Arakawa Shintarō 荒川慎太郎. Seika bun Konggo kyō no kenkyū (西夏文金剛經の研究). - Kyōto: Shoukadoh, 2014

Cover Page

Cite item



Full Text

Arakawa Shintarō 荒川慎太郎, Seika bun Konggo kyō no kenkyū (西夏文金剛經の研究, Kyōto: Shoukadoh, 2014) Recent years have seen some important advances in Tangut studies. Despite persisting difficulties, the Tangut texts, both those translated from Chinese or Tibetan and original compositions, are now generally readable. At the same time, much of the current research continues to concentrate on issues of historical linguistics and views the Tangut language through the prism of other languages. Another aspect of Tangut studies is publication and translation of various texts, predominantly of Buddhist nature. However, the linguistic and philological-cumhistorical approaches are rarely combined in one study. A recent publication by Arakawa Shintarō is one happy exception to this rule. This book successfully combines linguistic and philological approaches and concerns itself with issues of synchronic description, rather than with historical reconstructions. Historical linguistics deals with the Tangut materials from its own perspective, with little or no regard to questions of reading and understanding the texts. As a consequence, as Marc Miyake once commented, we know more about proto-Tangut than we do about the language which is presented to us in the written documents. Thus, despite fundamental achievements in the study of the Tangut phonomorphology in general, the structure of the Tangut verb, Tangut verb agreement and other important matters, current scholarship still lacks a comprehensive synchronic description of the Tangut language. Considering the rise of Tangut studies worldwide, this description has to be empirically based and practically oriented, that is to say, capable of providing clues to the understanding of texts. Obviously, the brief descriptions of the Tangut language produced by Berthold Laufer as early as 1916, and by Nishida Tatsuo and Hwangcherng Gong in more recent years, are of limited value in this respect and can be properly understood only by scholars already familiar with Tangut or by linguists who use these descriptions for reference purposes. That said, the recent publication Seika bun Konggo kyō no kenkyū by Arakawa Shintarō (荒川慎太郎) is a long-awaited step in the right direction. This publication is a valuable new contribution to the field. It is based on a meticulous analysis of an important group of texts that are connected in various ways with the Tangut version of the Vajracchedikāprajñāpāramitāsūtra. Exceeding the promise of its title, Arakawa’s book is a massive piece of work which covers issues beyond the topic specified. The publication also includes reproductions of the texts used in the research and careful transcriptions of them. Over two hundred pages of the book contain actual research which summarizes the author’s longtime efforts in the study of the Tangut language; the second section of the book consists of annotated translations of the Tangut texts fundamental to the research. The texts discussed in the volume include the sūtra itself, the version of it with gāthās by Liang Fu Dashi 梁傅大士, T 2732, and the Tangut version of the so-called “collected edition” (金剛般若經疏論纂要 Jinggang bore jing shulun zuanyao, T1701) by the famous Tang Buddhist Master Guifeng Zongmi (780-841). From a general perspective, this book is a very important achievement in the field of Tangut linguistics and philology which should be welcomed by the scholarly community worldwide. The translation needs to be discussed separately at length. For the moment it will suffice to say that it was made on the basis of clearly defined grammar principles formulated by the author in his research and thus the degree of guesswork and intuition, which still remains in translations from Tangut, is kept to a minimum. The author chose not to provide the Chinese originals for the Tangut texts; this impedes understanding, but is justified as it demonstrates that the author really does translate from the Tangut, and not from Chinese, afterwards disguised as the translation from Tangut. In his research the author articulates the grammar rules on which he bases his translation. This makes the translation reliable and worthy for future reference. My only objection to the translation is that in the reproduction of the actual Tangut text the author did not provide punctuation, however, this is remedied in the translation. The research part consists of several independent chapters devoted to a range of topics, from the textual history to questions of phonology and grammar. Of these, pages 2-66 are devoted to questions of the relationship between various versions of the Tangut translation of the sūtra and the textual corpus “generated” by this fundamental text. The author provides a comprehensive list of the Chinese and Tangut versions of the text discovered in Khara-Khoto and elsewhere, and establishes the relationship between various textual traditions. Arakawa pays special attention to the relationship between the Tangut and Chinese versions of the Jinggang jing zuan and verses by Liang Fu dashi. One reason for the analysis of the Tangut version of the Vajracchedikāprajñāpāramitāsūtra is that this text is one of the main scriptures whose study was mandatory for the Tangut monks (on pp. 62-66 the author provides his translation of the famous excerpt from the Tiansheng Law Code, where the important texts are listed). Through studying the surviving colophons, the author established the connection between the version of the sūtra and the gāthās by Fu dashi and the “schematic commentary” on the sūtra composed by Zongmi, whose Chinese version, as far as I am aware, has survived only within a larger commentary composed by Zixuan (子璿) during the Northern Song. The Tangut version of the text is apparently independent from the one prepared by Zixuan and is therefore indicative of the local peculiarities of Sinitic Buddhism in Xixia. Thus, the Vajracchedikā texts probably belong to the circle of Buddhist writings which demonstrate visible deviations of Sinitic Buddhism in Xixia from the perceived character of Northern Song Buddhism. Such works as the Recorded Saying of Nanyang Huizhong, Huizhong’s Commentary to the Prajñāpārimtahṛdaya and the works of Zongmi devoted to the Contemplation of the Dharma realm (法界觀) in all probability also belong to this circle, which defined the character of Sinitic Buddhism in Xixia. Following the general line of his research, the author traces the textual history of the Tangut version of the text which included the poems composed by Fu dashi (pp. 23- 24). Unfortunately, the author does not specifically discuss the text known as 綃癗腲茫阶監膳姑文維菢蓕瞲其佬粄牡沟屈虥挨吨 (pp. 24-25). This text definitely belongs to the Tibetan dimension of Tangut Buddhist literature, while its author 帘褪较构 (*Sumpa Sangs rgyas, “Sumpa Supreme in the World”) crops up in a variety of Tangut sources. Arakawa successfully brings together almost all the available texts of the Tangut translation of the sūtra and comes to the conclusion that a textual diversity existed which encompassed several versions of the text and tries to trace the origins of this variety. Unfortunately, the system of abbreviations which the author devised to indicate the various versions of the text (VMN, VMR, VPB, VPC, etc.) is overly complicated and I personally have trouble identifying the texts being discussed. Arakawa distinguishes 7 major versions of the text altogether (judging from the arrows in the diagram on p. 57; although he identifies 14 different publications). However, it appears to me from the discussion that the author discriminates not so much between the actual versions of the texts (e.g. “early” and “late,” that is published before or after the major “editing” project initiated some time during the reign of Renxiao), as between different editions. As can be seen from the publications by Nishida Tatsuo, the actual textual history of the Tangut translations is to be found in comparison of the various renderings of the dhārāṇi and in tracing new grammar patterns which replace the older forms in successive versions of the same texts. Although aspects of the textual history of the Tangut translations of the Diamond sūtra are a little vague, the reconstruction of the transmission of the “collected version” (Jingang jing zuan) with the poems by Fu Dashi is presented by the author with great clarity. From the perspective of Buddhist studies, the identification of the textual tradition of the Vajracchedikāprajñāpāramitāsūtra which is traceable to the late Tang Huayan tradition represented by Zongmi and resurrected by Jinshui Jingyuan (晉水凈源) in the Northern Song is well in tenor with previous observations concerning the nature of the Sinitic part of the Tangut Buddhist system: i.e. its connection with the Huayan teaching of the Northern China during the Liao and the Northern Song. The Buddhological part of the work concludes with the translation of the entry on Buddhist texts from the Tiansheng Law Code. It might be suggested here, though, that the Dunhuang materials are less relevant for the study of Tangut Buddhism, while the Liao versions of the scriptures, especially the ones available from the Fangshan stone sūtras, might have been useful in determining the general outline of the textual evolution of the Tangut versions of the texts and in determining the hypothetical source text. The second part of the research is obviously more important for the both the author and the reader: it contains the linguistic considerations and the results of a long-term study of Tangut phonology and grammar. One major advantage of the present publication is that unlike other scholarly works, Arakawa’s is actually based on the reading of large amounts of texts, which permits a systematized set of observations. These texts belong to a homogeneous tradition, thus the validity of grammar principles identified in the study can be attested by their recurrence throughout the set of the texts used in the research. However, the examples in the research section are not limited to the Vajracchedikā texts, thus implying a degree of universality for the interpretations postulated. To me, this approach appears more justified than the selection of individual sentences as examples without any reference to their general context. The first section of the second part is devoted to matters of phonology. This contains Arakawa’s own reconstruction of the Tangut phonetic system; among other things the author formulates the principles behind the Tangut transcription of the Sanskrit dhārāṇi. I find Arakawa’s reconstruction plausible; however, the author does not give his reasons for reconstructing the final nasal -n for some of the Tangut syllables, nor does he account for his reconstruction of the initial f-. For example, Hwangcherng Gong reconstructed the Tangut transcription for the Chinese fan 梵 as xiwã, on the basis of the sound change f>x shared by both Tangut and Northwestern Chinese. The dropping of the final nasal consonant is also established by Hwangcherng Gong as a common development for Tangut and Northwestern Chinese and it has to be accounted for if it is preserved in the reconstruction. The present review uses Hwangcherng Gong’s transcription. The most important part of the study is located on pages 125-192 of the publication and is devoted to the analysis of the Tangut grammar. One major advantage of the approach taken by the author is that he proceeds from the Tangut language itself and not from a comparative perspective. Thus the observations formulated in the study might not only be relevant (or might not be relevant at all) for general questions of Tibeto-Burman linguistics, but also useful for actual reading of the Tangut texts. Although the author lists several publications by scholars who worked on this subject before him, one should appreciate that the problems of the Tangut syntax and grammar were previously considered in relation to the study of morphology, structure of the Tangut verbs etc., whereas Arakawa attempts a systematized presentation of the subject. In many cases previous research failed to notice one obvious fact: Chinese sentences are short, whereas Tangut ones are long; this means that auxiliary words and various markers sometimes (not always) have to take on more than one function. So, rendering these classes of words through their Chinese equivalents will make Tangut texts ungrammatical. The Tangut philologists generally neglected the issues of grammar and assigned Chinese equivalents to the Tangut auxiliary words and markers according to what they thought would represent the primary meanings of the Tangut words. We can hypothesize that Chinese grammar was reanalyzed by Tangut scholars in order to accommodate various Chinese syntactic and grammatical markers to their own language. Thus a proper understanding of Tangut grammar is in a way impeded by direct association between the Tangut and Chinese words. For example, if we were to regard the Tangut rjir2 蔎 as yu 與, this would only be able to account for some aspects of this otherwise multifunctional word. Arakawa’s publication successfully escapes these shortcomings. This is because the approach adopted in his study is more appropriate for the study of the Tangut cases. Below I will try to discuss some issues which I find important in Arakawa’s publication. The examples are taken from the Tangut version of the Jingde Chuandeng lu 景德傳燈錄 (JDCDL), the Tangut translation of the Bodhicittotpādasamādānavidhi by Jitāri (Bodhi), The Forty Banners of Emptiness attributed to Atiśa (40), and other sources. As I see it, these examples might be useful for further research. The author begins his discussion with the presentation of morphological matters and specifies classes of words (p. 130), although he correctly indicates that the differences between, for example, verbs and nouns in Tangut are not always clear and are established on the basis of root vowel alterations, which in turn depend on the accuracy of the phonetic reconstruction. That is to say, although the existence of verb stems in the Tangut language is beyond doubt, one still has to be careful when assigning particular verbs to a specific stem. Discussing verbal nouns, the author specifies those formed by the combination of the verb with mjijr2 腞, lew2 籃, sji2 牡 and specifically discusses the particle njɨ2 弛.The first and the last present no difficulties: the first is analogous to the Tibetan ba and Chinese zhe 者, while the last is a noun quantifier. The usage of lew2 籃 as a nominalizer is discussed at some length, but the example given on p. 133 is not the best: here this auxiliary word can be interpreted in its modal capacity. A better one would be: (1) 籋蔲萅谍帝籃矖矂搓娘, 籋窾綀柴属怖 (JDCDL). “If I say that there is even one Dharma in what I give to the others , I am deceiving the people.” One nominalizer which was omitted from the discussion is probably śjij1 槽, normally treated as a suffix: (2) 舉篘毋笭槽, which translates as the “entry into the two truths.” (3) 筗氨篎缾焦臼, 礌祇谍出螏, 耫槽笒铜 (JDCL, “Zhicheng stepped forward from among the assembly, paid respect to the master and told him everything about his arrival.”). More examples of this sort can easily be supplied. Talking about Tangut cases, the author reproduces the classification once offered by Nishida Tatsuo: genitive, dative, accusative, all marked with jij1 谍, instrumental ŋwu2 嘻, accusative / dative (ɣa2 毋), locative do2 唐, inner u2 科 (marking “inside”), “middle” kha1 袭 (“in between”), “mutual” rjir2 蔎 (which can be translated as both “with” and “from”), “consecutive” bju1 瞭 (“to follow”), directional ljijr2 粮 and comparative dzjọ1 沪. Other than that, there are categories of the “nominative” tja1 落 and “fortified nominative”, 強調 dźjɨ·wji1 庭属, which is normally identified as the marker of ergativity. In my opinion, this division is purely empirical and depends not so much on the analysis of the Tangut texts themselves as on the reinterpretation of the Tangut texts on the basis of the corresponding Chinese originals. Of these, dźjɨ·wji1 庭属 and bju1 瞭 can hypothetically be considered examples of grammaticalization, as well as dzjọ1 沪, which is certainly a meaningful word and not a case marker. Taking the above approach, one has to deal with the problem that the Tangut case markers are not clearly distinguishable: only the nominative and fortified nominative (ergative) possess a degree of exactness, whereas the other markers represent several cases and might also take on the functions of phrase connectors, thus becoming syntactic. That is to say, the instrumental ŋwu2 嘻 (as in (4) 繰嘻秬弛菗葾監籃怖 (Bodhi), “one has to wash [one’s] face with water and make it clean”) can at the same time represent an instrumental / causative relationship between the clauses as in: (5) 订谍融袭疥禑弛篟癌, 蟨拓絧粵, 矖佬蔎锻嘻, 矖毋簕薀簧丑 (“Without striving after worldly acclaim, having abandoned the intention [to acquire] wealth and having established harmony with the Dharma, one thus becomes truly seeking for the Dharma.” 40) As the reader may have noticed, in the above two cases ŋwu2 嘻 operates in the same capacity as the Tibetan kyis / gyis, and in the second also connects the two clauses of the sentence, which is again characteristic for its Tibetan counterpart. The approach to Tangut cases taken by the author is more justified: he proceeds not so much from the formal criteria as from the idea that specific sets of markers demonstrate specific sets of relationships which might exist in a sentence. That is to say, he specifies the category of Core (Nom, Gen and Target (TG, equivalent of the Tibetan “Purposive”), represented by dźjɨ wji1 庭属, jij1 谍, ɣa2 毋) and Local cases (probably derived from the Tibetan “Locative”). The final category is the so called “miscellaneous markers”. This division obviously derives from the Tibetan case system. Arakawa’s adoption of this scheme evolves from an understanding that Tangut syntax and grammar cannot be thoroughly described by means of a precisely defined set of categories. The author proceeds from the idea that there are specific types of relationship between words in a sentence, represented by corresponding markers, which should be interpreted according to the context, thus some markers became multifunctional. If such a method is adopted, it would probably facilitate the understanding of the Tangut texts. The Core cases indicate the relation between Agent and Patient (Subject/ Object); locative cases describe directionality (“to,” “from,” “inside”, “outside”), while the “Miscellaneous” account for all the rest (instrumental, etc.). This gives a more systematized view of the Tangut grammar: such markers as jij1 谍 should not be treated as a marker of two cases and the corresponding subject-object relationship, but as the indication of a specific connection between subject and object (agentpatient) which can be described as Accusative or as Genitive, depending on the nature of the specific sentence: (6) 罶處瞪竛驳箁箁蒼礌谍颊 (The disciples of Shenxiu broadly criticized the southern school”, “disciples” and “school”: Subject- Object, Accusative JDCDL); (7) 礌祇禑: 舊祇菢缾谍矺漓碕瞪纓[…] (The master asked: how is your master instructing the Great Assembly; subject-object Accusative, JDCDL), etc. Using this device, interpretation of seemingly enigmatic Tangut syntactic structures becomes easier. Again, interpreting the Tangut jij1 谍 as the Tibetan kyi / gyi could be helpful in this respect. Also, the construal of dźjɨ·wji1 庭属 becomes relatively straightforward and its interpretation presents no difficulties, regardless of the presence or absence of ergativity in the Tangut. As for the “Target / Purposive case” (目的格 ɣa2 毋) I find Arakawa’s interpretation correct. The term “target” is not quite appropriate here and was probably adopted for lack of a better word: this Tangut marker is known to represent both the directionality “from” (礌毋 “from the beginning”) and “to” (毋[耳]維 “up to”, “until”). However, comparison of it with the actual locative do2 唐 “place” might have been useful. The coincidence of these two aspects in Tangut again indicates a proximity between ɣa2 毋 and the Tibetan la. The “purposive” meaning can be illustrated by the following: (7) 虃蔲亡蘀簕聁妹羏毋紹箾稧渡腞搓, 窾妹羏栏丑籃 (Then, if [among] the Bodhisattvas there those who enjoy strict adherence towards the precepts, then [I] must also take the precepts (modified with the causative suffix phji1 丑, Bodhi). Here the Tibetan original uses locative la. The relation between the locatives can be illustrated by the following (8) 菢亡蘀絧纚毋維, 礠緳肅唐瞭吐籋 (byang chub snying por mchis kyi bar / sangs rgyas la ni skyabs su mchi / “[When] I reach the abode of bodhicitta and position myself with the enlightened ones”; Bodhi. Here the Tangut and Tibetan texts do not match exactly: “sangs rgyas la ni skyabs su mchi” means “to take refuge in all the Buddhas”; the above translation is based on the Tangut). Here the “locative” is rendered through do2 唐 in the second clause, and “purposive” ɣa2 毋 emerges 毋維 which is the translation of kyi bar (up to). The above generally means that Arakawa’s approach to construing the Tangut case markers from the perspective of their actual role in the sentence and not from the point of view of the Chinese equivalents assigned to them by the compilers of the Zhangzhong zhu and Wenhai is justified and corresponds to the case structure of written Tibetan, which is the most closely related language with a rich written tradition. Arakawa further postulates “local cases”, demonstrating temporal / spatial relations and consisting of the words normally identified as postpositions. Some of the postpositions which normally represent spatial relationships, such as kha1 袭, are in fact not limited to spatial meaning, but evolved further to acquire syntactic functions: (5) 焊驳亡蘀絧蔰莀, 玛癦癦袭棍纚堡 ([by] maintaining this bodhicitta, in all times they will be living as in if in a palace = Temporal; Bodhi); (6) 罏玛竀紴癅袭毯, 竀紴竛袭耳誓籋 (“Now I am born into the family of the Buddhas, born among the disciples of the Tathāgata = Spatial); (6) 膳阶菢祇落, 挨技吞聁袭, 佬论帛阶 (“The Great Master Huineng, while (indicative of Huineng’s circumstances) not knowing even one character, how can he understand the truth?” = marks the relationship between the clauses in the sentence; JDCDL). The examples presented above do not challenge Arakawa’s basic conclusions, quite the opposite, they tend to support the author’s idea of defining the Tangut cases on the basis of what they actually represent in the texts and not on the basis of certain assumptions. This approach originally existed in the scholarship, and has now been further developed, refined and sufficiently justified on the basis of abundant textual materials by Dr. Arakawa. By this token, we should welcome his publication as a valuable tool which enhances our understanding of the Tangut language and culture. K. Solonin


About the authors

Kirill Yu. Solonin

Saint-Petersburg State University

Author for correspondence.
Russian Federation


  1. Arakawa Shintarō 荒川慎太郎, Seika bun Konggo kyō no kenkyū (西夏文金剛經の研究, Kyōto: Shoukadoh, 2014)

Copyright (c) 2016 Solonin K.

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

This website uses cookies

You consent to our cookies if you continue to use our website.

About Cookies