Buchbesprechungen Comptes Rendus

Author(s): Johannes Bronkhorst
Publisher: Johannes Bronkhorst
Catalog link: https://jainqq.org/explore/269481/1

Page #1
1210 BUCHBESPRECHUNGEN/COMPTES RENDUS BUCHBESPRECHUNGEN/COMPTES RENDUS 1211 zu verdanken, dass Japan ohne aufreibenden Burgerkrieg zu einer neuen Regierungsform fand und den Sprung in die Neuzeit schaffte. Shibas Werk bewegt sich ausschliesslich im Rahmen des oberen Schwert- und Hofadels. Dieses Milieu und die auch in der Ubersetzung nachvollzogene gehobene Sprache macht die Lekture zwar anspruchsvoll, verleiht ihr aber einen besonderen Reiz, den sich der historisch interessierte westliche Leser nicht entgehen lassen sollte. Im Gegensatz zum deutschen Sprachraum erfreuen sich in Japan historische Stoffe in Literatur, Film und Fernsehen grosser Beliebtheit. Das Staatliche Japanische Fernsehen NHK sendet jedes Jahr eine sich uber zwolf Monate hinwegziehende Serie zu einem historischen Thema. Die Folge von 1998 tragt den Titel "Tokugawa Yoshinobu", nicht zuletzt in Gedenken an den 1996 verstorbenen Schriftsteller Shiba Ryotaro. Das interessante Nachwort Eduard Klopfensteins behandelt die Person Shibas, seine Schreibhaltung und Bedeutung in der japanischen Literaturgeschichte und Gesellschaft. Es bietet zudem einen erleichterten Zugang zum vorliegenden, aus einem fremden Kulturkreis stammenden Werk. Um den Genuss der Lekture zu steigern, empfiehlt es sich, das Nachwort zuerst zu lesen. includes the items Parifiksa Savyakhya and Sarva sammata-siksa Savyakhya. The book under review illustrates that Wezler's hope has, to at least some extent, been fulfilled and that some of the work that Aithal planned to do has been taken up by someone else, Ralf Stautzebach (RS). It will not cause surprise that RS has prepared this book at the University of Heidelberg, where it has been accepted as dissertation in 1993; the University of Heidelberg is the institution with which also the author of Veda-Laksana is associated As indicated in the title, the book under review deals with two different Siksas of the Taittiriya-Sakha. It further contains a short general introduction and an appendix about present-day Taittiriya-recitation in Tamil Nadu. The present review will concentrate on the discussion, edition and explanation of the Parisiksa. It goes without saying that the book under review leans heavily on Aithal's Veda-Laksana, sometimes to the extent of being rather unintelligible without it. Consider, for example, the ms-basis on which the edition of the Parisiksa and of its commentary Yajusabhusana has been prepared. In the relevant section "Zur Texterstellung" we read (p. 13): "Der im fol. genden wiedergegebene Text der (Parisiksa) mit dem Kommentar (Yaju. sabhusana) grundet sich bis auf [Parisiksa) 265-84 auf einer Devanagarikopie des Grantha-Ms. MD 924 in (Sanskrit Texts on Phonetics (Lokesh Chandra 1981)] 317-94. Es ist trotz mehrfachen Bemuhungen von Herrn Dr. Aithal nicht gelungen, anderer Mss.-Kopien zu dieser Siksa habhaft zu werden." This manuscript, then, contains both text and commentary. Three other mss. are mentioned, which are stated to agree largely with the one used by RS. None of them contains verses 265-284. These verses figure nonetheless in the edition. Where do they come from? The following remark is meant to provide the answer (p. 13): "In dieser Hinsicht gibt der Schluss des Hamburger Ms. eine vollstandige Erganzung, wenn auch der letzte Vers nicht abschliesst." None of this is very clear. until one looks up Parisiksa in Aithal's Veda-Laksana (p. 429-432), where not only various mss of Parisiksa and Yajusabhasana (or both) are men. Ursula Koike-Good STAUTZEBACH, Ralf. Parifiksa and Sarvasarrumarafiksa. Rechtlautlehren der Taittiriya-Sakha. Stuttgart: Franz Steiner. 1994. (Beitrage zur Sudasienforschung, Sudasien-Institut, Universitat Heidelberg, Bd. 163.) VI + 419 pp. In the Foreword to Veda-Laksana: Vedic Ancillary Literature, 4 Descriptive Bibliography compiled by K. Parameswara Aithal (Franz Steiner, Stuttgart, 1991), A. Wezler recalled that the Vedic ancillary texts known by the name Veda-Laksana "have been virtually forgotten since about 40 years". Later on in the same Foreword he expressed the hope that Aithal's book "will fulfil its true purpose as a mighty incentive to resume the editorial and similar scholarly activities in this highly interesting field of traditional Indian learning". Aithal himself provided towards the end of his Introduction (p. 20), a list of Siksas which he intended to edit. This list

Page #2
1212 BUCHBESPRECHUNGEN/COMPTES RENDUS tioned and briefly described, but also the concluding verses in the Hamburg ms quoted. 1 Also elsewhere the clarity of presentation leaves to be desired. There can of course be no doubt that the Siksas constitute a highly specialized area of research, access to which is not easy for an outsider. But this can be no reason to make the book which tries to provide such access itself inaccessible. Unfortunately there is no other way to describe the book under review. Texts unknown to all but a few readers, even there where they are introduced for the first time, are referred to with the help (?) of obscure abbreviations. The "Einleitung" contains, for example, the following information: "Bei der Bearbeitung der pars und ssf konnte ich an folgende Publikationen anknupfen: tpr mit den Kommentaren tbhr, vaid und mah...[:] vyas...[:] ks...[:] kauns ...[:] vass...[:] bhvs... Weiterhin ... su ...[:] sams...] kns[:] [d]ie ars[:] [d]ie siddhs..." The list of abbreviations is found at the end of the book (pp. 415-16) and makes itself abundant use of abbreviations. The notes, they too full of abbreviations, are also at the end of the book (pp. 275-409). The result is that, in order to read even the least problematic passage of the book under review, one needs to permanently keep at least two fingers on other pages. This might easily have been avoided. The lack of effort to make the text accessible to a non-initiated public is a general feature of the book. This is to be regretted, for the very neglect into which this type of text had fallen calls for a work that introduces readers not accustomed to this kind of literature. More could have been done to make the work under review fulfill that role. Questions relating to the Parisiksa are discussed in a short introduction to the text. Here I will take up one of those questions, the one whether the author of the Parisiksa also wrote its commentary Yajusabhusana or not. RS dedicates less than a page to it and does not come to a clear conclusion. I will show that much more could be said about it, and that a very probable answer can be reached. 1 The Hamburg ms is "Hamburg (Staats- und Universitats-Bibliothek) (cod. Palmbl. III 8/133)" and is described separately on p. 549-550 of Aithal's Veda-Laksana (item 1195: Veda-Laksana (HB)). Strangely, the Parisiksa is not found among the 39 texts which this codex is here stated to contain. BUCHBESPRECHUNGEN/COMPTES RENDUS 2 1213 The question is taken up in a short section, SS 2.8 on p. 26-27 ("Sind Verfasser von [Parisiksa] und [Yajusabhusana] identisch?"). The question is only dealt with cursorily. Consider the following passage: "Die Identitat geht nicht, wie Varma meint, aus der Einleitung des Kommentars hervor. Mit [Cakra] wird lediglich der Autor eines Lehrwerkes zum varnakrama. benannt, was sich auch auf einen Kommentar beziehen kann." In other words, this introductory verse might identify the author of the Yajusabhusana as being Cakra. This is practically all RS says about the issue. Note here that the preceding introductory verse adds that Cakra's father's name was Rama. With regard to the Hamburg ms, considered above, Aithal's Veda-Laksana states (p. 549): "The Ms. must have been written by or belonged to Cakra, son of Rama Ayyangar (of Uruttiti?), whose writings are found in the codex." It seems likely that the two Cakras are identical. The Hamburg ms, as we have seen, contains the Parisiksa,2 but not the Yajusabhusana. This same ms does however contain commentaries on other Siksas (e.g., the Sarvasammatasiksa-vyakhya). If Cakra the son of Rama had composed both Parisiksa and its commentary Yajusabhusana, it would be hard to explain why he left out the commentary in this case. The situation becomes somewhat more understandable, without as yet becoming fully clear, if we assume that he is the author of the commentary only. In that case the Hamburg ms may be a collection of works he copied, perhaps against payment, and to which he could not, or did not wish to, add his own composition. This agrees with the circumstance that a colophon after Parisiksa 168 explicitly identifies the son of Rama as the author of the Yajusabhusana, a commentary on the Parisiksa (...ramasunuviracite parisiksavyakhyanabhute yajusabhusanakhye ...). Further research-beginning with a detailed inspection of the Hamburg ms-may throw further light on this issue, but RS has not even mentioned it. Internal criteria will have to be considered next. The use of the first or third person in the commentary to refer to the basic text does not help -here as so often-to determine identity or difference of authorship. Both occur, as in udaharisyamah introducing [221], against nirupayati to introduce [124] and following slokas, besides numerous occurrences of aha. See however note 1, above.

Page #3
1214 BUCHBESPRECHUNGEN/COMPTES RENDUS BUCHBESPRECHUNGEN/COMPTES RENDUS 1215 Occasionally RS points to a difference between Parisiks and Yajusabhusana, without discussing its relevance for the question of authorship. Parisiksa 65 defines the place of articulation (sthdna) of consonants as the place where contact takes place. The commentary adds that this definition does not literally apply to fricatives etc., because no contact takes place in their case. RS comments (p. 61): "Die Begrundung trifft nicht auf (die Parifiksa) zu, da (Parisiksa) 76 zu den Frikativen lediglich eine Offnung in der Mitte des Artikulators beschreibt." Does this have implications regarding the authorship question? RS does not raise the question, but one is tempted to interpret this difference as due to different authorship. In this connection it is to be noted that Parisiksa 3 announces an enumeration of sounds (varnasamannaya), but that no such list is given in that text. The Yajusabhusana, on the other hand, does list these sounds, 59 in number, in four verses. This might at first sight be considered an indication that the commentary is an integral part of the Parisiksa. However, it is equally possible to look upon these four verses as belonging to the Parisiksa rather than to the commentary. It is not clear by what criterion RS has relegated them to the commentary. Paninian terminology constitutes the background of the terminology of the Parisiksa. Indeed, the Yajusabhusana speaks of the agreement with the established conventions of grammar etc." (vyakaranadisastrasiddhasanketanusara) as an argument justifying certain expressions (p. 41). Many technical terms introduced in the Parisiksa coincide with those known from grammar. Occasionally a grammatical convention is used without it being introduced in the text. Consider the use of after a short vowel-in ar, i and w-to designate just the short vowel (Parisiksa 18). This convention should have been, but is not, explained in the initial section on technical terms (called paribhasprakarana in the commentary). The expressions ar, it and w are explained in the commentary, as akara, ikara and ukara respectively. Had the authors of the Siksa and of the commentary been one and the same person, one might have expected a definition of this convention. The same is true for the use of the Papinian pratyaharas. Ac, used for the first time in Parisiksa 25, covers all vowels, but nothing in the PariSiksa tells us why. The commentary explains the expression (akaradya ukaraparyanta svara; p. 43), and is clearly aware that it needs explanation Under Parisiksa 27 it similarly explains ac (svara) and hal vyanjana). Had its author been the author of the Siksa, he might then have added the required explanation in the section on technical terms. A strange reading is provided in Parisiksa 43-44 which, even more strangely, seems to be confirmed by the commentary. We read there: nadasya samvrte kanthe vasas tu vivre sari/ hakarah kriyate madhye .../. RS translates/paraphrases "Bei zusammengezogener Stellung im Hals wird Ton, bei geoffneter Hauch und in der Mittelstellung hakara erzeugt". This no doubt gives the intended meaning, but it only translates the Sanskrit if we assume as first word nominative nddah rather than genitive nadasya. The parallel passage in the Taittiriya-Pratisakhya has indeed samvrte kanthe nadah kriyate. Yet the Yajusabhusana introduces this verse with the words: ... nadaday[ah) ... cyante "nadasya" iryadina, thus confirming the reading nadasya. It does not however try to explain this reading, and comments as if the expected nominative were there. Only one conclusion seems possible here: the author of the Yajusabhusana found the incorrect reading nadasya in his ms. (The correct reading may have been nadas tu, nadas ca, or something of the sort.) This in its turn is only possible if the author of the Yajusabhusana was not identical with the author of the Parisiksa. Once again the situation might be further clarified by a detailed inspection of other mss. (A similar situation occurs in Parisiksa 51, but this time without confirmation by the commentary. The reading ekantarasya does not seem to make sense, and the corresponding sutra of the Taittiriya-Pratisakhya (2.25) has ekantaras tu. The Yajusabhusana appears to cite the text as ekantara ini. It would have been appropriate to explain why ekantarasya has been maintained, but RS has not done so.) It is also interesting to see that the term hanu "jaw" is feminine in the Parisiksa, but masculine in the Yajusabhusana (except where the latter cites the former). Cp, hanupasamhstatare in Parisiksa 53, aryupasamhrte hani in 54, nari vyaste hana in 57; against hanu ... atyupasamhrtau ... vivstau etc. in Yajusabhusana 48 (p. 52 1. 11 f.). 3 Surprisingly, the commentary on Parifiksa 135 explains the plural acah as acidayah svarar This should of course be har upasanhiatare, dual being pragthya (Panini 1.1.11). Is this a mistake? 4

Page #4
1216 BUCHBESPRECHUNGEN/COMPTES RENDUS BUCHBESPRECHUNGEN/COMPTES RENDUS 1217 In this context we also have to consider the relationship between Parisiksa 48 and the way it is explained in the Yajusabhusana. The Siksa reads nativyastam avarne harvosthart naryupasamhstam, which RS translates: "Bei den a-Vokalen sind die Kiefer und Lippen nicht zu weit geoffnet und nicht zu sehr angenahert". This line is practically identical to Taittiriya-Pratisakhya 2.12 (avame natyupasamhstam asthahanu narivyostam) but for the fact that the Parisiksa, unlike the Taittiriya-Pratisakhya, is metrical. Both the Yajusabhusana and the commentaries on the TaittiriyaPratisakhya interpret this statement in such a manner that the two adjectives concern different sounds: short a on the one hand, long a and pluta a3 on the other. The Yajusabhusana cites even another part of the Parifiksa to support its interpretations. RS does not comment in any detail on the significance of this apparent difference between text and commentary, and limits himself to saying that the citation is not very convincing. William D. Whitney (1868:55) is more outspoken and concludes that at least in the case of the Taittiriya-Pratisakhya) the commentator appears to go against the text he comments. If we draw the same conclusion in the case of the Parisiksa, one is led to think that its author was different from the author of the Yajusabhusana. What is more, one may then have to consider the possibility that the author of the Parisiksa was not influenced by any of the three surviving commentaries on the Taittiriya-Pratisakhya, with all the chronological consequences this may entail. The case is not however completely waterproof. One might still maintain that a supposedly single author of both Parisiksa and Yajusabhusana wished to imitate the Taittiriya-Pratisakhya and one of its commentaries (the Vaidikabharana). But this alternative would seem to be less convincing than the thesis of double authorship. Parisiksa 167 contains an obscure reading. RS presents it in the form apparently accepted by the commentator: ... niryah dtocyare 'sau kvacid ena orah. The problem lies in the last two words, which in the ms have the form esa orah. Neither reading is clear, but the commentator explains: enta oria ekarena okarena. This leads RS to the paraphrase "der nityakampa (wird) mit a, bisweilen mit e und o (gebildet)". There can however be no doubt that this interpretation does not fit the words of the Siksd, and indeed that the words of the Siksa must here be corrupt. The commentator forces an impossible interpretation on a nonsensical reading, which implies that he is different from the Siksakara. Elsewhere the commentator explains a grammatically incorrect line as being arsa "usage of the seers". Parisiksa 183 concludes with the words: evam ca saptasvarabhakty udahrtah "So sind Beispiele fur die 7 svarabhaktis gezeigt worden". The commentator observes: arra svarabhaktaya iti vaktavye svarabhaktity arseryadi vijneyam. It seems unlikely that the commentator describes his own language as arsa Parisiksa 244-245 express the following: "Ein tonlos unaspirierter Verschlusslaut am pada-Ende wird vor oder s aspiriert". However, "djer Kommentar) fuhrt weiterhin Beispiele zur Aspiration innerhalb eines pada auf: samvathsaram, takhsam. Dem entspricht die Regelfassung in (Sarvasammata-Siksa] 19-20.1".6 The text commented upon is however very explicit about the requirement that this operation can only take place at the boundary between two words: the expression padanta figures twice over, and the following s and are characterized as apadantavartin. It seems certain that commentator and Siksakara did not agree in this matter, and were therefore different people. This short survey shows, I believe, that all passages allow of the possibility that Parisiksi and Yajusabhusana had different authors, and that some more or less force us to draw this conclusion. I have no doubt that RS, if he had taken the trouble to take this issue somewhat more seriously, would have arrived at the same conclusion. As it is, he seems to be inclined to consider the two identical. The hypothesis that the Parisiksa could be older than the commentaries on the Taittiriya-Pratisakhya, discussed above, is not without consequences RS draws in the introduction to the Parisiksa attention to its similarity with that Pratisakhya. He then states (p. 24-25): "In einigen Fallen flossen hierbei Interpretationen ein, die (Tribhasyaratna) und (Vaidikabharana] (i.e., the two commentaries on the Taittiriya-Pratisakhya of that name, JB) gleichermassen entsprechen.... Unter den Auslegungen zum (TaittiriyaPratisakhya) geben insbesondere jene einen Impuls zur Bewertung der 5 A similar remark might have been appropriate under Parisiksa 179, which contains apparently an accusative plural svarabhakaayas (daharisye svarabhakrayas sah). Instead the commentator repeats the phrase without grammatical remarks. RS adds a reference to "wiss 10-1". Since this abbreviation does not occur in the list of abbreviations at the end of his book, this reference remains obscure.

Page #5
1218 BUCHBESPRECHUNGEN/COMPTES RENDUS BUCHBESPRECHUNGEN/COMPTES RENDUS 1219 (Siksa), die nur auf (Vaidikabharana) zuruckzufuhren sind ..." After some examples and remarks RS concludes: "ich (halte) es fur wahrscheinlich, dass zu den Vorlagen der (Parisiksa) ebenfalls [Vaidikabharana) zahlt". Why not assume the opposite, that the Vaidikabharana was influenced by the Parisiksa? The question is discussed, in the usual cursory manner, in the section dealing with the relationship between Parisiksa and Vyasasiksa. We read here (p. 25): "Dass letztere keineswegs der (Parisiksa] folgt, zeigt (Parisiksa) 239-40. Dieser Vers vereinigt bei der Definition eines Augmentes die Darstellung des [Taittiriya-Pratisakhya), der (VyasaSiksa) und des (Vaidikabharana). Die (Parisiksa] kann hier nicht zugleich von (Vyasasiksa) und (Vaidikabharana) ubernommen worden sein, da beide grundsatzlich verschiedene Ansatze vorbringen." This statement is not further explained, neither here nor under Parisiksa 239-40. And indeed, it is not easy to find what part of the Vaidikabharana supposedly exerted an influence here. The most likely candidate, as far as I can see, is the phrase: sa khaly abhinidhana iry ucyate/ abhinidhiyate praksipyata iry abhinidhanah (Shama Sastri & Rangacarya, 1906: 379). Something similar occurs in Parisiksa 240: sa cabhinidhiyate 'trabhinidhana ucyate. But obviously no Sanskrit author needs another text in order to link abhinidhana with abhinidhiyate. It is true that the Yajusabhusana cites the Vaidikabharana, but this proves nothing with regard to the relationship between Parisiksa and Vaidikabharana. Or does RS take it for granted that Parisiksa and Yajusabhusana have the same author? As so often, RS remains vague. der (Yajurveda) sind nur diese (gemeint sind offensichtlich die im Vers genannten) als Prapositionen anzunehmen. ..." In a note (no. 26 on p. 288) RS observes that the commentary here "(Vaidikabharana) und [Tribhasyaratna) sinnwidrig zusammengestellt hat." He then explains: "Nach Vorgabe des Merkspruches haben Prapositionen des Typs gari als ... upasarga zu gelten. Es musste dann aber nach (Parisiksa] 234 anu chandansi zu anu cchandarrisi erweitert werden, was der [Kommentar) ausschliessen will." This does not seem to make sense. The Merkspruch is, apparently, the etymology. But an etymology cannot be looked upon as a definition, nor as having more than approximate validity in the interpretation of a word, in this case upasarga. It is not therefore justified to conclude that passages from the Vaidikabharana and from the Tribhasyaratna have here "sinnwidrig" been combined. Note also that the Yajusabhusana does not hesitate to use the expression upasarga in connection with prepositions that are not included in the above list, such as sam (p. 31). This appears to mean that the term upasarga is only used in connection with the above enumerated list where the Yajurveda is concerned. Everywhere else Paninian terminology is used. The same attitude with regard to etymologies shows itself under Parisiksa 15-16. The commentary contains the following etymological explanation (p. 40): anusvaryate pascardhe svaravad uccaryata iry anusvarah "Weil er in der letzten Halfte (anu) wie ein Vokal (-svaryate) ausgesprochen wird, heisst er anus vara. A note (no. 5 on p. 288) comments: "Dieser Satz kann als Erganzung der anus vara-Definition 228-9.1 angesehen werden." This remark does no harm, if its sole aim is to derive information from the etymology. But the etymology was certainly not intended to be a definition, or a supplement to a definition. Parisiksa 133 explains the expression dhalvata with the help of the verbal form abhisandhiyare. RS comments (p. 89): "Der Name (dhaivata) wird offenbar als derivative veddhi aus einer angesetzten Wz. dhi (aus dha) entwickelt." However, etymology is differentiated in India from grammar, and does not require strict derivations. 7 The Yajusabhusana regularly gives etymologies (nirvacana) of key terms. RS seems to attach more value to these etymologies than they may deserve. This is what one is tempted to conclude from a note added to Parisiksa 1214. These lines assign the name upasarga to parl, a, ni, adhi, abhi, vi, prari, pra, ava and upa. The commentary contains the following two lines, which occur in (have been taken from?) the Vaidikabharana and the Tribhasyaratna respectively: nirvacanam tu garitaya karmapravacanlyataya va padantarair upasriyanta ity upasargah/yajurvedavisaye upasarga etavanta everi mantavyam/ RS explains (p. 39): "Weiterhin gibt der (Kommentar) die Ableitung: Die Prapositionen (upasarga) heissen so, weil sie mit anderen pada-s zusammengebracht werden (upastjyante) mit der Eigenschaft als gati oder als karmapravacaniya'. [The Yajusabhusana) verlasst nun den Bereich grammatischer Argumentation und fahrt fort: 'Im Bereich 7 Cp. Bronkhorst, 1984.

Page #6
1220 BUCHBESPRECHUNGEN/COMPTES RENDUS BUCHBESPRECHUNGEN/COMPTES RENDUS 1221 The Sanskrit text of the Siksa and its commentary is followed by an incomplete, but as a rule reliable paraphrase. Occasionally a literal translation is provided. This, too, is normally reliable, but there are exceptions. Consider the following. The Yajusabhusana (under Parisiksa 39-40) contains the following passage (p. 47 1. 4-7): dvividhah Sabdo nityah karyas ceti/ tatra nityah sarvadesavyapy avyakta ekah Sabdo brahmety abhidhyare/ tasmat karyah Sabda utpadyate/ sa vyaktah kvacitko 'nantabhedaf ca/ tasya varnatmakasyotpartir iha vyakh[yleyatayadhikriyatel. RS translates this as follows: "Der Laut ist zweifach: unverganglich und hervorgebracht. Der unvergangliche Laut durchdringt alle Orte und ist ungeschieden und einzig. Er wird brahma genannt. Aus diesem entsteht der hervorgebrachte Laut. Er ist isoliert, tritt bisweilen in Erscheinung und hat unendlich viele Arten. Er enthalt die Sprachlaute. Seine Entstehung, die einer weiteren Erklarung bedarf, wird zum adhikara erhoben." The German words in italics present cases where the translation leaves to be desired. Avyakta and vyakta do not, in this context, mean "ungeschieden" and "isoliert", but "non-manifest(ed)" and "manifest(ed)" respectively. And the manifested sound does not contain (enthalt) the phonemes, but is made up of them (varnarmaka). 8 The expression (utpartir) vyakhyeyatayadhikriyate, finally, does not mean "Seine Entstehung, die einer weiteren Erklarung bedarf, wird zum adhikara erhoben", but "Its production is made the subject-matter as something that is in need of explanation / as the thing to be explained". This last expression contains the instrumental of an abstract noun (vyakhyeyataya) in connection with the object of a verb. It seems that RS has difficulties with such constructions in general. Under Parisiksa 41-42 he paraphrases pratiniyatataya friyate as "Vernommenwerden durch Zuruckkommen". The correct translation is "it is heard as fixed for each single case"; cp. Filliozat, 1988: p. 82 9 27d. The fact that as a rule no literal translation is provided may account for the fact that at times Sanskrit readings are accepted that are untranslatable. Examples are nadasya and ekantarasya in Parisiksa 43 and 51 respectively, considered above. Also the line idaiddvitiyedrasavahnisamjnah in Parisiksa 83 seems to me hard to construe; the obvious emendation idaiddvitiyedrasa vahnisamjnah would go against the metre. Gakarasya in Parisiksa 229 must be something like gakarah sa, as is confirmed by the commentary. A particularly striking example is svaro na sandhanapade visargah in Parisiks, 198, which must be svaro na sandhau na pade visargah. In all these and similar cases one wonders how RS conceives of the task of editing a text. The notes at the end of the book show that he does not always simply reproduce the manuscript, but on many occasions ne apparently does, even when the result is plainly incorrect, or contrasts with the reading accepted by the commentator (recall that RS considers the commentator as being possibly identical with the Siksakara!). On p. 128 RS characterizes a passage from the commentary as being "leicht korrupt" without specifying what is wrong with it, nor proposing any emendation Parisiksa 241 reads, in RS's edition: rante pare sati tarhy anantai kagau, dvav api cagamau stah. This reading cannot be correct, for various reasons. To begin with, we learn from the commentary that this rule concerns the insertion of an augment k between 1 and 1. The rule in its present form says nothing of the kind, but a simple emendation from ante tortat te (confirmed by the commentary: nakarad iti kim and te takare ... pare sati) solves this problem. However, problems remain. The rule remains metrically chaotic, and still does not express all the commentary ascribes to it. RS could, and should, have recorded this, but he doesn't. Even less does he point out that the rule does fit the upajari metre (characteristic of many of the surrounding verses) if only some additional syllables be provided. The metrical scheme in its present, unsatisfactory, state is: --U-UU-U--u--UU-U-- By adding the three syllables in brackets, this becomes: --U-(-) UU-U--/ -)U--UU-U-- which is a perfect upajati. On the basis of the elements presented in the commentary, but that are missing in the incomplete verse, one can make the following conjecture as to its full form: nar te pare (dhe) sati tarhy anantat/ skramar) kagau dua api agamau stah// 8 Under Parisiksa 41-42 RS paraphrases again warmatmaka as "Sprachlaut enthaltend".

Page #7
1222 BUCHBESPRECHUNGEN/COMPTES RENDUS BUCHBESPRECHUNGEN/COMPTES RENDUS 1223 read 'etdy etan', with y. This may look strange at first sight, but is regular Paninian sandhi for ete etan, elision of y being optional by P. 8.3.19 lopah Sakalyasya. 10 The Taittiriya Pratisakhya prescribes elision of y and v by sutra 10.19, but adds (10.20) that Ukhya disagrees with it. This example shows that the authors of the Parisiksa and of its commentary did not necessarily always know, or accept the reading of the Taittiriya texts known to us. 11 To conclude. With some more attention to details the book under review might have been considerably improved. It is unlikely that someone else will anytime soon edit and interpret the Parisiksa and Sarvasammatasiksa, so the book will, in spite of its shortcomings, become the basis of future studies concerning these Siksas and related issues. In the situation one can only advise its readers to use it with caution. There is of course no guarantee that this is the correct reading (which can be checked, and possibly refuted, with the help of the other mss of the text known to exist), but unlike the text presented by RS, it may well be. A reasonable conjecture might have been made in the case of Parisiksa 249, too. Consider to begin with the first line of the commentary on 1248-249): idaidaya ikarapurva aikarapurvo yo yakaro dvinipau dvitvam na bhajen napadyate .... A note attached to the word dvinipou says: "ms: dvinipau dvi". This allows us to conclude that RS planned to correct this reading-no doubt into dviripam dvir am, which alone makes sense- but somehow forgot to do so. This sentence, thus emended, paraphrases the expressions dvinipam and na bhajer, which one would therefore expect to find in the passage commented upon. They are not there, but the edited version contains a lacuna, which we must consider in some detail. The second half of (249) reads, in the edition: sparsottarasthe... A note gives the ms reading, which is: visargottarasthobhadvirupya//. This cannot, of course, be the correct reading, but it does contain similarities to the missing expressions na bhajet and dvirupam. RS changed the beginning on the basis of its citation in the commentary (usmasparfortarasth(e), where usma occurs at the end of the first half of [249]). If we add na bhajed dvinipam, we arrive at: sparfotarasthe na bhajed dvirupam, which is metrically impeccable, and which makes perfect sense in its context. An interesting case is to be found under Parisiksa 144. The Siksa reads: ivarnakotor yavakarabhave yah svaryate ksalpra udattayoh syd "When there is y or v in the place of udatta i oru, the (resulting) svarita is [called) ksaipra". The commentary raises the question why ivarnakotoh "in the place of i ory" had to be expressed, in the following passage: ivarnakotor iti kim/ eta eran' (TS 6.6.8.3), 'asav adityah' (TS 2.1.2.4), 'samyartah' (TS 1.5.1.1), 'sam vadante' (TS 4.2.6.5). RS paraphrases: "Gegenbeispiele: (a) andere Laute als i oder u tragen den udatta: 'eta etan', 'asav adirydh' etc. ..." He has clearly misunderstood the passage, for the context requires that the examples illustrate cases where there is y or v that do not replace i or u. Three of the four examples fit without difficulty: 'asav adiryah', 'sarrtyartah' and 'sam vadante' all contain y or v that do not replace i or u. What about the first example? It clearly has to Some suggested improvements in the edition: (Note that no systematic search for errors has been made, and that the following enumeration merely lists some of the errors that a superficial reading brought to light.) p. 411.2 p. 431.21 p. 461.11 p. 471. 7 p. 491.5 p. 50 1. 26 p. 54 1.4 p. 65 1.25 p. 651. 27 praripsitasya akha procyamana sadvimati vyakheyatayao nadasya "nupradhanam ekantarasya kantham sthanam kanthasthanam ttttttt praripsitasya (?) akhya procyamane sadvimsatir vyakhyeyataya nadas tu (?) "nupradanam ekantaras tu (?) kanthah sthanam kanthah sthanam 10 The Kasika under this rule gives, among other examples, the contrasting pair a adiryah / asav adiryah, precisely the quotation from the Taittiriya Samhith also given in the Yajusabhsana. (This quotation has not been identified in Wilhelm Rau's Die vedischen Zitate in der Katika Viti (1993).) 11 Note that the counterexample ta enam bhisayari (TS 2.3.11.4) under (196-197) and in [207) shows that here a hiatus (and not y) separates the two vowels 'ae. 9 Besides Idaidaya/h), cited from Parifiksi 248. I have no idea how to understand this form.

Page #8
1224 BUCHBESPRECHUNGEN/COMPTES RENDUS BUCHBESPRECHUNGEN/COMPTES RENDUS 1225 REFERENCES AND ABBREVIATIONS (in these last two cases the correct reading might conceivably be kantha sthanam, in accordance with Taittiriya Pratisakhya 9.1) p. 67 1. 12 vargavatsthanayogo + vargavat sthanayogo p. 67 1. 12 mukhasabdenatra + mukhasabdenatra p. 69 1. 27 uktaprakarane uktaprakarane p. 861. 1 madhyakasya madhyamasya (?) p. 87 1. 23 sabdasyoudattao sabdasyodatta p. 89 1. 19-20 degpradhanyotkarsal: + opradhanyotkarsal p. 94 1. 11 eta etan etay etan p. 99 1. 26 bhihate 'bhihate p. 100 1. 20 sa idhanah sa idhanah p. 109 1. 12 sya dhursam ityo + sya dhursad ityo (?) p. 117 1. 2 svaro na sandhanapade vio + svaro na sandhau na pade vio p. 117 1. 24 svarah svarah p. 120 1. 3 ce p. 123 1. 11 prakarena prakarena p. 128 1. 4 gakarasya gakarah sa (?) p. 134 1. 32-33 nante pare sati tarhy anantat kagau, dvav api cagamau stah - nat te pare (dhe) sati tarhy anantat/ (kramat) kagau dvav api cagamau stah|| p. 136 1. 7 kim artham kim artham kimartham p. 1361. 8 tatrafiks tatra Sikso p. 137 1.23 prathamsparsah prathamasparsah p. 139 1. 4 sparsottarasthe ... + sparsottarasthe (na bhajed dvirupam) p. 139 1. 5 dvirupau dvirupam BRONKHORST, Johannes (1984): "Nirukta, Unadi Sutra, and Astadhyayi." Indo-Iranian Journal 27, 1-15. FILLIOZAT, Pierre-Sylvain (1988): Grammaire sanskrite panineenne. Paris: Picard. RAU, Wilhelm (1993): Die vedischen Zitate in der Kasika Vstri. Stuttgart: Franz Steiner. (AAWL Jg. 1993, Nr. 5.) SHAMA SASTRI, R., and RANGACARYA, K. (1906): The Taittiriya-Prarisakhya. With the commentaries Tribhasyaratna of Somayarya and Vaidikabharana of Gargya Gopala Yajvan, with an English Introduction, and Sanskrit Introduction by K. Rangacarya. Reprint: Motilal Banarsidass, 1985. WHITNEY, William D. (1868): The Taittiriya-Pratisakhya, with its commentary the Tribhashyaratna: Text, Translation and Notes. Reprint: Motilal Banarsidass, Delhi, 1973. ca AAWL Abhandlungen der Akademie der Wissenschaften und der Literatur, Mainz, Geistes- und Sozialwissenschaftliche Klasse Jahrgang Paninian sutra Taittiriya Samhita 11111 Johannes Bronkhorst Other corrections: ++ p. 47 1. 23 Sabdasyodbhava p. 471.34 1.12.12 p. 109 1. 22 hastini p. 132 1. 29 laksyanusara p. 290 1. 35 (n. 7) nicht nicht ++ odbhava 2.4.2 hastini laksyanusara nicht +