standpoint of Agra, so too we
might consider the similar validity of histories of Tibet between the mid-17th and the
mid-18th centuries constructed from the standpoint of the Potala and histories covering
the same period constructed from the standpoint of Dam.

40 Ngag-dbang blo-bzang rgya-mtsho, Ngag-dbang blo-bzang rgya-mtshoi rnam-thar [= Du-k-lai gos-bzang]

(Lhasa, 1989), pp. 256257. The passage cited is translated (with some obvious errors) by Ahmad, op. cit., p. 142.
See also Bod rang-skyong-ljongs rig-dngos do-dam U-yon lhan-khang, ed., op. cit., pp. 3637, which erroneously
places the return of the image in the year 1642 and asserts that it was in Khalkha custody. This work also writes
the name of the Mongol noblewoman who saw to the return of the image as Dalai Gun-ci rgyal-mo and identifies
her as the chief wife of Guri Qan. The rulers chief wife is identified by Sum-pa mkhan-po, op. cit., p. 995, only
as Thor-gwod (i.e., Torud) dpon-mo, an identification repeated by Dung-dkar, op. cit., p. 507.
R O C Z N I K O R I E N T A L I S T Y C Z N Y, T. LXV, Z. 1, 2012, (s. 212223)

PIOTR TARACHA

Hittitology Up to Date: Issues and New Approaches

Abstract

This paper is a stock-taking of the present state of research in Hittitology and ancient
Anatolian studies in general, including the ongoing publication and digitalization projects
concerning Hittite texts and iconographic sources as well as spectacular archaeological
discoveries made in the past few decades. New study perspectives and still debatable
issues are also highlighted, with reference, among others, to Hittite history, geography,
written legacy and text dating, and a new approach to descriptions of cult festivals and
magical rituals known from the archives of the Hittite capital Hattusa.

Keywords: Hittitology, Asia Minor, Hittite history, Hattusa, Hittite archaeology

After more than a century of research on the languages, history and culture of Asia
Minor in the second millennium BC, Hittitology as this field is called, has come of age.
Hugo Winckler, who started the first regular excavations in Hattusa (now Boazkale about
150 km as the crow flies east of Ankara) in 1906, discovered over the course of four
digging seasons more than 10,000 complete and fragmentary cuneiform tablets.1 After a
century, the German excavations in the old capital of the Hittite empire are still ongoing.2

1 For the first excavations in Boazky-Hattusa, see now S. Alaura, Nach Boghaski! Zur Vorgeschichte
der Asugrabungen in Boazky-Hattua und zu den archologischen Forschungen bis zum Ersten Weltkrieg
(13. Sendschrift der Deutschen Orient-Gesellschaft), Berlin 2006; J. Seeher, Die Adresse ist: poste restante
Yozgat Asie Mineure: Momentaufnahmen der Grabungskampagne 1907 in Boazky. In: J. Klinger, E. Rieken,
Ch. Rster (eds), Investigationes Anatolicae: Gedenkschrift fr Erich Neu (Studien zu den Boazky-Texten 52),
Wiesbaden 2010, pp. 253270.
2 Preliminary reports are regularly published in the Archologischer Anzeiger. For archaeological results of

the excavations, see, e.g., W. Orthmann, Die Ausgrabungen der letzten 50 Jahre in Hattua und die Geschichte
HITTITOLOGY UP TO DATE: ISSUES AND NEW APPROACHES 213

The cuneiform texts unearthed to date during the 67 field seasons their number is
fast approaching 30,000 as well as quite numerous Luwian hieroglyphic inscriptions
from the second millennium BC3 have been instrumental in resurrecting the forgotten
culture and history of the Hittite state which existed in central Asia Minor from about
the middle of the 17th century BC. Under a new dynasty (from the 1460s BC) the Hittite
empire eventually expanded to cover vast territories from the Aegean coastal area of
Asia Minor to northern Syria, becoming the third, after Egypt and Kassite Babylonia,
superpower in the ancient Near East. The peak in the might of this state came in the reign
of Shuppiluliuma I (c. 13551322), who subjugated the land of Mittani and conquered
Syria. The fall, in still unexplained circumstances, came in the 1180s BC. With it came
the end of the archives of cuneiform documents in Hattusa.
Equally useful in resurrecting the Hittites are the spectacular archaeological discoveries
made in the past few decades in Hattusa itself, as well as in other important Hittite urban
centers like Tapigga (Maat Hyk), Sapinuwa (Ortaky), Sarissa (Kuakl), and in recent
years Oymaaa near Vezirkpr in the lower reach of the Kzlrmak, near the Black
Sea littoral, and Kayalpnar in the rivers upper reach, tentatively identified by German
archaeologists as the Hittite towns of, respectively, Nerik and Samuha.4 The excavations
in Hattusa especially have brought groundbreaking results, providing a new perspective
on the evolution of the city itself, and the everyday life and history of the Hittites and
their state in general. The breakthrough nature of the discoveries currently being made
in Hattusa is best illustrated by the title of a paper by Jrgen Seeher, the head of these
excavations from 1994 to 2005: Farewell to Certainties.5

der hethitischen Kunst. In: G. Wilhelm (ed.), Hattua-Boazky: Das Hethiterreich im Spannungsfeld des Alten
Orients: 6. Internationales Colloquium der Deutschen Orient-Gesellschaft, 22.-24. Mrz 2006, Wrzburg (Colloquien
der Deutschen Orient-Gesellschaft, Bd. 6), Wiesbaden 2008, pp. 1533.
3 Th.P.J. van den Hout, The Written Legacy of the Hittites. In: H. Genz, D.P. Mielke (eds), Insights into

Hittite History and Archaeology (Colloquia Antiqua 2), Leuven Paris Walpole, MA 2011, pp. 4784, offers
an overview of all preserved texts from the period of the Hittite kingdom, recorded in cuneiform on clay tablets,
one bronze tablet and some objects as well as in the Luwian hieroglyphic script on seals and seal impressions,
stone surfaces, and metal objects. Cf. also idem, Institutions, vernaculars, publics: the case of second-millennium
Anatolia. In: S.L. Sanders (ed.), Margins of Writing, Origins of Cultures (Oriental Institute Seminars 2), Chicago
2007 (Second Printing with Postscripts and Minor Corrections), pp. 221262.
4 Generally, H. Genz, D.P. Mielke, Research on the Hittites: A Short Overview. In: H. Genz, D.P. Mielke

(eds), op. cit., note 3, pp. 130. Cf. also M. Sel, Bir Hitit Bakenti Ortaky-apinuva, Ankara 2008; A. Sel,
Another Capital City of Hittite State: apinuwa. In: F. Pecchioli Daddi, G. Torri, C. Corti (eds), Central-North
Anatolia in the Hittite Period. New Perspectives in Light of Recent Research: Acts of the International Conference
Held at the University of Florence (79 February 2007) (Studia Asiana 5), Roma 2009, pp. 193205; R.M.Czichon,
Archologische Forschungen am Oymaaa Hyk in den Jahren 2005 and 2006, ibidem, pp. 2530; A. Mller-
Karpe, Recent Research on Hittite Archaeology in the Upper Land, ibidem, pp. 109117; idem, aria. B.
Archologisch. In: M.P. Streck et al. (eds), Reallexikon der Assyriologie und Vorderasiatischen Archologie,
vol. 12/1-2, BerlinNew York 2009, pp. 6264; A. Mller-Karpe, V. Mller-Karpe et al., Untersuchungen in
Kayalpnar und Umgebung 20062009, Mitteilungen der Deutschen Orient-Gesellschaft 141, 2009, pp. 173238.
5 J. Seeher, Abschied von Gewusstem: Die Ausgrabungen in Hattua am Beginn des 21. Jahrhunderts.

In: G.Wilhelm (ed.), op. cit., note 2, pp. 113.


214 PIOTR TARACHA

We shall see that this interpretational uncertainty regarding new findings in


archaeology concerns Hittitological debate overall in recent years. The philology
archaeology interface is becoming increasingly important for future research.6 The pace
of studies should increase as a result and fields hitherto considered inaccessible from
the point of view of these disciplines separately should gain ground. In Hittitology
itself interesting new research perspectives have appeared. Suffice it to quote one of the
coryphaei of Hittitology today, Theo van den Hout from the Oriental Institute in Chicago:
Die Ausgrabungen der Deutschen Orientgesellschaft in Hattua haben uns eine Flle an
Material beschert, die unerschpflich ist, und dies nicht nur, weil die Grabungen noch
immer andauern, sondern vor allem weil jede Zeit ihre eigenen Fragen an die Vergangenheit
stellt. Das erste Jahrhundert der Hethitologie hat sich vor allem um die Grundlagen der
Forschung bemht, es ist aber schon in den letzten Jahrzehnten eine ganz klare Tendenz
zur Synthese und zu weiterreichenden Problemstellungen sichtbar.7
Let me first emphasize that the breakthrough nature of current research in this field
of studies is something that Hittitologists and archaeologists digging on Hittite sites are
both well aware of. Current approaches and research projects, either just undertaken or
planned for the near future, increasingly often charged to bigger teams,8 will set new
objectives and highlight new study perspectives for the next few decades. The first hundred
years of Hittitology were summed up nicely at the 6th international colloquium of the
Deutsche Orientgesellschaft: Hattua-Boazky: Das Hethiterreich im Spannungsfeld des
Alten Orients, organized at the University of Wrzburg on the centennial of the start
of regular excavations at Hattusa.9 Two other international conferences and workshops
summing up research and setting new study perspectives also deserve note: Structuring
and Dating in Hittite Archaeology: Requirements Problems New Approaches, at
the German Archaeological Institute in Istanbul in November 2004,10 and Central-North

6 A recently published volume, H. Genz, D.P. Mielke (eds), op. cit., note 3, provides contributions on several key

issues in Hittite studies based on new developments and approaches from historical, philological and archaeological
points of view. Cf. also U.-D. Schoop, Wo steht die Archologie in der Erforschung der hethitischen Kultur?
Schritte zu einem Paradigmenwechsel. In: G. Wilhelm (ed.), op. cit., note 2, pp. 3560, who states on p. 54:
Wie man sieht, stehen unsere beiden historischen Informationsquellen, die Texte und der materielle Befund, nicht
in Konflikt, sie vervollstndigen einander.
7 Th.P.J. van den Hout, Verwaltung der Vergangenheit: Record Management im Reiche der Hethiter. In:

G.Wilhelm (ed.), op. cit., note 2, pp. 9394.


8 On a new research project combining expertise of archaeologists and Hittitologists, see M.F. Fales, S. de

Martino, S. Ponchia, K. Strobel, Austro-Italian Archaeological Investigations in the Region of Yozgat (Turkey).
In: F. Pecchioli Daddi, G. Torri, C. Corti (eds), op. cit., note 4, pp. 3137. Cf. also S. de Martino, F.M. Fales,
S. Ponchia, Archaeological Investigations at Yass-Hyk (Yozgat): The Site of Yass-Hyk within the Overall
Picture of Hittite geography. In: A. Sel (ed.), VII. Uluslararas Hititoloji Kongresi Bildirileri: orum 2529 Austos
2008 = Acts of the VIIth International Congress of Hittitology: orum, August 2529, 2008, vol. 1, Ankara 2010,
pp. 189198.
9 G. Wilhelm (ed.), op. cit., note 2.
10 D.P. Mielke, U.-D. Schoop, J. Seeher (eds), Strukturierung und Datierung in der hethitischen Archologie:

Voraussetzungen Probleme neue Anstze / Structuring and Dating in Hittite Archaeology: Requirements
HITTITOLOGY UP TO DATE: ISSUES AND NEW APPROACHES 215

Anatolia in the Hittite Period: New Perspectives in Light of Recent Research, which
took place in February 2007 at Florence.11
Many of the issues this article deals with have been taken up in papers presented at
these conferences and the discussions that followed. If so much has already been said on
the subject, why then this brief analysis of the position of Hittitology today and tasks and
challenges facing the discipline in the coming years? I have felt an obligation to do so,
not only because I appreciate the importance of such reviews of the state of research in
my field,12 especially to an esteemed body of representatives of different Oriental studies.
The other reason is that the Ancient Near Eastern Studies Department of the Oriental
Studies Faculty of the University of Warsaw was holding the eighth international congress
of Hittitology on 5-9 September 2011. The congresses, which take place every three years
starting from 1990, alternately in orum, capital of the Turkish province where Hattusa is
located, and in one of European centers of Hittite studies (Pavia, Wrzburg, Rome, and
Warsaw to date), have already become the most important international event in world
Hittitology. Entrusting us with the organization of the congress was an expression of
international esteem for the Warsaw school of Hittitology established by Maciej Popko.13
As usual, the congress provided a forum for Hittitologists and archaeologists to discuss the
future of studies on Hittite Anatolia. This paper, which brings some general reflections,
is an outcome of this important event.
I have cited here van den Houts conclusion that the first hundred years in Hittitology
were dedicated to creating a base for future research. In this context, the Hethitische
Forschungen (Hittite Studies) project is of greatest importance. For fifty years it has
been run under the auspices of the Deutsche Forschungsgemeinschaft by an institution
established in 1961, the Kommission fr den Alten Orient of the Academy of Sciences and
Literature in Mainz, and directed successively by Heinrich Otten (until 1992), Erich Neu
(died in 1999) and Gernot Wilhelm (since 2000).14 The first objective was to collect and
prepare lexicographically all existing Hittite texts. Ottens lexical files collection, which is
indispensable for any text edition, has grown from 90,000 cards in 196115 to more than
amillion today. As of May 2010, according to Silvin Koak, it will no longer be expanded
in the paper version, but rather in digitalized form.16 The lexical collection held physically
in the western wing of the building of the Academy in Mainz, on Geschwister-Scholl-

Problems New Approaches: Internationaler Workshop, Istanbul, 26.27. November 2004 (BYZAS 4), Istanbul
2006.
11 F. Pecchioli Daddi, G. Torri, C. Corti (eds), op. cit., note 4.
12 See, first of all, E. Neu, Hethitologie heute. In: G. Wilhelm (ed.), Akten des IV. Internationalen Kongresses

fr Hethitologie: Wrzburg, 4.8. Oktober 1999 (Studien zu den Boazky-Texten 45), Wiesbaden 2001, pp. 111.
13 P. Taracha, Hittitology in Warsaw: Past, Present and Future, Rocznik Orientalistyczny 62/1, 2009,

pp. 213221.
14 G. Wilhelm, Die Edition der Keilschrifttafeln aus Boazky und das Projekt Hethitische Forschungen der

Akademie der Wissenschaften und der Literatur, Mainz. In: G. Wilhelm (ed.), op. cit., note 2, pp. 7386.
15 H. Otten, Kommission fr den Alten Orient. Bericht. In: Jahrbuch 1961 der Akademie der Wissenschaften

und der Literatur, Mainz 1962, p. 143.


16 Dr. S. Koak pers. comm.
216 PIOTR TARACHA

Strasse, is available to all Hittitologists, including younger colleagues. It is noteworthy


that the projects coordinated by the Mainz Academy engage researchers from around
the world. In this sense, Hittitology is a model of international cooperation in science.
Gernot Wilhelm informs that the publication of Hittite texts from the newer excavations
in Hattusa (in 19311939 and from 1952) should be completed by 2015 at the latest.
The texts will be published in the Keilschrifttexte aus Boazky series (published in
19161923 and from 1954), which by the same will close on volume 69.17 The parallel
series Keilschrifturkunden aus Boazky, published since 1921 (after World War II by
the German Academy of Sciences, from 1972 the Academy of Sciences of the German
Democratic Republic), was interrupted on volume 60 (1990) after the GDR returned
to Turkey in November 1987 the tablets with Bo numbers originating from Wincklers
excavations in 19061912, kept from 19161917 in the Berlin Museum. They are now
held by the Museum of Anatolian Civilizations in Ankara, but Turkish colleagues have
yet to undertake the task of editing these texts. Similarly, a few thousand Hittite tablets
uncovered by an expedition from the University of Ankara in Ortaky (Hittite Sapinuwa),
directed by Aygl Sel, await edition.
Two projects of the Mainz Academy have become indispensable tools in Hittitology
today. The first is a lexicon of Hittite signs (Hethitisches Zeichenlexikon), published by
Erich Neu and Christel Rster in 1989 after almost ten years of work.18 The second,
commenced in 1985 and nowadays still being conducted, is of much greater importance.
The Concordance of Hittite Cuneiform Texts (Konkordanz der hethitischen Keilschrifttexte)
is the lifework of Silvin Koak, who managed to publish four volumes between 1992 and
1999 in the Studien zu den Boazky-Texten series,19 and then another five volumes
reflecting the state of research in 2005.20 In the case of the Concordance, however, the
Internet has demonstrated great potential. Putting in order thousands of text fragments
and finding ever new joins (with the inestimable help of Hittitologists around the world)
required constant changes, making hardcopy versions of the concordance obsolete by the

17 Prof. G. Wilhelm pers. comm.


18 Ch. Rster, E. Neu, Hethitisches Zeichenlexikon: Inventar und Interpretation der Keilschriftzeichen aus den
Boazky-Texten (Studien zu den Boazky-Texten Beiheft 2), Wiesbaden 1989.
19 S. Koak, Konkordanz der Keilschrifttafeln I. Texte der Grabung 1931 (Studien zu den Boazky-Texten 34),

Wiesbaden 1992; idem, Konkordanz der Keilschrifttafeln II. Texte der Grabung 1932 (Studien zu den Boazky-
Texten 39), Wiesbaden 1995; idem, Konkordanz der Keilschrifttafeln III/1. Texte der Grabung 1933: 1/c-1300/c
(Studien zu den Boazky-Texten 42), Wiesbaden 1998; idem, Konkordanz der Keilschrifttafeln III/2. Texte der
Grabung 1933: 1301/c-2809/c (Studien zu den Boazky-Texten 43), Wiesbaden 1999.
20 S. Koak, Konkordanz der hethitischen Keilschrifttafeln 1. Teil: Die Textfunde der Grabungen in Boazky

19061912 (Hethitologie Portal Mainz Materialien 1), Wiesbaden 2005; idem, Konkordanz der hethitischen
Keilschrifttafeln 2. Teil: Die Textfunde der Grabungen in Boazky 19311939 (Hethitologie Portal Mainz
Materialien 2), Wiesbaden 2005; idem, Konkordanz der hethitischen Keilschrifttafeln 3. Teil: Die Textfunde der
Grabungen in Boazky 19521963 (Hethitologie Portal Mainz Materialien 3), Wiesbaden 2005; idem, Konkordanz
der hethitischen Keilschrifttafeln 4. Teil: Die Textfunde der Grabungen in Boazky 19642004 und Texte aderer
Provenienz (Hethitologie Portal Mainz Materialien 4), Wiesbaden 2005; idem, G.G.W. Mller, Konkordanz der
hethitischen Keilschrifttafeln 5. Teil: Die Textfunde der Grabungen in Boazky: Indizes der Konkordanz (Hethitologie
Portal Mainz Materialien 5), Wiesbaden 2005.
HITTITOLOGY UP TO DATE: ISSUES AND NEW APPROACHES 217

time they had been printed and slowing work on new volumes. The online version of
the Concordance, updated more or less every six months, has been generally available
on the Hethitologie Portal Mainz (www.hethiter.net) since it was established in 2001.
Technical supervision and coordination is handled by Gerfrid G.W. Mller. The portal
also offers access to images of Hittite texts, the target being about 50,000 images (there
are plans for 3D rotational views of tablets in the future21), and a Hittite bibliography.
The Mainz Portal provides links or direct access to digitalization projects concerning
Hittite texts and iconographic sources, conducted in various academic centers, for example,
Hititte state treaties (Gernot Wilhelm, University of Wrzburg), magical rituals (Doris
Prechel, University of Mainz), Hittite myths (Elisabeth Rieken, University of Marburg),
and the Anatolian Hieroglyphic glyptics project (Massimiliano Marazzi, University of
Naples, Clelia Mora, University of Pavia, Gerfrid Mller, University of Wrzburg).22
A project for the edition and possibly also digitalization of historical texts is being
carried out by Italian colleagues, including Onofrio Carruba (Pavia), Franca Pecchioli
(Florence) and Stefano de Martino (Turin). The goal of all projects of the digitalization
type has been well phrased by Gernot Wilhelm: Das Hauptziel all dieser Bemhungen
ist es, den Zugriff auf hethitologische Daten so zu organisieren, dass nich jeder einzelne
Hethitologe den grten Teil der knappsten Resource, ber die er verfgt nmlich seine
Zeit dafr einsetzen muss, Daten aller Art zu sammeln, sondern dass er gezielt und
unter mglichst geringem Aufwand die Daten erheben kann, die fr die Beantwortung
der Fragen, die er bearbeiten mchte, ntig sind. (...) Heute erhlt er diese Informationen
in Minutenschnelle und kann seine kostbare Zeit dafr nutzen, die bereitgestellten Daten
kritisch zu prfen, wo ntig, zu korrigieren und sich auf die kognitiven Aspekte seiner
Arbeit zu konzentrieren.23 I am sure the crucial argument about saving time is one we
all agree with.
The Concordance has given Hittitologists an excellent research platform and so have
two context dictionaries of the Hittite language that have shown great progress recently.
They are based on lexical collections of two late eminent scholars in our field, respectively,
Hans Gustav Gterbock and Annelies Kammenhuber: The Hittite Dictionary of the Oriental
Institute of the University of Chicago (volumes from M to have been published since
1980), and the second edition of Hethitisches Wrterbuch (Johannes Friedrich was the
author of the first edition), prepared in the University of Munich (fascicles from A to H
have appeared since 1975).24 Moreover, the long awaited new descriptive grammar of

21 Such new technical means will bring about fast progress in restoring larger parts of original tablets from
dispersed text fragments. See G.G.W. Mller, Tausend Teile: Technische Mittel zur Rekonstruktion der Festrituale.
In: G.G.W. Mller (ed.), Liturgie oder Literatur? Die Kultrituale der Hethiter (in print).
22 M. Marazzi, C. Mora, G. Mller, The Anatolian Hieroglyphic Glyptics Project and its Digitalisation into

the Mainz Portal. In: F. Pecchioli Daddi, G. Torri, C. Corti (eds), op. cit., note 4, pp. 259268.
23 G. Wilhelm, op. cit., note 14, p. 84.
24 J. Friedrich, A. Kammenhuber et al., Hethitisches Wrterbuch. Zweite, vllig neubearbeitete Auflage auf der

Grundlage der edierten hethitischen Texte, Heidelberg 1975 ff.


218 PIOTR TARACHA

the Hittite language by Harry Hoffner, Jr. and Craig Melchert was published in 2008,25
replacing Johannes Friedrichs classic but already obsolete grammar of 1960.26
The last decade has also seen a bumper crop of publications, from brief contributions
to comprehensive monographs, concerning different population groups and languages
in Hittite Anatolia, Hittite history and society, the Hittite states international relations,
cross-cultural connections with the Achaeans and other neighbors, literature, religion,
magic, divination, geography, and other aspects of Hittite culture. Theo van den Houts
observation concerning the trend toward synthesis in recent years finds full confirmation
in these publications. Hittitologists have also made a concerted effort to translate Hittite
texts into modern languages, including German and English within the framework of the
most important projects.27
Even so, many issues remain unclear and debatable, despite the decades of research
that have passed. An objective impediment in many instances is the still insufficient body
of sources and difficulties in their interpretation. To mention Hittite history as one example,
a key and still uncertain issue concerns the circumstances accompanying the founding
of the Hittite state about the middle of the 17th century BC, as well as the events from
the later 16th to the middle of the 15th century BC preceding the emergence of the New
Kingdom that should be connected in my opinion (but which is not shared by many
colleagues) with the takeover of the throne in Hattusa by a new dynasty of Kizzuwatnean
origin. Finally, there is the fall of the state in shady circumstances in the 1180s.
New approaches are possible and recommended with regard to at least the latter
two questions. The results of Hittite archaeology must be taken into account. In 2009,
Andreas Schachner published an article under the meaningful title: The 16th century
BC a time of change in Hittite central Anatolia.28 He pointed out changes not only
in the city of Hattusa, which was extended to include the fortified Upper City with at
least three large temples 2, 3 and 4 located in it, but also outside the capital, where
new towns with monumental architecture were established, for example, Sarissa/Kuakl
excavated by German archaeologists.29 Unification processes in the state are attested
to, for instance, by the unifying of ceramic forms throughout the region situated in the
bend of the Kzlrmak River. At the same time, imported pottery vessels appeared in

25 H.A. Hoffner, H.C. Melchert, A Grammar of the Hittite Language, Part 1: Reference Grammar; Part 2:
Tutorial, Winona Lake, Indiana 2008.
26 J. Friedrich, Hethitisches Elementarbuch, 1. Teil: Kurzgefate Grammatik, 2nd ed., Heidelberg 1960.
27 See the relevant chapters in The Context of Scripture, Vols. 13, Leiden 19972002, and Texte aus der

Umwelt des Alten Testaments. Neue Folge, Gtersloh 2005 ff., as well as the successive volumes of the Writings
from the Ancient World series, including H.A. Hoffner, Jr., Hittite Myths. Second Edition (WAW 2), Atlanta, GA
1998; G. Beckman, Hittite Diplomatic Texts. Second Edition (WAW 7), Atlanta, GA 1999; I. Singer, Hittite Prayers
(WAW 11), Atlanta, GA 2002; H.A. Hoffner, Jr., Letters from the Hittite Kingdom (WAW 15), Atlanta, GA 2009;
G. Beckman, T. Bryce, E. Cline, The Ahhiyawa Texts (WAW 28), Atlanta, GA 2011.
28 A. Schachner, Das 16. Jahrhundert v. Chr. eine Zeitenwende im hethitischen Zentralanatolien, Istanbuler

Mitteilungen 59, 2009, pp. 934.


29 Preliminary reports on these excavations have been regularly published by A. Mller-Karpe in the Mitteilungen

der Deutschen Orient-Gesellschaft since 1995. See also note 4.


HITTITOLOGY UP TO DATE: ISSUES AND NEW APPROACHES 219

Anatolia, like spindle bottles and libation arms, both representing the so-called Red
Lustrous Wheelmade Ware. Ulf-Dietrich Schoop comments: Whrend mit den Spindle
Bottles und ihrem Inhalt neue Objekte im Bereich der Selbstdarstellung erscheinen, lt
das Auftauchen der Libationsarme ebenfalls importierte nderungen ritueller Natur
erkennen. (...) Die oben besprochene Befundsituation deutet auf einen schnellen Wandel
gegen 1400 v. Chr. hin; diesem ging eine lang ausgezogene Vorlaufphase voraus, die
das ganze 15. Jahrhundert umfate.30
This breakthrough moment has also been observed by Hittitologists. For example,
in his considerations on the rise of Hittite literacy Theo van den Hout states: Only by
the time of the earliest charters does the typical Hittite cuneiform seem to become more
firmly established. Following Wilhelm these can be dated to the later 16th century, to the
reigns of Ammuna and Huzziya at the earliest. (...) [The typical Hittite cuneiform] would
thus have gained its final foothold somewhere in the course of the later 16th century. Not
coincidentally, the charters are also the first direct expression of administrative activity.
Administration and literacy are very much interdependent (...). With his administrative
reforms and renewal of international diplomacy Telipinu is an excellent candidate for a
king who may have given the decisive push in getting the cuneiform script definitively
established in the Hittite kingdom. Besides the charters his so-called Proclamation describes
a nation-wide administrative reform and his also seems to be the first of a series of
treaties with Kizzuwatna in south-eastern Anatolia.31
The said Proclamation, regulating among others issues of royal succession from father
to the eldest son, put the Hittite state in line with the other kingdoms of the ancient Near
East, also in the administrative and ideological sense. One should also emphasize the
opening onto Kizzuwatna, which should be seen perhaps as a key (impeded, however,
by the scarcity of sources) to interpreting the circumstances accompanying the takeover
of the throne in Hattusa a few dozen years after Telipinu (in the 1460s?) by a new
Hurrianized dynasty of Kizzuwatnean origin. This new royal family may well have been
established at the court in Hattusa long before the coup dtat of Muwattalli I. I am in
full agreement with van den Hout when he says that with the accession of Tudhaliya I,
Kantuzilis son, a new era begins when the Hittite state anew established itself as one of
the major players of the global politics of the second millennium B.C.32 Observations
on the consequences of this change for the image of the state and for the cultural picture
of Hittite Anatolia in general, presented in works concerning specific aspects of the

30 U.-D. Schoop, op. cit., note 6, p. 54.


31 Th. van den Hout, Reflections on the Origins and Development of the Hittite Tablet Collections in Hattusa
and Their Consequences for the Rise of Hittite Literacy. In: F. Pecchioli Daddi, G. Torri, C. Corti (eds), op. cit.,
note 4, pp. 9293.
32 Th. van den Hout, A Century of Hittite Text Dating and the Origins of the Hittite Cuneiform Script, Incontri

Linguistici 32, 2009, p. 35. However, he identifies this Tudhaliya with his namesake, the spouse of queen Nikalmadi
who ruled Hatti one or two(?) generations later. For a summary of the discussion on the political history of the
Early Hittite Empire, reflecting different points of view, see now S. de Martino, Some Questions on the Political
History and Chronology of the Early Hittite Empire, Altorientalische Forschungen 37, 2010, pp. 186197.
220 PIOTR TARACHA

culture, deserve a holistic analysis, including numerous, weighty arguments for the new
dynasty having seized power in Hattusa around the middle of the 15th century BC. This
is definitely a major proposition for future studies on Hittite history and culture.
The results of archaeological excavations have also prompted a rethinking of the
meaning of texts referring to the fall of the Hittite state around 1180 BC. We now
know that Hattusa did not succumb to a single concerted attack by an outside enemy,
but was depopulated and impoverished gradually in the course of the latter half of the
13th century BC (it was at this time, among others, that 16 out of the 25 temples in the
main sacred precinct in the Upper City fell into ruin).33 Can it be that the defeat in the
conflict with Assyria in the reign of Tudhaliya IV (c. 12401212) weakened the state
to such a degree that even the short revival under his son Shuppiluliuma II could not
lift it from ultimate decline? Is this the reason why a number of monumental building
projects, undoubtedly fueled by ideological motives, such as the stone fortifications of the
Upper City in Hattusa and the cult reliefs on the city walls of Alacahyk, were never
completed?34 Hittitologists should also reconsider in this context the character of the Hittite
tablet collections in the capital archives. Was the state administration reflected in these
archives operating until the fall around 1180 BC according to the scenario tentatively
sketched by Theo van den Hout in his recent works,35 or was it reduced and its tasks
redefined during this last period (maybe, with a new role of the House on the Slope)?
Some of the debate is burdened with dubious methodology and false assumptions. Let
issues of Hittite geography serve as a classical example in this case. Comparing Hittite
toponyms with similarly sounding place names from ancient and Byzantine periods is
misleading as a rule. In effect, the locations of some of the Hittite cities proposed in the
course of a debate lasting several dozen years have been numerous and often quite far
apart. Even more loaded with far-reaching consequences is the identification, accepted
by many Hittitologists, of A(m)kuwa (now Aliar Hyk) of the Assyrian Colony period
with the city of Ankuwa mentioned in Hittite texts, in spite of the fact that the sources
are fairly clear in locating this initially Hattian center called Hanniku (it is only the
Hittitized name Ankuwa that starts to resemble Amkuwa) north of Hattusa and in the
neighborhood of the city of Hanhana.36 The identification of Hittite Ankuwa with Aliar
Hyk in recent works by many different authors results in a shift of the entire geography
of the land of Hatti to the south of the capital.37 A new methodological approach would

33 J. Seeher, Die Zerstrung der Stadt Hattua. In: G. Wilhelm (ed.), op. cit., note 12, pp. 623634; cf. also

idem, op. cit., note 5, p. 9.


34 Cf. P. Taracha, The Iconographic Program of the Sculptures of Alacahyk, Journal of Ancient Near

Eastern Religions 11, 2011, pp. 143144.


35 Th.P.J. van den Hout, op. cit., note 7, pp. 8794.
36 Cf. P. Taracha, Religions of Second Millennium Anatolia (Dresdner Beitrge zur Hethitologie 27), Wiesbaden

2009, p. 101 n. 528; idem, Local Cults in the Zuliya Basin. In: Y. Cohen, A. Gilan, J.L. Miller (eds), Pax
Hethitica: Studies on the Hittites and their Neighbours in Honour of Itamar Singer (Studien zu den Boazky-
Texten 51), Wiesbaden 2010, p. 351.
37 See M. Forlanini, The Central Provinces of Hatti: An Updating. In: K. Strobel (ed.), New Perspectives

on the Historical Geography and Topography of Anatolia in the II and I Millennium B.C. (Eothen 16), Firenze
HITTITOLOGY UP TO DATE: ISSUES AND NEW APPROACHES 221

be greatly recommended in future studies on Hittite geography and this is exactly what
may be forthcoming from a doctoral dissertation currently being prepared under my
supervision by Adam Krysze, who is applying cluster methodology to the analysis of
the relevant textual evidence.
A key issue in Hittitology is text dating. An overview of developments in this aspect
of Hittite studies has been presented by Theo van den Hout in his article published in
2009.38 The debate commenced after the discovery in 1952, in stratified archaeological
context, of a fragment 29/k, soon published as KBo 7.14, which became known as the
Zukrai Text. By the end of the 20th century, there was general agreement about this text
being an Old Hittite original written in the Old Script. And consequently, a detailed
paleographic system in three major stages (Old Script c. 16501500 Middle Script
15001350 New Script 13501180 BC), each with subdivisions, had found all but
general acceptance.39 With this it was taken for granted that the internal development
of the Hittite text corpus and language had been explained satisfactorily. Once again,
however, it has turned out that every age has its own questions regarding the sources.
Let me again quote Theo van den Hout: Recently, new dissenting voices are being heard
and the currently used dating system has come under pressure. It has been shown by
Jared Miller and Maciej Popko, for instance, that some texts that had been labeled OS,
could only have been written after 1500 BC, that is, when the OS period according to
the above system had already ended. (...) Finally, the fragment of the Zukrai tablet that
had triggered the entire system and provided its sole mooring is nowadays classified as
undetermined (ah.?/mh.?). In traditional chronological terms this means that it could
have been written anywhere between 1650 and 1400 BC. This dating renders it useless
as the basis for the system it was supposed to provide.40 Hittitologists are now faced
with the challenge to verify current ideas about text dating, evolution of the Hittite
language, chronology of the Hittite tablet collections, and as a further consequence, the
actual significance of texts at our disposal for conclusions concerning the Old Hittite
period and Old Hittite cultural traditions.41 We have to reconcile ourselves to the idea that
we have no texts written in Hittite that can be placed securely in the period preceding
the reign of Tudhaliya I son of Kantuzzili, the founder of the new dynasty (mid 15th

2008, pp. 145188, and other papers in the same volume. Cf. also S. de Martino, F.M. Fales, S. Ponchia, op. cit.,
note 8.
38 Th. van den Hout, op. cit., note 31, pp. 1135.
39 Ibidem, p. 22.
40 Ibidem, pp. 28, 29. Cf. also M. Popko, Althethitisch? Zu den Datierungsfragen in der Hethitologie. In:

D. Groddek, M. Zorman (eds), Tabularia Hethaeorum. Hethitologische Beitrge Silvin Koak zum 65. Geburtstag
(Dresdner Beitrge zur Hethitologie 25), Wiesbaden 2007, pp. 575581.
41 Theo van den Houts paper on the ductus of the Alalah VII texts has recently given a new incentive to

the discussion of the Old Script and the origin of Hittite cuneiform, see Th. van den Hout The Ductus of the
Alalah VII Texts and the Origin of Hittite Cuneiform. In: E. Devecchi (ed.), Paleography and Scribal Practices
in Syro-Palestine and Anatolia in the Late Bronze Age: Papers Read at a Symposium in Leiden 1718 December
2009 (PIHANS CXIX), Leiden 2012, pp. 147170. Redefining Old Hittite was the topic of a special, Hittological
workshop during the 58th RAI in Leiden, 1620 July 2012.
222 PIOTR TARACHA

century), even though some documents written in the Old Script might also belong to
the late 16th or the first half of the 15th century BC and some others certainly go back
to earlier texts or oral traditions. Alas, there are no criteria to date these earliest Hittite
texts more precisely.
To conclude and referring once again to the every age has its own questions with
regard to the past idea, I would like to draw attention to a new approach to descriptions
of cult festivals and magical rituals, which constitute the subject of close to 70% of all
the texts found in Hattusa. Large numbers of copies and parallel versions are typical of
this text category, these compositions having been written down over and over again for
close to two hundred years. Juggling this laborious puzzle can lead to a reconstruction of
a master text from a multitude of small fragments. Modern Hittitologists, however, are
reaching beyond a simple philological edition of the texts to ask questions of the Sitz im
Leben kind, taking an interest in the technical side, the scribes workshop, meaning the
mind set behind compilations of earlier texts based on old tablets found in the archives,
etc. There is ample evidence suggesting that repeated stereotypical lists of offerings, and
occasionally also significant parts of a text, may have been copied from descriptions of
entirely different festivals.42 It is also demonstrable that in copying some texts the scribe
was determined more by tablet format than by text content. This working model for the
Hittite scribe in many cases leads to doubts as to whether there was actually any one
master text for the numerous copies and parallel versions in existence.43 Moreover,
the specialized, abbreviated language of these ritual scenarios also holds promise as a
topic for future research.44
The question concerning a common original is even more justified with regard to
magical rituals, that consist as a rule of a series of regularly repeated magical techniques
commonly taken as effective among the ritualists. Considering the Kizzuwatna rituals,
Jared L. Miller observed that at least a significant portion of the Kizzuwatnean ritual
literature at Hattusa was taken over from a previous scribal tradition in Kizzuwatna.
Perhaps much of the original composition of the ritual tradition took place not at Hattusa,
but in Kizzuwatna, and the material was recorded not by Hittite scribes, but by scribes
associated with the state archives of Kizzuwatna.45 This observation can be extended to
include many other rituals descriptions of which have survived in the Hattusa archives,
taking under consideration the Luwian and Hurrian names of ritualists and their origins
in the heavily Hurrianized communities in south-eastern Anatolia and northern Syria. It
42 Cf. M. Popko, Arinna: Eine heilige Stadt der Hethiter (Studien zu den Boazky-Texten 50), Wiesbaden

2009, p. 2.
43 P. Taracha, Hittite rituals as literary texts. What do we know about their original editions? In: M. Hutter,

S. Hutter-Braunsar (eds), Hethitische Literatur: berlieferungsprozesse, Textstrukturen, Ausdrucksformen und


Nachwirken: Akten des Symposiums vom 18. bis 20. Februar 2010 in Bonn (Alter Orient und Altes Testament
391), Mnster 2011, pp. 276281.
44 E. Rieken, Fachsprachliche Merkmale in den hethitischen Ritualbeschreibungen. In: M. Hutter, S. Hutter-

Braunsar (eds), op. cit., note 43, pp. 207216.


45 J.L. Miller, Studies in the Origins, Development and Interpretation of the Kizzuwatna rituals (Studien zu

den Boazky-Texten 46), Wiesbaden 2004, p. 254.


HITTITOLOGY UP TO DATE: ISSUES AND NEW APPROACHES 223

can be assumed that a large majority of texts from Hattusa, which are defined as words
of a foreign ritualist, had very little to do with a Kizzuwatnean or North Syrian original
that could have been older than its travesty from Hattusa even by a few hundred years. In
many cases there are grounds to think that we are dealing with adaptations from Hurrian
or Luwian. The Hittite translators or interpreters, however, had often difficulties not only
with understanding foreign technical terms, but also with the Hurrian (and sometimes
perhaps also Luwian) grammar.46 Besides, they are likely to have compiled foreign texts
in the same way that they edited the Hittite ritual texts instead of attempting a faithful
translation. This new approach in recent research on Hittite rituals is certainly a promising
avenue for future studies.

46 P. Taracha, op. cit., note 43, pp. 281282; cf. also idem, Ugarit Forschungen 41, 2009, pp. 704705.
R O C Z N I K O R I E N T A L I S T Y C Z N Y, T. LXV, Z. 1, 2012, (s. 224230)

JERZY TULISOW

Mongolian and Manchu Studies in Poland

Abstract

Paper describes development of Mongolian and Manchu Studies in Poland starting


with the activities of Jan Potocki, Ignacy Wojciechowski and Jzef Kowalewski. It depicts
works of Marian Lewicki and Stanisaw Kauyski as well as of modern scholars, such
as Byambaa Ragchaa, Agata Bareja-Starzyska, Jan Rogala and Jerzy Tulisow.

Keywords: M
 ongolian Studies, Manchu Studies, Kowalewski, Kauyski, Lewicki,
Kotwicz

Talking about the achievements of Mongolists and Manchurists in one paper might
seem a bit strange, because in the world Manchu studies are often associated with Sinology.
I do not follow this trend here, because, firstly, the Manchu language is much closer
to Mongolian than to Chinese, and secondly, in Poland Manchu studies are covered by
Mongolists.
Currently, both disciplines are being developed in Poland at the University of Warsaw,
in a special department, established in the 1950s. Their beginnings, however, are much
older. Not going back as far as the time of the battle of Legnica, when there was a big
interest in Mongolia among the Poles, we may say that the tradition of Polish research
on the Mongols and the Manchus goes back to the turn of the 18th and 19th centuries. It
was started by two people: Count Jan Potocki and Ignacy Wojciechowski.
Potocki we know: the author of The Manuscript Found in Saragossa and
a globetrotter who travelled the world from Morocco to Mongolia.1 His Journey to the

1 A. Brckner, Jana hr. Potockiego prace i zasugi naukowe, Warszawa 1921.


MONGOLIAN AND MANCHU STUDIES IN POLAND 225

Steppes of Astrakhan2 may still arouse interest today, especially among the scholars of
Kalmyk studies. Wojciechowski,3 on the other hand, is less known. We know that he was
born in 1793 in a Polish family concluding from his surname and that he was a doctor
between 1820 and 1838 we meet him as a physician of the Russian Orthodox Mission
in Beijing. That is where, most probably, he learnt Manchu and Chinese. He mastered
these languages so well, that later on (18441850) he could teach them at the University
of Kazan. Unfortunately, he did not leave any written works. He allegedly did work on
some compendiums of the above-mentioned languages, as well as the Chinese-Manchu-
Russian lexicon, but according to my knowledge, none of them was ever published.
The biography of Wojciechowski is all the more interesting, as it intertwines with the
life of Jzef Kowalewski. And his life was much complicated indeed. Before becoming the
father of Mongolian studies, Kowalewski was a classical philologist.4 The requirement
for him to deal with the exotic eastern languages was a form of repression he suffered
for being a member of the Philomaths Association. Initially, he studied in Kazan and
subsequently in Transbaikalia, where he learnt the basics of Mongolian. In order to master
the language better, he went via Mongolia to China and met Wojciechowski in Beijing. It
seems that the doctor, who had already spent many years in China, was of great help to
him, also with the Manchu language. Altogether Kowalewski spent five years in the East.
In 1833 he went to Petersburg, passed his exams, and became professor of Mongolian
at the University of Kazan. And thus, when years later Wojciechowski came to town,
Kowalewski was already in his eleventh year of work at the university.
Wojciechowski died in 1850 in Kazan, and Kowalewski moved to Warsaw in 1863.
At that time he was already a renowned scholar. He gained the greatest acclaim for his
three-volume Mongolian language dictionary,5 containing, besides French equivalents,
numerous Manchu, Sanskrit and Tibetan parallels for many reasons an invaluable
book even today. It might have been expected that along with the appearance of such a
luminary, Warsaw would become an important centre of Mongolian studies. Unfortunately,
that did not happen. The program of studies in the [Main School] Szkoa Gwna
where Kowalewski was employed did not cover Oriental studies and the outbreak and
subsequent failure of the January Uprising were not helpful for a fruitful academic work.
What is worse, during the Uprising Kowalewski lost a major part of his archives. This
resulted in his final farewell to Oriental studies, after which he devoted himself with
not much conviction, though to teaching history.
2 J. Potocki, Voyage dans les steps dAstrakhan et du Caucase, Paris 1829.
3 R.M. Valeev, Kazanskoe vostokovedenie: istoki i razvitie (XIX v.20 gg. XX v.), Izdatelstvo Kazanskogo
Universiteta, Kazan 1998, p. 149; P.E. Skachkov, Ocherki istorii russkogo kitaevedeniya, Moskva 1977, pp. 195197.
4 See W. Kotwicz, Jzef Kowalewski orientalista (18011878), Prace Wrocawskiego Towarzystwa Naukowego,

Seria A, nr 11, Wrocaw, 1948. Also: Szamow, G.F., Profiessor O.M. Kowalewskij. Ocherk zhizni i nauchnoi
deatelnosti, Izdatelstwo Kazanskogo Universiteta, Kaza, 1983; S. Kauyski, Kowalewski Jzef Szczepan
(18011878). In: Polski sownik biograficzny, PAN, Pastwowy Zakad im. Ossoliskich, 1969, pp. 525528;
Mongoloved O.M. Kovalevskij: biografia i naslediye (18011878), R.M. Valeev (ed.), Kazan 2004, especially
J.Tulisow and R.M. Valeev, Osnovnye etapy zhizni O.M. Kovalevskogo: Vilnus, Kazan, Varshava, pp. 95138.
5 J. Kowalewski, Dictionnaire Mongol-Russe-Franais, Kasan, 1844.
226 JERZY TULISOW

Does that mean that Kowalewskis stay in Warsaw was a lost time for Oriental
studies? Well, not necessarily, because, although Kowalewski himself did not act as an
expert of the East, the word of his adventures spread around, here and there arousing
the already existing interest in the Orient. Who knows if the inexplicable decision of
young Kotwicz to follow the Oriental studies was not influenced by some stories about
Kowalewski? That might have been the case, particularly because both men came from
the same circle of minor gentry of the eastern borders.
Anyway, Wadysaw Kotwicz,6 already as a school student must have been dreaming
of studying eastern languages, and precisely those in which Kowalewski had specialized
before. He made his dream come true when he enrolled in 1891 to the Petersburg
Universitys Faculty of Eastern Languages. Among the languages taught there were also
Chinese and Japanese, but Kotwicz focused on Mongolian and Manchu. At the time of his
graduation (1895) his career developed in two directions: in 1896 Kotwicz took up job in
the Ministry of Finance, and a little later in his parent university. He conducted classes in
two chairs: of Mongolian and of Manchu languages, and soon became their director. As
for his scholarly interests, at that time they focused on the language and folklore of the
Kalmucks. It is worth emphasizing that it was Kotwicz who contributed to the publishing
of the Kalmyks national epic, The Jangar, and wrote the first grammar of the Kalmyk
language. The change of interests came in 1912 with a journey to Mongolia. It resulted
in the discovery of previously unknown inscriptions and archeological monuments and,
more than before, made Kotwicz inclined to history and historic linguistics.
Many of valuable materials brought from that journey and others, collected during
Kotwiczs whole lifetime, including never published texts, are stored in the Kotwicz
collection in the Archive of Science of the Polish Academy of Sciences (PAN) and the
Polish Academy of Arts and Sciences in Cracow under call number K-III-19, archival
units 55-134.7
After he left from the Ministry of Finance (1917), Kotwiczs academic career gained
momentum. In the same year 1917 he became a full-time docent, in 1923 aprofessor,
and in 1924 correspondent member of the Russian Academy of Sciences. However,
in 1924 he was no longer in Russia. When, after the First World War, the independent
Poland was re-established, Kotwicz began his attempts to be repatriated and in 1923 he
succeeded.
In Poland two universities tried to have him: Jagiellonian University and the University
of Jan Kazimierz in Lvov. Kotwicz chose Lvov. There, he took control of the Chair of the
Far-Eastern Philologies, established especially for him. He taught Mongolian and Manchu,
as well as history, with special emphasis on the history of nomads. He continued to do so
almost throughout the whole between-wars period (19241939). When the Second World
War broke out, he moved to the area of Vilnius. He died there on October 3rd, 1944.
6 M. Lewicki, Wadysaw Kotwicz (20.II.18723.X.1944), Wrocaw, 1948; J. Tulisow, Wadysaw Kotwicz,
Hemispheres, 1986, No 2, pp. 199215; J. Tulisow, At the Source of Vilnius Altaic Studies, Acta Orientalia
Vilnensia, 2009, vol. 10, Issues 1-2, ed. by Andrius Beinorius and Marek Mejor, Vilnius 2011, pp. 97110.
7 J. Tulisow, Wadysaw Kotwicz, Hemispheres, 1986, No 2.
MONGOLIAN AND MANCHU STUDIES IN POLAND 227

Kotwicz left an impressive output. Apart from the above-mentioned works, it contains
editions of texts, works on the history of the steppe peoples, biographies of Polish
Orientalists, articles devoted to different linguistic problems, as well the posthumously
published opus magnum: Studies on Altaic languages. It is interesting that this heritage
does not contain almost any strictly Manchurist works. However, we know that Kotwicz
did not stop at just teaching the Manchu language, he also collected texts written in
Manchu and encouraged his students to study it. For example, he induced Muromski to
collect materials in the Sibe dialect.8
And one more thing when speaking about the scholarly output of Kotwicz, we
have to take into account the works which, for various reasons, were never published by
the author, but kept in his archives. Some of them, like Studia nad jzykami atajskimi
[Studies on Altaic languages],9 were published after his death, others still await publication.
Among the latter there is the article La langue mandchoue et sa lexicographie. Although
it has on its front page the words Do not print!, handwritten by Kotwicz, perhaps this
ban is no longer valid today?
Kotwicz had many students, often outstanding, who in time also made academic
careers. Among the listeners to his Lvov lectures, this concerns two people: the Ukrainian
historian (who later worked in the USA) O. Pritsak, and the Polish Mongolist M. Lewicki.10
Here we are more interested in the latter. He had a versatile education (studied also
under P. Pelliot), he easily dealt with various topics, but gained fame for his studies of
the Middle Mongolian language11 and commentaries to Marco Polos Description of the
World.12 Had he lived longer, most probably he would have also had his contribution
to Manchu studies. The author of this paper had in his hand a certain Manchu text
transcribed by M. Lewicki.
After the war M. Lewicki divided his time between Cracow, where he lived, and
Warsaw and Wrocaw. At the University of Wrocaw he was a guest lecturer (at the
Philology Faculty) and in Warsaw he did his habilitation (1949) and took control of the
Chair of the Philology of the Inner Asian Peoples, established (1950) on the basis of
the Indian Philology Seminar of the Institute of Oriental Studies. In 1951 he employed
in his Chair Stanisaw Kauyski13 who had just (1951) received his M.A. diploma in
Wrocaw. At that time Kauyskis specialization was Turkology, but under the influence
of Lewicki, it was soon overshadowed by Mongolian Studies. Kauyski (from 1964 head
of the Chair of the Philology of the Inner Asian Peoples, and from 1979 full professor),
8 Edited later by S. Kauyski, see infra, note 19.
9 In Rocznik Orientalistyczny XVI, 1958.
10 S. Kauyski, Marian Lewicki (15 VIII 190813 XI 1966), Przegld Orientalistyczny, 1956, No. 3 (19),

pp. 299312; an entry in the Polish Biographical Dictionary t. XVII, 1972 and Marian Lewicki (na 25-lecie
mierci), Przegld Orientalistyczny, 1980, no 4.
11 La langue Mongole des transcriptions chinoises du XIVe sicle. Le Houa-yi yi-yu de 1389, Wrocaw 1949.
12 Opisanie wiata [Description of the World], Warszawa 1954, 1975, 2010.
13 See Opera altaistica professori Stanislao Kauyski octogenario dicata, Rocznik Orientalistyczny, vol. LVIII,

No. 1, ed. by Agata Bareja-Starzyska, Jerzy Tulisow, with the assistance of Jan Rogala, Warszawa 2005 with the
biography by Jerzy Tulisow, pp. 912 and the complete bibliography of Prof. Kauyski, pp. 1532.
228 JERZY TULISOW

is the author of the Classical Mongolian language grammar,14 a brilliant translation of


the Secret History of the Mongols15 and many other valuable works on the history and
culture of Mongolia. At least partially devoted to Mongolian studies are also his other
works, like Mongolische Elemente in der Jakutische Sprache (PWN, Warszawa, 1961).
Kauyskis Manchurist works deserve a separate mention. There are more than
a dozen of them. Apart from one the translation of Manchu maxims in the anthology
Voices from the Yurt16 they concern the Sibe people and their language, which is in
fact a Manchu dialect. I already mentioned it when I talked about the Manchu interests
of Kotwicz. Works of Kauyski are based precisely on the materials of Kotwicz, and
in particular on those given to him by Muromski. By publishing them,17 Kauyski gave
a strong impulse to start new research of the so far rather neglected Manchu language.
In 1997 Professor Kauyski was replaced as head of the Chair by Professor
Godziski.18 This scholar specializes in the studies of Mongolian language. Even though,
as a student of Kauyski, he also possesses deep knowledge of other Altaic languages,
yet he was never interested in researching them. Instead, he is successful in the field
of Tibetan studies. We might even say that he is a pioneer in this discipline in Poland.
Others have already spoken about Godziskis achievements as a Tibetologist.19 I will
focus on his Mongolist works. There is no doubt, that one of the most important is the
monograph on the Middle Mongolian language: Jzyk redniomongolski: sowotwrstwo,
odmiana wyrazw, skadnia [The Middle Mongolian Language: Word Formation,
Morphology, Syntax], (Rozprawy Uniwersytetu Warszawskiego, Warszawa, 1985) his
habilitation dissertation and the book Modern Mongolian Language.20 These works,
along with the articles on Mongolian affixes and scripts used by the Mongols in different
epochs, established Godziskis reputation as a linguist. However, a true Mongolist does
not limit himself to linguistic problems only. Thus, Godziski also authored an article on
Mongolian art,21 an exemplary publication of Dagur/Daur fairy tales from the collection
of Kotwicz,22 and a popular publication In the Circle of Lamaist Legends and Myths.23

14 Klasyczny jzyk mongolski, [Classical Mongolian Language], Wydawnictwo Akademickie Dialog, Warszawa
1998.
15 Tajna historia Mongow, [The Secret History of the Mongols], PIW, Warszawa 1970, 1989, 2005.
16 Gosy z jurty [Voices from a Yurt], Wiedza Powszechna, Warszawa 1960.
17 Kauyski S., Die Sprache des mandschurischen Stammes Sibe aus der Gegend von Kuldscha, PWN,

Warszawa 1977.
18 See Altaica et Tibetica. Anniversary Volume dedicated to Stanisaw Godziski on His Seventieth Birthday,

Rocznik Orientalistyczny, vol. 63 (1), 2010. Biography on pages 1315.


19 J. Rogala, Stanisaw Godziski. In: Altaica et Tibetica. Anniversary Volume dedicated to Stanisaw Godziski

on His Seventieth Birthday, Rocznik Orientalistyczny, vol. 63 (1), 2010, pp. 1415.
20 Wspczesny jzyk mongolski, [Modern Mongolian Language], Wydawnictwo Akademickie Dialog, Warszawa

1998.
21 Sztuka mongolska [The Mongolian Art], Etnografia polska, 24, 1980.
22 Deux contes dagours de la rgion de Kouldja... in: Mongolian Studies, Budapest 1970.
23 W krgu lamajskich legend i mitw, [In the Circle of the Lamaist Legends and Myths], Iskry, Warszawa,

1981.
MONGOLIAN AND MANCHU STUDIES IN POLAND 229

This scholar also gave us the translation of The History of Mongolia by the Mongolian
publicist Baabar.24
In 2003 Godziski quit his post as head of the Chair and in the same year was
replaced by Dr Agata Bareja-Starzyska. She is a graduate of Mongolian studies, but
studied also Tibetan and Sanskrit, and took part in the Buddhology seminar. During her
studies she had a language practice in Ulan Bator (19861987) and an internship in
Budapest (1988). As an assistant in the Institute of Oriental Studies of the University
of Warsaw, she had scholarships at the universities in Oslo (1993), Bonn (1996) and,
as part of the university exchange, in Bloomington (2000). In 1999 she defended her
Ph.D. dissertation on ikula Kereglegi, a Mongolian handbook of Buddhism.25 The main
sphere of interest of Dr Bareja-Starzyska is Buddhism in Mongolia, Mongolian-Tibetan
relations, as well as literatures of Mongolia and Tibet.26
When talking about the research on Mongolian Buddhism carried out in Warsaw, we
cannot omit R. Byambaa, who is employed as a language teacher. This former member
of the Mongolian Academy of Sciences and the director of Gandan monasterys library,
has for many years studied Mongolian writings in Tibetan language and published several
works devoted to this topic.27
As for other employees in the Chair, we may say that Mongolists are also Dr Jan
Rogala and Dr Jerzy Tulisow. The former specializes in contemporary Mongolia and
in linguistics. His Ph.D. dissertation was entitled The Verb Word Formation in Modern
Mongolian Language.28 The latter published a work devoted to the language of the Secret

24 Dzieje Mongolii, [History of Mongolia], Wydawnictwo Akademickie, Dialog, Warszawa 2005.


25 Chikula kereglegchi. Zasady buddyzmu, [Chikula kereglegchi. Buddhist Principles] by Shireget Gushi
Chorji, Wydawnictwa Uniwersytetu Warszawskiego, Warszawa 2006.
26 The History of Ancient Tibet According to the 17th Century Mongolian Chronicle Erdeni-yin tobi by Sagang

Seen. In: Proceedings of the 5th International Seminar on Tibetan Studies in Narita, Japan 1989, Narita 1992,
pp.341351; A. Bareja-Starzyska, H. Havnevik , A Preliminary Study of Buddhism in Present-day Mongolia. In:
Mongolians from country to city. Floating boundaries, pastoralism, and city life in the Mongol lands, Kopenhaga,
2006, NIAS (Nordic Institute of Asian Studies), pp. 212236; The Growth of the Religious Authority of the First
Jebtsundampa of Mongolia (16351723). In: Biographies of Eminent Mongol Buddhists, PIATS 2006: Tibetan
Studies: Proceedings of the 11th International Seminar on Tibetan Studies in Knigswinter 2006, Johan Elverskog
(ed.), International Institute for Tibetan and Buddhist Studies GmbH: IITBS Halle (2008), pp. 4957; The Mongolian
Incarnation of Jo nang pa Trantha Kun dga snying po : ndr Gegeen Zanabazar Blo bzang bstan pai rgyal
mtshan (16351723). A case study of the Tibeto-Mongolian Relationship. In: Proceedings of the International
Seminar on Tibetan and Himalayan Studies, Held at the Library of Tibetan Works and Archives, September 2009
on the Occasion of the Thank you India Year, Roberto Vitali (ed.), The Tibet Journal, Autumn 2009Summer
2010, vol. XXXIVXXXV, s. 243261.
27 The Square Script of Zanabazar (1997, new ed. 2005), Biographical Guide of Mongolian Writers in the

Tibetan Language and Mongolian Translators, vols. IIII, 20032004; Reedition of the Collected Works of emitent
Mongolian Buddhist Masters including The Collected Biographies of Jebzundampa Khutugtus of Khalkha, Ulaan
Baatar 2006.
28 Sowotwrstwo czasownika wspczesnego jzyka mongolskiego derywacja i kompozycja, [The Verb

Word Formation in Modern Mongolian Language Derivation and Composition, Wydawnictwa Uniwersytetu
Warszawskiego, Warszawa 2007.
230 JERZY TULISOW

History of the Mongols29 and several translations of Mongolian poetry.30 Later, however,
he gave up Mongolian studies in order to focus on the culture of peoples living near the
borders of Mongolia, the history of the steppes and Manchu studies. He wrote, among
other things, a grammar of Manchu31 and a handbook of this language,32 as well as two
books about the folklore of the Tungusic people,33 including the Manchu and the Sibe.
It is worth stressing that under his direction already two M.A. theses devoted to Manchu
studies have been written, so there is hope that this discipline will be developed in Poland.

29 Terminologia kultury materialnej w Tajnej historii Mongow, Acta Philologica nr 5 (r. 1972), pp. 153195.
30 Aforyzmy mongolskie, przekad z mongolskiego, Przegld Orientalistyczny nr 86 (r. 1973), pp. 114,
120; C. Damdinsuren, Tol Genden, przekad z mongolskiego, Przegld Orientalistyczny nr 129-132 (r. 1983),
pp. 107115; D. Nacagdord, Mj kraj, przekad z mongolskiego, Przegld Orientalistyczny nr 141 (r. 1986),
pp. 8385.
31 Jzyk mandurski [The Manchu Language], Wydawnictwo Akademickie Dialog, Warszawa 2000.
32 Krtki kurs jzyka mandurskiego, [A short Course of the Manchu Language] Wydawnictwo Akademickie

Dialog, Warszawa 2004.


33 Legendy ludw Mandurii, [Legends of the Manchu Peoples], vol. I and vol. II, Wydawnictwo Akademickie

Dialog, Warszawa, 1997, 1998; Folklor ludw tunguskich, Folklore of the Tungusic Peoples] Wydawnictwo
Akademickie Dialog, Warszawa 2009.
R O C Z N I K O R I E N T A L I S T Y C Z N Y, T. LXV, Z. 1, 2012, (s. 231235)

VLADIMIR USPENSKY

Rivalry of the Descendants of Chinggis Khan and His Brother


Khasar as a Factor in Mongolian History

Abstract

Relations between Chinggis Khan and his younger brother Khasar were somewhat
strained at times. In accordance with Mongolian tradition Chinggis Khans brothers and
their descendants were allotted lands in the eastern part of the Mongolian Plateau. This
fact became a crucial point in the early seventeenth century when descendants of Khasar
turned rather to the resurging Manchu state than to the last Mongolian ruler Ligdan Khan.
At that time another famous Khasars descendant Gushi Khan crushed the remaining
supporters of Ligdan Khan in the Qinghai area and established the paramountcy of the
Gelugpa school in Tibet with his military forces. Owing to continuous intermarriages
with Manchu Qing emperors Khasars descendants became closely linked to the ruling
dynasty during its entire reign.

Keywords: Mongolia, Chingis Khan, Khasar, Mongolian Empire, Manchus, history

Relations between the world conqueror Chinggis Khan and his brothers and other
relatives were somewhat strained at times. This is also true about his relations with Khasar
(also styled as Jochi Khasar, Khabutu Khasar; ca. 11641227), his younger brother. They
both were sons of Yisugei Bagatur and Ogelun Uijung; they both killed their half-brother
Bekter in their childhood. Khasar participated in many small and major military campaigns
waged by his elder brother and became famous for his marksmanship. Cases of discord
between the two brothers were vividly described in the Secret History of the Mongols
(Mong. Nigua tobiyan; Chin. Yuan chao mi shi ).
The land, people and resources of Mongolia and later of the Mongolian Empire were
regarded as being the common property of all the Chinggis Khans family. This was
232 VLADIMIR USPENSKY

an unchangeable political principle of the Mongol state. Already at the early stages of
his conquests Chinggis Khan allotted lands and people to his relatives and the highest
military chieftains. His brothers and their descendants were allotted lands in the eastern
part of the Mongolian Plateau along both sides of the Great Hinggan Range.1 For the
time being, his youngest son Tolui inherited the lands in central Mongolia while his elder
sons had their appanages far to the west of the Altai Mountains.
The collapse of the Mongol rule in China in 1368 led to the Western Mongol Oirat
hegemony in the first half of the fifteenth century. Under the Oirat pressure some of the
Mongols who were under the rule of Khasars descendants migrated even further to the
East the area of the River Nonni . They became neighbours of the Jurched the
descendants of the Jin dynasty, which was crushed by the Mongols in 1234. At the end
of the sixteenth century the Jurched2 tribes were unified by their able chieftain Nurhachi
(b. 1558; r. 16161626). At the same time, the last Mongolian great Khan, Ligdan Hutugtu
Khan (r. 16041634) made an attempt to unify a rather loose confederation of the Mongols
and to establish a nomadic empire. This led to a conflict with the resurging power of
the Jurched and with Mongolian local rulers as well. The eastern Mongol rulers soon
turned to Nurhachi and not to Ligdan Khan. The Mongolian aimags which were ruled
by hereditary rulers, descendants of Khasar were: Khorchin, Jalayid, Dorbed, Gorlos,
Muumingan, Durben-Khukhed, Aru-Khorchin and Urad. The largest of them was Khorchin
called the state (Mong. ulus or tmen, Chin. guo ). In 1624 Khorchin ruler Uuba
swore an oath of allegiance to Nurhachi. This led to a punitive expedition by Ligdan
Khan which was repelled with the assistance of the army sent by Nurhachi. Within afew
years all Mongolian tribes which were ruled by the descendants of Khasar accepted the
suzerainty of Jurched rulers.3 In 1632 Nurhachis heir Hong Taiji (r. 16271643) launched
a massive military campaign against Ligdan Khan. In this campaign he was assisted by
his eastern Mongolian allies with their military units. Ligdan Khan retreated to Kukunor
(Qinghai ) and died of smallpox in 1634.4
The Manchus regarded establishing dynastic marriages as being very instrumental
for gaining loyalty of their Mongol allies and subjects. During the formative stage of the
Manchu dynasty intermarriages between the eastern Mongol and Manchu ruling clans
were widespread.
Genealogy of Khasars Descendants (a fragment)
Khabutu Khasar
15th generation Buyandara
16th generation Ceceg Namsai
17th generation Ongodai Mangus
18th generation Uuba (d. 1632) Jayisang
19th generation Manjushiri

1 Qura 2001: 1213.


2 The ethnic name Manchu was introduced in 1635.
3 Yermachenko 1974: 2345.
4 Heissig 1979: 2340.
RIVALRY OF THE DESCENDANTS OF CHINGGIS KHAN AND HIS BROTHER KHASAR... 233

A daughter of Mangus, named Jere (15991649) was Hong Taijis empress. She
gave birth to three daughters one of whom, Makata, was married to the elder son of
Ligdan Khan. Another daughter of Mangus was married to Dorgon (16131650), the
mighty regent during Shunzhi Emperors infancy. A daughter of Jayisang named
Bumbutai (16131688) was Hong Taijis concubine and gave birth to his ninth son Fulin
who later became Shunzi Emperor (r. 16441661). She was given the title of Xiao
zhuang Wen huanghou .5
Intermarriages between the ruling Manchu family and the Mongol nobility, especially
the descendants of Khasar, continued until the very end of the dynasty.6 Qianlong
Emperor (r. 17351795) while travelling through Khorchin ayimag composed verses
which began with the following words:

Though this northern land is said to be a distant one


It has been linked to the Golden Clan since old days.7

Another faction of Khasars descendants became hereditary rulers of the Oirat tribe
of the Hoshot. They were captured and moved to the West in the middle of the fifteenth
century in the days of the Oirat domination. However, with the decline in power of
the Choros tribe they became a dominating force among the Oirats. The rulers of the
Hoshots were the only ones among the Oirat nobility who belonged to Borjigid clan.8 In
the first part of the seventeenth century their able leader Gushi Khan (15821655) who
was a seventeenth generation descendant of Khasar managed to subjugate the whole of
Tibet and Kukunor.
Thus Gushi Khan was a relative of the Manchu rulers and descended from the same
famous progenitor. The vital role of genealogical ties is basic in the polity of nomadic
societies such as those of the Mongols and the early Manchus, and the importance of
this fact should not be underestimated.
Also, Gushi Khan was an ally of the Manchus in their fight against the Mongol
supporters of Ligdan Khan. The most powerful of them was Chogtu Khungtaiji (15811637),
a mighty chieftain from Khalkha Mongolia who moved to the Kokonor area but failed
to join Ligdan Khan because of the latters death. Gushi Khan moved his armies to
Kokonor and defeated Chogtu Khungtaiji in a decisive battle in 1637.9
In this way the Manchus with their eastern Mongol allies and the Hoshot Mongols
led by Gushi Khan defeated the last Mongolian Great Khan Ligdan and his potential
supporters. After this the Manchus turned their further conquests to China proper and

5 Hummel 1943, vol. 1: 300301; Veit 1990, Teil I: 1819.


6 Jagchid 1984.
7 Jang Mu 1988, vol. 1: 1011. By the Golden Clan (Aisin Gioro in Manchu) here is meant the Manchu

imperial clan.
8 Okada 1987: 203207; Qura 2001: 103113.
9 Ho-Chin Yang 1969: 3637.
234 VLADIMIR USPENSKY

Gushi Khan directed his attention to Central Tibet. Eastern Mongolian rulers were active
in the military campaign in China in 1644 and afterwards.
It should not be regarded as an exaggeration that in the seventeenth century the
descendants of Khasar reached their utmost might and glory. The most famous descendant
of Khasar in the late Qing period was Senge Rinchin (18111865), a Khorchin prince.
He was one of the powerful military commanders during the Taiping Rebellion and the
Second Opium War. When he was killed in an ambush while suppressing rebels in the
Shandong province, a tablet with his name was installed in the Imperial Ancestral Temple.10
Khasar was much glorified in the traditional Mongolian historiography, especially
in the works written by the authors who originated from Mongolian aimags ruled by
his descendants. Altan tobi by Mergen Gegen or Bolor toli by Jambaldorji may
serve as examples of such works. Khasar replaced his elder brother Chinggis Khan in
many glorious episodes of the ancient Mongolian history described in these chronicles.11
The close relationship between the Manchu ruling house and the descendants of Khasar
even found its indirect reflection in the imperially commissioned voluminous collection
of biographies of the Mongol princes commonly known under its shortened Mongolian
title Iledkel astir. It begins with the genealogy of the Khorchin princes thus speaking
about their distant progenitor Khasar prior to his elder brother.12
The territorial neighbourhood between the Manchus and the Eastern Mongols who
were ruled by the descendants of Khasar turned out to be a crucial point in the emergence
of the Qing empire. Also it was his descendants who almost simultaneously changed the
history of Tibet. This affirms the fact that the genealogical factor is not to be neglected
in the studies of Asian history.

Bibliography

Heissig 1979 Heissig, W. 1979. Die Zeit des letzten mongolischen Grosskhans Ligdan (16041634).
Opladen: Westdeutscher Verlag.
Ho-Chin Yang 1969 Ho-Chin Yang. 1969. The Annals of Kokonor. Bloomington: Indiana University.
Hummel 1943 Hummel, A.W. (ed.). 1943. Eminent Chinese of the Ching Period (16441912).
Washington: United States Government Printing Office, vols. 1-2.
Jagchid 1984 Jagchid, S. 1984. The Sinicization of the Mongolian Ruling Class in the Late Manchu
Ching Period. Transactions of the International Conference of Orientalists in Japan. Tokyo: The Toho
Gakkai, Nos. 2829.
Jang Mu 1988 Jang Mu . 1988. Mongol-un qosiu nutu-un temdeglel. [Beijing]: ndsten-u
keblel-n qoriy-a, vol. 1.
Okada 1987 Okada, Hidehiro. 1987. Origin of the Drben Oyirad., Ural-Altaische Jahrbcher,
Bd. 7, pp. 203207.
Qura 2001 Qura. 2001. ang Ming. Qorin tobiyan. [Beijing]: ndsten-u keblel-n qoriy-a.

10 Hummel 1943, vol. 2: 632634.


11 Tsendina 2007: 5767.
12 Veit 1990: 9293.
RIVALRY OF THE DESCENDANTS OF CHINGGIS KHAN AND HIS BROTHER KHASAR... 235

Tsendina 2007 Tsendina, A.D. 2007. Mongolskiye letopisi XVIIXIX vekov: povestvovatelnye
traditsii. Moscow.
Veit 1990 Veit, V. 1990 . Die vier Qane von Qalqa. Ein Beitrag zur Kenntnis der politischen Bedeutung
der nordmongolischen Aristokratie in den Regierungsperioden Kang-hsi bis Chien-lung (16611796)
anhand des biographischen Handbuchs Iledkel astir aus dem Jahre 1795. Wiesbaden: Harrassowitz, Teil I.
Yermachenko 1974 Yermachenko, I.S. 1974. Politika manchzhurskoi dinastii Tsin v Yuzhnoi
i Severnoi Mongolii v XVII v., Moscow: Nauka.
R O C Z N I K O R I E N T A L I S T Y C Z N Y, T. LXV, Z. 1, 2012, (s. 236253)

MONIKA ZIN

When Stones are All that Survived:


The Case of Buddhism in Andhra

Abstract

The conclusion of the project of the University of Munich, aiming at a new publication
of all narrative reliefs from the Amaravati School of sculptures (Andhra Pradesh 1st c.
BCE the 4th c. CE) is that there are many reliefs in Andhra for which no explanation
can be given, neither in Pali nor in the texts of the northern schools. The reason for
this is that not a single text is available today from the Buddhist culture of the region
(predominantly the Caitika or aila Schools associated later with the Mahsghikas), so
that the narrative reliefs are all that survived today. A typical example presented here is
a relief from Amaravati (Fig. 1), explained previously as illustration of the Morajtaka,
which in fact must be a different narration of the peacock that gave or saved a child,
perhaps from a serpent bite. The story is not known today.

Keywords: Buddhist art, Amaravati, sculpture, Buddhist narrative literature, Mahsghika

The University of Munich hosts the project undertaken by the German Research
Society (DFG), aiming at a new publication of reliefs from the so-called Amaravati
School of sculptures. The project includes narrative reliefs not only from Amaravati
but also sites like Ghantasala or Gummaddidurru, as well as from the later centres of
Nagarjunakonda and Goli, i.e. it includes narrative reliefs from all the Buddhist sites in
Andhra Pradesh from the 1st c. BCE till the 4th c. CE. Initially the project was to be
based on the black and white photographs taken in the early 1980s. Thanks to the DFGs
generosity in covering all travel expenses (including renting an off-road car) and thanks
to the permission granted by the Archaeological Survey of India to take pictures, the
WHEN STONES ARE ALL THAT SURVIVED: THE CASE OF BUDDHISM IN ANDHRA 237

project can benefit not only from standard photographs but also from numerous digital
photos that allow for an enlargement of details.
Since the relative chronology of the Amaravati School based on a stylistic analysis
of the reliefs has been established1, the Munich Project can concentrate on their narrative
content. As a matter of fact, the primary objective is not solely an interpretation of reliefs
but rather an identification of the school affiliations of Buddhism in the region and their
characteristics. It would indeed be of utmost importance to ascertain if the reliefs had
Theravda Buddhism as their literary tradition or the Hnayna Schools of northern
Buddhism, in which Mahyna Buddhism developed over the centuries. The schools
known from the inscriptions in the region are predominantly the Caitika or ailas,2 which
were associated in later sources with the Mahsghikas.3
The main reason for undertaking the study, which should culminate in the publication
Narrative Reliefs of the Amaravati School, was the observation that technically poor
publications of very many reliefs not only excluded the possibility of using them for
the sake of research but also that the research was in a way hampered by the erroneous
interpretations provided in those publications.

Anyone who at least generally has been dealing with the art of Amaravati, will in
all probability be surprised to hear that there is a need for re-interpreting the reliefs. After
all, in the standard publication by Sivaramamurti,4 which influenced later interpretations
(including the sculptures in the British Museum),5 only a handful of the reliefs are described
as unidentified, the rest have been given either Sivaramamurtis own identifications or
interpretations taken from earlier works.6 However, these identifications are frequently
uncertain or simply erroneous, as they only take as starting point some detail of the
relief.

1 For the dating of the Amaravati reliefs cf. e.g. Knox 1992, pp. 31ff. (with references to earlier research);

for more resent research cf. a study by Akira Shimada (2006) on the great railing in Amaravati.
2 Lamotte 1958, pp. 580581.
3 Bareau 1955, pp. 2333, for the characteristics of the schools cf. ibid., pp. 87105.
4 Sivaramamurtis book Amaravati Sculptures in the Madras Government Museum, published in Madras 1942

was reprinted several times (1956, 1977, 2000) and is still considered a standard publication on the subject.
5 For interpretations of the Amaravati reliefs kept in London cf. Knox 1992, ibid. references to earlier

research.
6 Sivaramamurti gives references and also lists the scholars he took the identifications from (1942: XV): Since

the publication of The Buddhist Stupas of Amaravati and Jaggayyapeta by Burgess in 1887, excavations by Rea
have added a number of sculptures to the already rich collection in the Madras Museum. Great scholars such as
Foucher, Coommaraswamy, Vogel and Dubreuil have written discussing problems concerning the Amaravati stupa
and have identified some of the scenes in the sculptures. Bachhofer, Mlle. Linossier, Rene Grousset, Ramachandran,
and Barnet Kempers have identified other scenes. Mr. Ramachandrans paper on two labelled early scenes on
apillar from Amaravati has definitely brought early Amaravati and Bharhut together. Hultzsch and Burgess edited
the Amaravati inscriptions and later Franke corrected their readings. Many inscriptions of the old collection that
remained unpublished and those on the slabs dug out later and brought to the Museum were edited by Chanda.
But many scenes still remained to be identified, some inscriptions still unread and some uncorrected.
238 MONIKA ZIN

A typical example is a relief from Amaravati (Fig. 1),7 identified as the Morajtaka.8
The depiction was placed on a railing pillar from which only the lower part is preserved
today, covering the lower part of the medallion and the fluted area below.
The representation inside the medallion is partially preserved but the significant
elements that could link it with a particular story, are missing. The scene had originally
depicted a royal court encamped in the park, in front of the beautiful lotus pond. On
the left-hand side, to the left of the wide throne on which a male person surrounded by
ladies is seated, there are rests of a tiny woman moving to the left in a bent posture. The
woman displays the characteristic pose of a female guard, pratihr, who in this case is
probably going to receive someone arriving to meet the royal personage.
The reason for identifying the relief as the Morajtaka is given below. Mora, i.e.
the peacock, is actually shown here, in the centre of the fluted area. The jtaka9 narrates
astory about the golden peacock which was brought to the royal court where he answered
the kings questions and instructed him. According to generally applied rules of Indian
pictorial representations, the peacock should be seated at a height that is at least on par
with that of the king whom he instructs, if not even higher; this is however not the
case here. More importantly, there are details in the relief which can by no means be
matched with the jtakas content: above the peacock a monk (? or at least a tonsured
person wearing a simple cloth covering his shoulders) is represented. The man is holding
his left arm in a particular way while touching his left shoulder, while with his right
he is pointing to a little infant lying on the lap of the seated male figure. The man, as
also a woman behind him, clasp their hands together in the gesture of worship which
is undoubtedly meant for the peacock. The scene plays outdoors, as is suggested by the
depiction of a tree at the back.
The register on the right provides numerous possibilities for interpretation and can
be linked to different narratives, since it shows a male person of rank sending off or
receiving a messenger. What may be significant here is the portrayal of the man on the
throne in a state of apparent grief.
The left side, though seriously damaged, reveals several elements of a distinctive,
obviously unique, story. In the upper section, we see a male person standing and holding
in his arms another smaller person, probably a child. The right foot of this smaller person
(of whom we see only the right leg preserved) has an instrument (knife?) being applied
to it by another man. Below we see a woman kneeling in front of another figure seated

7 Chennai Government Museum, No. 158, ill. e.g.: Burgess 1887, Pl. 8.2; Bachhofer 1929, Pl. 113.1;
Sivaramamurti 1942, Pl. 32.1; Stern/Bnisti 1961, Pl. 36b; Sugimoto 2001, Fig. 15.
8 The identification of the relief as the Morajtaka was first given in 1914 by Foucher at the conference in

Muse Guimet (the paper was published only in 1928) and repeated since without any reservation. Sivaramamurti
incorporated the interpretation into his standard work (1942, p. 226): The Peacock preaches the Law, Mora Jtaka.
9 Morajtaka No. 159, ed. Vol. 2, pp. 3338; transl. pp. 2326. The story as told in the Mahmorajtaka,

No.491 (ed. Vol. 4, pp. 332342; transl. pp. 210216) should not be taken as a possible textual basis, since there,
the peacock is not brought to the court at all. The hunter, moved by his teachings, stays in the woods as the
Pratyekabuddha.
WHEN STONES ARE ALL THAT SURVIVED: THE CASE OF BUDDHISM IN ANDHRA 239

on a chair (only the legs of this person have been preserved). The kneeling woman is
holding an object, apparently a round bowl.
All these details of the depiction the infant, the monk (?), the operation on the foot
which do not find mention in the Morajtaka, make it clear that it is not this Jtaka that
is illustrated here. What we see is, rather, an illustration of a story in which the peacock
participates, without however preaching. The story must be the narration of the peacock
that gave or saved a child (or perhaps its mother?) perhaps from a serpent bite thus
the operation on the foot may represent the cleaning of the wound after a serpent bite.

In the standard publication Amaravati Sculptures in the Madras Government Museum


Sivaramamurty citing earlier interpretations or presenting his own sometimes notices
that the details do not match the literary sources, after which he writes that the artist has
taken the liberty of adopting the story. However in those reliefs whose interpretations
are certain, such as Vivantara or adanta, no deviations from literary traditions can
be found.
Sivaramamurty a priori accepts Pali literature as the literary basis for the reliefs,
and he uses such texts as the Nidnakath, the Dhammapadahakath or prose from
the Jtaka. These texts, from the 5th century, should not be applied without reservation
to older reliefs. Naturally the sources may contain older stories, but what is it that really
tells us to draw upon them to provide explanations for the Amaravati reliefs in the first
place? Even in the usual representations of episodes from the life story of the Buddha,
there appear incidents not known in the Pali tradition at all. One such incident is the
presentation of the new-born Bodhisattva to the Yaka of the kya clan, very popular
in the relief depictions of the Amaravati School10, the quarrel in the house of Mra,
where his good sons do not allow Mra to disturb the Buddha in his efforts to achieve
enlightenment,11 or the subjugation of the Nga Apalla12. Also several jtakas illustrate
versions of the stories known from the northern traditions as I wrote in another
publication.13 Sometimes the stories are totally unknown in Pali.

10 The episode is represented in the Amaravati-School at least 15 times (e.g. British Museum, No. 44, it

was illustrated several times, for example in Knox 1992, No. 61, p. 121); it shows the Yaka coming out of his
tree-sanctuary to pay homage to the new-born Bodhisattva. The episode is not known in the Pali scripture and
corresponds with the story as rendered in the Vinaya of the Mlasarvstivdins (cf. Schlingloff 2000, No. 64(13),
pp. 338339). The earliest depiction of the episode was recently discovered in the stpa at Kanaganahalli.
11 Also this episode is not rendered in the Pali tradition and known only from the northern sources (cf.

Schlingloff 2000, No. 80). The quarrel in which some of Mras sons try to stop their father is depicted several
times in the Amaravati School, among others in two particularly fine representations on the ua stones kept in
the Chennai Government Museum, No. 10 (illus. e.g: Sivaramamurti 1942, Pl. 57.1; Stern/Bnisti 1961, Pl. 23a)
and No. 153 (illus. e.g: Burgess 1887, Pl. 21.2; Sivaramamurti 1942, Pl. 42.1).
12 For literary sources and pictorial representations cf. Zin 2006, chapter 3.
13 Zin 2004, with the identification of one Amaravati relief (Chennai Government Museum, No. 148) as the

narrative of Mukapangu according to the northern sources, and of one relief from Nagarjunakonda (Nagarjunakonda
Museum, No. 19) as the story of Prabhsa, not known in relevant Pali sources at all.
240 MONIKA ZIN

It must also be noticed here that in Nagarjunakonda (3rd c.), where the affiliations
of several monasteries are substantiated by inscriptions, no narrative reliefs have been
found on stpas belonging to the Theravda, nor in the monasteries of the Mahsakas
and the Bahrutiyas. These schools were apparently not too much interested in the
pictorial representations of the stories, even if the Buddha images were discovered in
their monasteries.
The main problem with the Amaravati School, however, is that there are many reliefs
there for which no explanation at all has been given, neither in Pali nor in the texts
of the northern schools. Coming back to our relief (Fig. 1), it may be concluded that
the relief does not illustrate the Morajtaka, also that the story that it illustrates is not
known today.
What is characteristic of the entire Amaravati School is that, after rejecting the doubtful
interpretations, it must be stated that the literary basis of the reliefs is not known today.
The Mora-Jtaka is just one example among many.
The question that arises here is whether we can ever get to know the correct
interpretations. It is possible that we never may. These interpretations may, for instance,
have belonged to the Vinaya of the Caitika or aila Schools, known from inscriptions in
the region (cf. fn. 2). These Vinayas are lost. We can only hope that the narrative that
has been depicted perhaps survived in the texts of other schools.
For the time being we can only rely on the pictorial tradition which should be analysed
much more carefully and considered as important as the textual sources.

The reliefs of the Amaravati School only seldom illustrate a story just once; usually
we are dealing with several representations of one and the same narrative. This is not
only true of reliefs whose contents have been explained but also unidentified ones; i.e. the
representation may be unexplained but it is clear that it has been repeated. Acomparison
of the depictions might bring further details of the story or ascertain which of the details
are important for the illustration of the narrative, on the basis of their being repeated in
every relief.
The representations of the same story reappear in two forms. The first one involves
areplica of the visual form. A good example of this is a scene on a frieze in the Chennai
Government Museum, No. 105 (Fig. 2)14 showing a king, or some other person of rank,
being attacked by several soldiers. It is quite obvious that in view of the violent actions
of the assailants, so clearly depicted, the frightened looks of the women, the harassment
of the main character one of the assailants is pulling at his cloth while on the verge of
striking him with the object held in his right hand the explanation given by Sivaramamurti
that Prince Siddhartha lives in three pleasant palaces carefully guarded from the ills of
life15 is absolutely unacceptable. Even if the scene is not properly identified, we must

14 Chennai Government Museum, No. 105, 3rd register, illus. e.g.: Burgess 1887, Pl. 21.2; Sivaramamurti 1942,

Pl. 42.1; Parimoo 1982, Fig. 119; Rao 1984, Pl. 241; Zin 2004, Fig. 10 (drawing).
15 Sivaramamurti 1942, Pl. LIX, I c.
WHEN STONES ARE ALL THAT SURVIVED: THE CASE OF BUDDHISM IN ANDHRA 241

agree that the story was obviously known since its representation was reproduced. The
replica is to be found on one of the Amaravati slabs depicting a stpa, to be found today
in the British Museum, No. 69 (Fig. 3 a, b)16. The tiny register repeats quite accurately
the entire scene from the frieze in Chennai. Except for one servant in the upper right
corner (for whom there was no more space), all the other persons are shown in the same
arrangement and making the same gestures.
The second type of representing what is apparently the one and the same narrative
differs from the first in that the content of the story is represented but the composition
of the representation differs. This kind of depiction is characteristic of the majority of
the narrative reliefs; first and foremost it applies to all those whose explanations are
certain: there are always distinctive elements of the iconography that make identification
of the entire story possible, like for instance, the hare jumping into the fire in the aa
narrative, or the king cutting off his flesh for ibi (in the Amaravati School actually
Sarvadada) etc., but the rest of the depiction, the arrangement of the persons in the
composition etc., varies.
What is of great importance is to distinguish the iconographic features of the story,
which can only be done by comparing the various reliefs. To avoid a feeling of frustration
that the Amaravati reliefs defy identification, it is necessary that a positive example be
provided here.
A relief in the British Museum, No. 37 (Fig. 4)17, has been identified by Parimoo18
as a representation of the Kusajtaka.19
Only by comparing the relief with two other illustrations of the same narrative (Figs.
5 and 621) does a clear identification emerge. All three reliefs represent a king leaving
20

the city to war; the battle is depicted in front of his elephant. All three representations
show the king riding on an elephant together with his consort, and in all three of them
there is a flying god above the couple, carrying a necklace. Only in one of the depictions,
however (Fig. 6), is the god identifiable as Indra through his distinctive crown. This
corresponds as Parimoo observed with regard to the relief in London (Fig. 4) with the
text of the jtaka, precisely speaking with the gths 78-81,22 which tell us that Kusas
wife Pabhvat sat on the elephant behind her husband to whom Indra brought a solar

16 British Museum, No. 69, illus. e.g.: Stern/Bnisti 1952, Pl. 43; Knox 1992, No. 68, p. 130.
17 British Museum, No. 37, illus. e.g.: Burgess 1887, Fig. 8, p. 38 (drawing); Barrett 1954, No. 33, Pl. 24 (only
upper area); Rao 1984, Pl. 178; Knox 1992, No. 15, p. 67; Parimoo 1995, Fig. 2.
18 Parimoo 1995, pp. 130132.
19 Kusajtaka, No. 531, ed. Vol. 5, pp. 278312; transl. pp. 141164. The identification did not seem to be

wholly conclusive because Parimoo took the typical depiction of a woman holding a fan (in the small preserved
fragment of the fluted area below the central lotus rosette) to be the princess Pabhvat refusing to take a palm-
leaf fan from Kusa.
20 Amaravati, Archaeological Site Museum, No. 61, illus.: Roy 1994, Pl. 145; Gupta 2008, Pl. 9.
21 Mackenzie drawing of August 1817, British Library, No. WD1061, folio 74, illus.: Fergusson 1868, Pl. 66.
22 Ed. Vol. 5, p. 310; transl. p. 163: Mounted on back of elephant, the queen behind her lord, Kusa descending

to the fray with voice of lion roared. All beasts, when Kusas lion-voice thus roaring loud they hear, And warrior
kings flee from the field, smitten with panic fear. Life-guardsmen, soldiers, horse and foot, with any a charioteer,
242 MONIKA ZIN

jewel (mai verocana). The couple leaving the city on an elephant, the flying god
bringing the necklace and the scene of the battle must be recognized as the iconographic
features of the narrative on king Kusa.
Such a small deviation between text and depiction, as between jewel and necklace,
can actually be understood as the artists liberty in adapting the story. For, the gem
would not be noticeable in the relief.23

Unfortunately it is very seldom that a comparison of the reliefs yields positive results,
and there are a large number of reliefs from the Amaravati School, also those in which
elements repeatedly appear that must be iconographic features of the narratives, that are
still not identified.
Let us go back to our relief with the peacock. This relief too may be compared with
other depictions.
The first representation is small and does not say much about the content of the
narrative. The depiction is placed in one of the registers on the drum-slab of a stpa
(Fig. 7).24 The scene shows a court, with a peacock standing in front of the throne.
The entire composition of the scene is different, as shown in Fig. 1. Since the relief is
known today only from the old drawing, it is not possible to ascertain if further details
were represented here and perhaps not recognized by the draftsman. One thing is certain,
namely that it is also not the Morajtaka, since the bird also stands on the floor, i.e. he
is not instructing the king. The representation provides no further details, yet it provides
an important piece of evidence that the illustrated narrative was well known when it was
placed in the depiction of the depiction as a tiny register on the stpa slab.
The second representation which can be taken for comparison is more interesting even
though it survived only in the old drawing (Fig. 8)25. The piece is a part of the railing
pillar which, at the time when the drawing was made, only had its lower section preserved,
this ending beneath the middle of the central medallion. The peacock is represented in
the medallion; he is shown surrounded by a group of women and children and, as he is
standing on the ground here too, he is not preaching but shown as receiving great respect.
There is a bowl of food placed in front of him. The scene with the peacock appears in
the lowest part of the medallion which mainly consists of the representation of a court
scene. Still recognizable are the lower parts of two persons sitting on chairs awoman
on the viewers left, a man on the right. The persons are seated so far away from each
other that it can be taken for granted that somebody or something was represented between

At Kusas voice break up and flee, all paralysed with fear. Sakka right glad at heart looked on in forefront of the
fight, And to king Kusa gave a gem, Verocana twas hight.
23 The relief corresponds quite precisely with the Pali gths and not with the versions in northern Buddhism

(for references cf. Panglung 1982, p. 38); yet only from these versions is it possible to understand the significance
of the gem: Lord Indra gave it to Kua to make him handsome (e.g. Schiefner 1906, p. 28; Schmidt 1843, p. 95).
Kusas ugliness is the main motif of the story; it is also referred to in the Pali gths.
24 Mackenzie drawing of September 1817, British Library, No. WD1061, folio 78, illus.: Burgess 1887, Pl.37.2.
25 Mackenzie drawing of March 1817, British Library, No. WD1061, folio 43, illus.: Burgess 1887, Pl. 15.2.
WHEN STONES ARE ALL THAT SURVIVED: THE CASE OF BUDDHISM IN ANDHRA 243

them, which placed as it was in the centre of the medallion must have once formed
the essential part of the scene.
In most of the pillars from the Amaravati railings, the same narrative was represented
in the medallion and in the fluted areas above and below; this may also be presumed for
our piece. The preserved lower section divided into three compartments is characteristic
of such pieces and designed more or less symmetrically since there are persons lying
on beds in both the outer registers. The scene on the viewers left shows a lady lying
on the bed. In front of her is a long object which the artist had drawn without really
understanding what it was. Around the bed, there are women sleeping on the floor, and
above, a female deity is shown flying in the air. The goddess seems to carry something
in her right hand.
The scene on the right is even more unusual: here the person on the bed is holding
an infant, but the person, unmistakably male, and is surrounded by the usual royal court.26
Below, on the left, a lady is sitting on a chair and in front of her another woman is
kneeling holding a bowl in her hands.
In the middle compartment there appear again the turbaned man, several women,
the courtiers and the infant. The scene is set inside an enclosure surrounded by a wall,
probably representing a city or a palace area; trees grow outside the walls. The man,
accompanied by two male and one female servant, stands in front of a building on whose
balcony three women appear, the middle one holding the infant.

The peacock represented at the royal court in all three reliefs allows us to assume
that they all illustrate one and the same narrative. In Fig. 1 and Fig 8, we find not
only the peacock appearing but also a reoccurrence of other elements, first of all the
man holding the infant which is very unusual. One common element of the depiction
is apparently also the woman kneeling in front of another holding a bowl, since this
group has been repeated in both representations. It might also be significant that in both
cases the group is placed beneath the man holding the infant (in Fig. 8) or the child (?)
on whose foot someone operates (Fig. 1). That not all elements of one depiction can
be recognized in the other might be the result of the fact that both representations are
incomplete. Perhaps it was the monk (?) from Fig. 1 who was represented in the middle
of the medallion in Fig. 8?
As long as a text is not found, it will be impossible to interpret the depictions.
Nevertheless, we may be sure that the narrative is not the Morajtaka. The reliefs cannot

26 Burgess (1887: 41) who was not able to identify the scene writes thus: What scene in Buddhist legend this
is intended to represent I do not recognize. Is it a case of couvade? This curious practice is in vogue among the
Erukalavalu or people of the Erukala or Yerukala tribe, that wanders about the Kish, Godvar, and Nelur
districts to the present day []. When a child is about to be born, the husband goes to bed, and as soon as the
infant is born it is at once placed beside him. But this does not explain the other two connected scenes. [] Of
the central disc [] the object engaging their attention is a peacock, to which the two principal figures in the
foreground, attended by a chauri-bearer, are paying marked attention. This does not suggest any connection with
the story in the Nachcha Jtaka, and perhaps without the complete scene it will be difficult to identify it.
244 MONIKA ZIN

be adequately identified on the basis of just one element here: the peacock besides,
differences between the text and the relief cannot be interpreted as the artists liberty
of adapting the story.
It is apparent through several examples similar to the Mora Jtaka that the problem
is much deeper and concerns Andhra Buddhism as a whole, from whose flourishing
Buddhist culture not even a single text is available today. The study of the pictorial
representations must play a crucial role in research.
In the forthcoming publication Narrative Reliefs of the Amaravati School, all the reliefs
with unidentified content will be brought together and released for further research
in the hope that one day, a related text or representation from another region of South,
Southeast or Central Asia will be found to explain their meaning.

The narrative reliefs of the Amaravati School are of the greatest importance, not just
for art historians alone. They are the repositories of lost literature for, today, it is only
the stones that survive.

Bibliography

Jtaka, Fausbll, V. (ed.). The J together with its Commentary, being Tales of the Anterior Births
of Gotama Buddha, 1-6, PTS, London, 18771796; transl. Cowell, E.B. (ed.). The J or Stories of the
Buddhas Former Births, Translated from the Pli by Various Hands, 1-6, PTS, Cambridge, 18951907.

***

Bachhofer, L. 1929. Early Indian Sculpture, MnchenNew York, 1929 (repr. 1972).
Bareau, A. 1955. Les sectes bouddhiques du Petit Vhicule, Saigon.
Barrett, D. 1954. Sculptures from Amaravati in the British Museum, London.
Burgess, J. 1887. The Buddhist Stupas of Amaravati and Jaggayyapeta in the Krishna District, Madras
Presidency, Surveyed in 1882 = Archaeological Survey of Southern India, 1, London (repr. Varanasi 1970).
Fergusson, J. 1868. Tree and Serpent Worship: or Illustrations of Mythology and Art in India in the
First and Forth Centuries after Christ from the Sculptures of the Buddhist Topes at Sanchi and Amaravati,
London (repr. Delhi 1971).
Foucher, A. 1928. Les sculpture dAmarvati. In: Revue des Arts Asiatiques, 5, Paris, pp. 924
(Confrence au Muse Guimet en mars 1914).
Gupta, S.S. 2008. Sculptures and Antiquities in the Archaeological Museum, Amarvat, Delhi.
Knox, R. 1992. Amaravati, Buddhist Sculpture from the Great Stpa, London.
Lamotte, . 1958. Histoire du Bouddhisme Indien, des origines lre aka = Bibliothque du
Muson, 43, Louvain.
Panglung, J.L. 1981. Die Erzhlstoffe des Mlasarvstivda-Vinaya, Analysiert auf Grund der
Tibetischen bersetzung = Studia Philologica Buddhica, Monograph Series, 3, Tokyo.
Parimoo, R. 1982. Life of Buddha in Indian Sculpture. An Iconological Analysis, New Delhi.
Parimoo, R. 1995. On Re-identification of ndhra Buddhist Jtaka Relief Sculptures. In: Artibus
Asiae, 55, Zrich, pp. 125154.
Rao, R.P.R. 1984. Andhra Sculpture, Hyderabad.
Roy, A. 1994. Amarvat Stpa, Delhi.
WHEN STONES ARE ALL THAT SURVIVED: THE CASE OF BUDDHISM IN ANDHRA 245

Schiefner, A. 1906. Tibetan Tales derived from Indian Sources. Translated from the Tibetan of the
Kah-Gyur, transl. from German by W.R.S. Ralston, London.
Schlingloff, D. 2000. Ajanta Handbuch der Malereien / Handbook of the Paintings 1. Erzhlende
Wandmalereien / Narrative Wall-paintings, Wiesbaden.
Schmidt, I.J. 1843. Dsanglun oder Der Weise und der Thor, 1-2, St. PetersburgLeipzig.
Sivaramamurti, C. 1942. Amaravati Sculptures in the Madras Government Museum = Bulletin of the
Madras Government Museum, N.S. 4; Madras (repr. 1956, 1977, 2000).
Shimada, Akira. 2006. The Great Railing at Amaravati: An Architectural and Archaeological
Reconstruction. In: Artibus Asiae, 66.1, Zrich, pp. 89142.
Stern, P.; Bnisti, M. 1961. volution du style indien dAmarvat = Publications du Muse Guimet,
Recherches et documents dart et darchologie, 7, Paris.
Sugimoto, T. 2001. Jtakas from Amaravati. In: Kanazawa-daigaku Bungakubu Ronsh, Kd-kagaku-
Tetugaku hen [Studies and Essays, Behavioral Sciences and Philosophy, Faculty of Letters, Kanazawa
University] 21, pp. 9747 (in Japanese).
Zin, M. 2004. The Mkapagu Story in the Madras Government Museum: The Problem of the
Textual Affiliations of the Narrative Reliefs in Amaravati and Nagarjunikonda. In: Annali dell Istituto
Universitario Orientale, 64, Napoli, pp. 157180.
Zin, M. 2006. Mitleid und Wunderkraft. Schwierige Bekehrungen und ihre Ikonographie im indischen
Buddhismus, Wiesbaden: Harrassowitz.
Zin, M. (in print). Nagarjunakonda its Monasteries and their School Affiliations. In: Proceedings
of the Conference on Buddhism across Asia: Networks of Material, Intellectual and Cultural Exchange.
Singapore, February 2009.
246 MONIKA ZIN

Fig. 1. Chennai Government Museum, No. 158, Wojtek Oczkowski


WHEN STONES ARE ALL THAT SURVIVED: THE CASE OF BUDDHISM IN ANDHRA 247

Fig. 2. Chennai Government Museum, No. 105, 3rd register, Wojtek Oczkowski
248 MONIKA ZIN

Fig. 3a

Fig. 3b

Fig. 3. British Museum, No. 69, after Knox, 1992, No. 68, p. 130
WHEN STONES ARE ALL THAT SURVIVED: THE CASE OF BUDDHISM IN ANDHRA 249

Fig. 4. British Museum, No. 37, after Knox, 1992, No. 15, p. 67
250 MONIKA ZIN

Fig. 5. Amaravati, Archaeological Site Museum, No. 61, Wojtek Oczkowski


WHEN STONES ARE ALL THAT SURVIVED: THE CASE OF BUDDHISM IN ANDHRA 251

Fig. 6. Mackenzie drawing of August 1817, after Fergusson, 1868, Pl. 66


252 MONIKA ZIN

Fig. 7. Mackenzie drawing of September 1817, after Burgess, 1887, Pl. 37.2
WHEN STONES ARE ALL THAT SURVIVED: THE CASE OF BUDDHISM IN ANDHRA 253

Fig. 8. Mackenzie drawing of March 1817, after Burgess, 1887, Pl. 15.2
R O C Z N I K O R I E N T A L I S T Y C Z N Y, T. LXV, Z. 1, 2012, (s. 255)

Lista autorw / List of authors

Prof. Jerzy Baczerowski Ph.D., Samuel B. Linde College of Modern Languages, Pozna
Agata Bareja-Starzyska Ph.D., University of Warsaw
Byambaa Ragchaa M.A., University of Warsaw
Michael Farris M.A., Adam Mickiewicz University, Pozna
Prof. Eli Franco Ph.D., University of Leipzig
Szymon Grzelak Ph.D., Adam Mickiewicz University, Pozna
Prof. Joanna Jurewicz Ph.D., University of Warsaw
Prof. Karnina Kollmar-Paulenz Ph.D., University of Bern
Norbert Kordek Ph.D., Adam Mickiewicz University, Pozna
Jowita Kramer Ph.D., University of Oxford
Prof. Marek Mejor Ph.D., University of Warsaw
Prof. Mikoaj Melanowicz Ph.D., University of Warsaw
Prof. Nina Pawlak Ph.D., University of Warsaw
Prof. Klaus Sagaster Ph.D., University of Bonn
Prof. Tadeusz Skorupski Ph.D., University of London
Prof. Elliot Sperling Ph.D., Indiana University, Bloomington
Prof. Piotr Taracha Ph.D., University of Warsaw
Dr hab. Jerzy Tulisow, University of Warsaw
T L nh Ph.D., Hanoi University; Adam Mickiewicz University, Pozna
Prof. Vladimir I. Uspensky Ph.D., University of St. Petersburg
Prof. Monika Zin Ph.D., University of Ludwig-Maximilians, Munich; Freie Universitt,
Berlin
Instrukcja dla Autorw Rocznika Orientalistycznego

Kady nadesany materia zostanie poddany recenzji, kwalifikujcej tekst do publikacji.


Zarwno nazwiska recenzentw, jak i rezultaty ich pracy s danymi poufnymi Redakcji.
RO prosi o nadesanie wraz z publikacj wszystkich niestandardowych czcionek
wniej uytych oraz, pliku w formacie *.rtf lub *.doc, a w uzasadnionych przypadkach
take jako *.pdf.
Przyjmujemy teksty w jzykach angielskim, niemieckim, francuskim i rosyjskim.
Cao tekstu powinna skada si z nagwka tytuowego, tekstu gwnego, sw
kluczowych (co najmniej 45.) i abstraktu w j. angielskim (od 150 do 200 sw).
Autor zapewnia, e posiada stosowne zezwolenia na reprodukowanie w swoim artykule
cudzych materiaw. Autor owiadcza, e nadesany tekst nie by wczeniej w adnej
formie opublikowany. Redakcja jednoczenie ostrzega, e zarwno ghostwriting,
jak i guest authorship s przejawem nierzetelnoci naukowej, a wszelkie wykryte
przypadki bd demaskowane, wcznie z powiadomieniem odpowiednich podmiotw.
Zgaszajc artyku do Rocznika Orientalistycznego Autor jednoznacznie wyraa
zgod na udzielenie autorskiej licencji eksploatacyjnej do nadesanego tekstu
(dot. postaci drukowanej i/lub elektronicznej), co obejmuje prawo Redakcji do:
kopiowania, publikacji, reprodukcji, cytowania, umieszczenia w formie elektronicznej
w bazach danych, itp. Czas trwania udzielonej licencji jest zgodny z dyrektyw UE.
Licencja zostaje udzielona RO nieodpatnie.
Autor otrzymuje 2 egz. numeru RO w przypadku artykuu i 1 egz. w przypadku
recenzji.
Redakcja zastrzega sobie prawo do skracania i redagowania zoonych tekstw.
Po przyjciu tekstu do druku i zredagowaniu Autor otrzyma drog elektroniczn
tekst do korekty, ktr jest zobowizany uczyni i odesa do Redakcji w moliwie
najszybszym terminie.
Instructions for Authors of Rocznik Orientalistyczny

Submitted papers will be subject to peer review by appropriate referees. The names
of the reviewers and the results of their work are confidential data.
The Publisher accepts complete scripts of the Work in *.rtf, *.doc or (in justified
cases) in *.pdf format. All the special fonts used by the Author should be sent as an
attachment to the text.
We accept articles written in English, German, French and Russian.
Every text should consist of a title, main body text, keywords (45 at least), and
a synopsis in English (150 to 200 words).
By submitting the text, the Author warrants to the Publisher:
1. That the Author is the sole author of the Work (both ghostwriting and guest
authorship) are manifestations of scientific misconduct. All detected cases will
be exposed and appropriate authorities will be informed;
2. That the Author holds the full power and authority to grant these rights;
3. That the Work has not been published in any form with any company or person
that may still own proprietary rights to the Work.
By submitting the text, the Author grants, assigns, and transfers to the Publisher,
during the full term of copyright and all renewals thereof, the sole and exclusive
right to print, publish, distribute, market and sell the Work in any and all editions
and formats throughout the World. The assignment is granted free of charge to RO
and shall be effective as long as it complies with the EU directives.
The Author shall receive two free copies of the RO issue in which the Work
appears (the Author of a review shall receive one free copy).
The Editor reserves the right to copy, edit and proof texts accepted for the publication.
After copy editing and final proofing, the text, in an electronic format, shall be sent
to the Author for his approval. After revisions and clarifications (if necessary) the
text must be submitted to the Editor as soon as possible.

You might also like