वेदे विश्वासटिप्पन्यः

Veda Revelation according to Bhartr-hari

Intro

In the present paper,* I approach Veda revelation principally as a phenomenon or experience, more for the logic of its assumption and conception than for the details of what it reveals. The paper thus supplements Aklujkar 199la summarized in appendix 1 with some changes.

A very short version of this paper was presented on 30 March 1992 at the 202nd meeting of the American Oriental Society held at Cambridge, Massachusetts. It discussed only the TKV passages 1 and 2 specified below in their philological relationship with Nirukta 1.20. It was given the present extended form over the intervening years. My extraordinarily learned friend Professor Albrecht Wezler kindly referred to one such extended form in an article of his published in 2001. That form has undergone many changes in the present version. If any incongruence is noticed between the present version and what Wezler attributes to me in his 2001 article, the attribution in Wezler’s article should be set aside, without any implication of error on his part; my view should be thought of as changed.
In my statements as well as the statements I cite, I italicize only those non-English words which are mentioned (as distinct from used). The titles of book length texts/works, volumes, journals, etc. are italicized only in the bibliography at the end.
The abbreviations I have employed are easy to figure out in the context of their occurrence. Yet, in order to leave no doubt, I have explained them in the bibliography.

I assume that the author of both the kārikās and Vrtti of the Trikāndi or Vākyapadīya is BH. Even if the Vrtti author were to be thought of as a different person, he would be a student of BH, not far removed in time and not expressing any significantly different views.

The main questions I ask of myself are: What could be the nature of the revelation process to which Bhartr-hari gives expression? What world-view does he presuppose as providing the logical basis for the assertion that Veda revelation takes place? Where does the phenomenon fit, if it does, in his philosophy as a whole? Does it agree with the dharmābhivyakti view which he presupposes? What do we learn from the exercise as to the role played by the Veda in the thinking of those who shaped India’s intellectual history?

It is necessary to ask these questions not only to gain as complete an understanding of BH’s philosophy as possible, but also because even a preliminary attempt made to answer the questions will help in making sense of an important but little understood part of India’s history.

BH’s Veda revelation view is principally expressed in two passages that have been interpreted differently (appendix 2). One of them contains a quotation from Yaska’s Nirukta and shares several of its key terms with the quotation. As a consequence, our exploration requires a study of what is quoted from the Nirukta, how the traditional commentators of the Nirukta, Durga and Skanda-Maheśvara, understand the quoted part (appendix 3) and how the Nirukta tradition is related to the Trikāņdī (or Vākyapadīya) tradition as far as the quoted part is concerned (appendix 4).

Finally, the issues addressed at various points in the essay make it necessary that I should take into account the views expressed by modern scholars such as Halbfass, Wezler, Falk and Carpenter and by the translators of the Nirukta and the Trikāndi.

The issue to which the issues to discuss lead $1.1 In recent Indological literature, reference has been made to the astute observation by the late French Indologist, Professor Louis Renou that the traditional Indian recognition of the Veda amounts to tipping one’s hat, the action one engages in when one passes someone respectable or when one sees someone respectable pass by. The suggestion is that the appeal made to the Veda as authority or as the ultimate source of all knowledge frequently amounts to nothing more than a traditional etiquette. Often, the persons making such an appeal or bestowing praise have no direct or logical use of the Veda and little or no personal knowledge of it, even if they happen to be quite knowledgeable in other areas and could justifiably be venerable to Indians for other reasons. Without attempting to determine the precise extent of truth in Renou’s observation as it may apply to different periods and different thinkers of Indian history, I would like to raise the question, “Why so?”! Do we have examples in other

Other questions that occur to me in the context of the Renou proposition are: Is the Christian recognition of the Bible or the Muslim recognition of the Qur’an significantly different in extent or essence? If it is, why would it be different? Has any Indian author made a remark similar to that of Renou, since the tradition of urging people to understand the meaning of the

Veda Revelation according to Bharty-Hari

intellectual, reason-dominated traditions of a text or text-complex being so distant in daily life and yet being so much like a knowledgeable and experienced older relative living nearby who can be called on for help even at odd hours? If not, what thing could be there in the minds of thoroughly or mostly rational ancient Indian thinkers that makes them reserve a certain space for the Veda?

Should the thought spring to my reader’s mind, ‘Oh, but the Jina and the Buddha left no such space or had no particular respect for the Veda,’ let me add that, in my view, the rejection of the Veda in the Jain and Buddhist traditions begins only with certain later philosophers. Earlier, what we have is a rejection of those who misunderstand or misuse the Veda associated institutions such as sacrificial worship and varņa but no rejection of the true vedagu or vedāntagu or of the Veda as a body of literature. On the contrary, we have explicit acceptance of the view that in an earlier period there were true Brahmins and spiritually advanced individuals called rșis, implying that the Veda-based tradition itself was not viewed as an object of criticism.

$1.2 Let me introduce, in another way, the gain anticipated from my question stated above. While doing so, I will also provide a possible Indian analogue to Renou’s ’tipping of the hat.’ One of the outstanding short story authors of Marathi, the late G.A. Kulkarni, has written a rather long story titled “Svāmī.“3 To summarize this story is, in a way, to kill it — to sacrifice the thrill of how it affects one as a reader. Yet the present context dictates that I summarize:

‘A person travelling from one town to another realizes that the next stop of his bus is going to be near the village in which he spent his early childhood years. His mother had mentioned several things about the village over the years. On the spur of the moment, he decides to break his journey and to get down at the next stop to visit the village. As he walks back from a disappointing visit, he misses the last bus that would have enabled him to resume his journey and reach the

Veda — not to be satisfied with acquaintance of its words or sound— has existed in India? With meaning emphasized, it is unlikely that no one would query: ‘What is so great about the Veda?’ In fact, the Cārvāka rejection of the Veda (which must stand for a rejection of all scriptures if the Cārvākas are to be consistent) and the critique by Dharma-kirti etc. of Veda-prāmānya make it almost certain that such a query was made more than once and in frequently studied texts, especially of the materialists and the Buddhists.

?I intend to discuss in a separate publication if the current depiction of the Jain and Buddhist traditions as anti-Vedic is valid.

3 The story was originally published in the Dīpavalī (“Diwali” in common Marathi usage) 1973 issue of the magazine Dīpāvali (probable popular spelling “Deepawali”) published from Mumbai. I am not aware of any European language translation of it. It is reprinted in the following collection of G.A. Kulkarni’s stories: 1977. Pingalāveļa. Mumba.ī [= Bombay]: Popyulara Prakāśana [= Popular Prakashan]. Some libraries may have “Kulakarņī” as the transcription of the author’s last name.

originally intended destination. While he is anxiously thinking at the bus stop about how and where he should spend the night, a mahanta (head of a religious establishment) talks to him and persuades him to accompany him (the mahanta) to the mațha on top of a nearby small hill and to spend the night there, assuring that the matha is well-equipped to handle even a hundred unexpected guests. The man is taken up courteously, asked to take a bath, requested to change his travel-soiled clothes into a clothing appropriate to matha life and given fruit and a glass of warm milk as pre-dinner refreshment. In his conversation with the mahanta during the walk to the matha and the reception inside, the man learns that the matha is a well-furnished and self-sufficient establishment with modern amenities such as electricity and bathroom showers and incorporating some aspects of an asrama, that its head svāmī passed away three months ago and is to be replaced within 108 days, and that the mahanta, who came from a very rich family, would not himself take over as svāmī because people follow his directives without objections and resistance when he tells them that they came from the svāmī. After the refreshment, the mahanta guides the man down several levels of the matha building, constructed along a side of the hill, to show the meditation room of the recently dead svāmī. After the man enters the narrow quarters of the svāmī and sits on the seat of the syāmī, partly at the suggestion of the mahanta and partly to test the truth of a miracle to which the mahanta had referred earlier, he notices that a heavy steel door has closed behind the narrow opening through which he entered the meditation room and that the mahanta is on the other side of the door. In the ensuing frantic conversation, he learns that he is now the new svāmī of the matha and will be spending the rest of his life without any contact whatsoever with the outside world. He has access to the absolutely necessary facilities, his well-prepared meals arrive on time through a shaft, and, if he wishes, the appropriate new matha clothing can be supplied to him through the same shaft. There is no physical discomfort, except for the small space available for movement, but there is no talk, no communication with him even through any written note, book etc. Gradually, his pitying for himself and his family comes to an end. His sense of time and attachment to life disappear. He raises a vine that asserted itself through a crack outside his bath area, guides it through the only duct that brings him fresh air, thinks of it as the new carrier of his life, allows its stem to block the supply of air as it rushes toward the Sun and

abandons his life.’ Kulkarni’s carefully conceived story, full of passages touching upon the fundamental problems of life in a disturbingly direct language that alternates with unusual poetic imagery, could be read as using the concept of svāmitva (“mastery’) in various ways. Its emphasis may not square exactly with the emphasis of the present essay. The remark in it that the mahanta needed the inaccessible svāmī to ensure his own effectiveness and its suggestion that, as our traveler was confined in the remote meditation place and was undergoing a thorough transformation, the matha above was being run in his (the traveler’s) post-initiation name are the aspects we should particularly note. Does the Veda attain svāmihood of traditional India’s intellectual matha in a way similar to Kulkarni’s svāmī? Is it a document like any other that has been lifted out of its historicity and enabled to transcend its ordinariness or limitations by a myth-making institution of pre-modern India’s society? And, no matter how great the benefits of its being at the

Veda Revelation according to Bharty-Hari

highest node may be, is it historically not the case that it began its career as any other text would? Or, is there something more to its positioning than what some ancient managers of the religio-spiritual-philosophical complex engineered? 4 Do the great ancient Indian intellectuals, however uncompromising they may be in their use of the faculty of reason, become powerless before the tradition of paying lip service to the Veda that some public relations experts started?

I hope that my attempt to determine what sort of Veda revelation BH presupposed and why he so presupposed will establish that the standing the Veda enjoys in Indian life is not merely a success story in managing public relations. There were, from time to time, at least a few thinkers in India who had a good rational understanding of why the Veda had to be respected the way it was (probably in addition to several mystics who were willing to accept the Veda as a source of fundamental knowledge or wisdom simply because what they read in it, or part of it, agreed with their own experiences). Even the possibility that such an understanding existed in some form when the Veda began to be arranged in the way we find it now need not be discounted. The excesses of stuyding the Veda only as a primitive document or of stuyding it only for gleaning historical or linguistic information should not be committed. One can, in fact, be nearly certain (a) that it was not so continuously studied simply because it was an ancient document or had a tradition of being preserved with considerable self sacrifice, (b) that it was compiled and preserved because the agents of its compilation and preservation had not only a distinctive but also a sophisticated view of individual and social life and (c) that the same agents also had a specific understanding of how it was to function as a means.”

The first TK passage having a

bearing on Veda revelation $2.1 In a 1991 paper entitled “Bhartr-hari’s concept of the Veda,” I have pointed out the distinctive features of BH’s concept of the Veda and

4 (a) I mention religion, spirituality and philosophy together because, despite the abundance of all three in ancient Indian life, they were not separated from each other with separate names as academics do now.

(b) Some Indologists and historians of India write as if the complex I mention here was like the military-industry-politics complex of certain modern societies — a hegemony that did not allow the will or welfare of the people at large to prevail. I am not sure that such indeed was the case.

5 (a) In making these remarks, I am not denying that the Veda body grew over time. Nor am I discounting the possibility that the criteria for the application of veda could have been different from one time to another or from one region to another.

(b) Some readers may find it convenient to read SS4.1-6 before reading the philological analysis in SS2.1 - 3.6

its relationship with some other elements of his thought. A part of that publication relevant to our present concern is summarized in appendix 1, with some changes and additional clarifications that would connect it and the present one smoothly. Therefore, presupposing knowledge of the contents of appendix 1, I will concentrate below on passages which have a direct bearing on BH’s concept of Veda revelation as a process. These passages are two.

Passage 1: context Trikāņdi (= TK) 1.5: prāpty-upāyo ’nukāraś ca tasya vedo maharşibhiḥ/eko ‘py aneka-vartmeva samāmnātah prthak prthak//6

“The means of reaching and the representative likeness of that (sabda-tattva brahman) is Veda. (It) is set down for transmission’severally by the great seers as if it has more than one path, although it is one.”

Trikāņdi-vrtti (= TKV) 1.5: anukāra iti. ‘yām sūksmām nityām atīndriyām vācam rsayaḥ sākṣāt-krta-dharmāņo mantradrśaḥ paśyanti tām asākṣāt-krta dharmabhyo ‘parebhyaḥ pravedayisyamāņā bilmam samāmananti, svapna-vịttam iva, drsta-śrutānubhūtam ācikhyāsanta ‘ity esa purā-kalpaḥ. āha khalv api. “sākṣāt-kyta-dharmāna rşayo babhūvuḥ, te ‘parebhyo’ sākṣāt-kyta-dharmabhya upadeśena mantrān samprāduḥ. upadeśāya glāyanto ‘pare bilma-grahaņāyemam grantham samāmnāsişur vedam ca vedangāni ca. bilmam bhilmam [>bilvam?) bhāsanam iti ve” ti. [Yāska, Nirukta 1.20].

$2.2 I will offer a translation and an explication of this passage in $82.21-23. First I need to make some preparatory determinations and offer clarifying observations.

According to its presentation above, the passage contains statements of BH as well as Yaska. The former does not name the latter or his Nirukta. The introductory expression āha khalv api can at the most be understood as alerting the reader that the source is different from the Purā-kalpa mentioned immediately before (cf. agamāntaram used by the commentator Vrsabha, p. 25). Secondly, there is no reflection in the non-quotation part of the passage, that is, in the part which owes its content directly to BH, of the quotation’s phrase imam … ca. Theoretically, therefore, the possibility remains that BH cited from a source other than the currently available Nirukta. However, I see no reason to doubt that the passage incorporates Yaska’s that is, the Nirukta author’s, statement. The expressions rsayaḥ sākṣāt kyta-dharmāṇaḥ, asākṣāt-kyta-dharmabhyaḥ, bilma, and sam+h+mnā are too many to be accidentally common, especially given the brevity of the passage

My translations will not be literal to the extent of matching the number of sentences in the original, and I will not always translate the ’loaded’ words like dharma and mantra. If we stop at each occurrence to discuss what they mean, we will get nowhere or get somewhere when it no longer matters whether we got somewhere.

My reasons for translating samāmnāta in this way will be given in a separate article.

8 Commentator Vrsabha mentions specific referents for vartman (‘paths such as Sāma veda, Rg-veda etc.’) and prthak pythak (’through distinction in the form of mantra etc.,’ If the emendation of maunādi to mantrādi is accepted). However, it seems more likely that BH has deliberately used very general words like vartman and prthak prthak; see appendix 1, note to point 5.

Veda Revelation according to Bharty-Hari

and the fact that s-k-d (or its negation), bilma and sam+a+mnā are not the kind of expressions that would be used in widely divergent contexts (as ysi and apara/avara, for example, would be). Further, there are several indications in BH’s works which cumulatively establish that he must have known Yaska’s Nirukta. To look at the evidence from the side of the Nirukta tradition, there are indications in the commentary of S-Mo to the effect that BH was thought of as commenting on the Yāskīya Nirukta in parts of his works (Aklujkar 1994) — as someone who offered observations on certain Nirukta sections from a distance. It is therefore, justified to proceed on the assumption that it is in fact the currently available Nirukta that BH cites in the passage under consideration, although BH could have followed a version or recension of the Nirukta in which apara was the reading in the place of avara (see appendix 4).

$2.3 Two details available in BH’s statement (sūksmāṁ nityām atīndriyām vācam paśyanti and svapna-vịttam iva dysta-śrutānubhūtam ācikhyāsantaḥ) that undoubtedly pertain to the revelation process are not found in Yaska’s statement. Therefore, we need to ask if a historically identical understanding of the revelation process is reflected in TKV 1.5 (and TKV 1.173 to be discussed later) and Nirukta 1.20. Could BH not have differed from Yāska in his view of how the revelation of the Veda took/takes place? Or, does BH simply help us in recovering what has been lost in the other intellectual traditions of India, including perhaps the Nirukta tradition itself? If he did not essentially differ from Yaska, was there any progress in the theorization about revelation such that the statements of the theoreticians concerned came to possess greater detail and clarity, or was there practically the same theory all along that was expressed in slightly different words as time moved on? Alternatively, is it possible that there was no theory as such in Yaska — he simply made an assertion, and it is philosophers like BH that gradually filled his assertion with deep significance and gave it a logical backbone that was not in the purview of the older and (hence?) simple-minded Yāska?

To give my four-part answer right away: (a) It is not possible that Yāska did not have a theory. (b) There is no evidence to prove that Yaska and BH differed, although they might have. (c) A combination of the possibilities we have entertained is what probably happened between the times of Yāska and BH. There could have been addition of detail and gain in clarity. There could also have been changes of terminology. (d) BH does help us, at least to a significant extent if not fully, in recovering, in a historically justifiable way, Yaska’s thought and the thought behind the authority and sanctity accorded

9 For the sake of convenience, I shall speak here of Skanda-svāmin and Maheśvara as if they were one author. I shall not try to decide which one of them wrote which part of the Nirukta commentary going under their name.

to the Veda. The guidance he offers is limited and scattered. Yet if we study it comprehensively and minutely we can open, at least to some extent, the door to a closed chamber of India’s history. By the end of our discussion, we will probably have a relatively well-integrated account and—what is more important in historical research — an account that has an evidential, textual, basis.

I do not need to justify each part of the preceding answer individually or at length. Even if I prove that Yāska had a theory, it should suffice. Appendices 3 and 4 should take care of part (b). My part (c) is only a probability and it follows from historical commensense. There is no evidence of knowledge traditions having been totally or irrevocably lost between Yāska and BH as far as the views or theories regarding the Veda are concerned.10 So, philosophers must have gone on refining ideas and expressing them in different words. Is this not what they most commonly do? Further, the differences between Yaska and BH are all of the nature of addition. Just as they strengthen the probability that the philosophers were elaborating on and re-expressing the views, they indicate that BH need not be read as clashing with Yāska:11 As there is no other theoretician between Yaska and BH known to us at present who echoes Yāska’s words, 12 it is also commonsense to proceed on the assumption that BH would help us in recovering a part of Yaska’s world. Where there is almost total darkness, even a single source of feeble light can be put to use. Our reconstruction of BH’s understanding of how Veda revelation took/takes place will simply demonstrate that this commonsense, like most flashes of commonsense, has put us on the right path.

Now, how can it be proved that Yaska had a theory? One only needs to look at the words sākṣāt-kyta, dharma/dharman, rși and mantra. Each of these is a charged word of Indian culture (cf. SS2.10-16 below). Each embodies a notion that no other known culture has taken to the height to which India has taken it. How could Yāska then be without a theory while

10 Loss of vyākaraṇāgama, ‘inherited knowledge in the area of grammar,’ is mentioned in TK 2.481-485 in the context of Pāṇinian grammar and with specific reference to Patañjali.

11 Halbfass (1991: 48 note 69): “Bhartr-hari cites and accepts the statement from the Nirukta … But at the same time, he transforms and reinterprets Yaska’s dictum.” The details given by Halbfass after this remark are all by way of addition and specification of BH’s view. They do not suggest that Yaska was definitely unaware of them or would have taken issue with them.

12 (a) I have checked as many pre-BH texts as I could for a passage in which the ‘load bearing’ words of the s-k-d passage occur reasonably near each other. No one can assert that we will never come across a post-Yaska and pre-BH passage reminiscent of the first or second sentence of TKV 1.5. But at present we must proceed on the assumption that non-occurrence is the reality.

(b) Post-BH occurrences are found in the Yukti-dipikā and Helā-rāja 3.1.46 (see appendix 4). Allusion to the passage through its opening word exists in Medhātithi 2.12. I am leaving out modern citations and allusions.

Veda Revelation according to Bhartr-Hari

employing sākṣāt-kşta, dharma/dharman, ?și and mantra? Further, what author would bring the entities dharma and mantra together, suggest that access to the first results or can result in access to the second, and leave things there, if he did not presume readers familiar with the connection between dharma and mantra, that is, if he did not presume knowledge of some theory on the part of his readers (cf. appendix 4, point 2)? Is the connection between dharma and mantra, under any common understanding of these terms datable to Yaska’s time, obvious? Of course, not. Even the literal or etymological meanings of these terms are not such as would naturally lead to their being put together in a seriously intended statement. There must, therefore, have been a reasonably widespread undertaking, containing what we could call a theory, in Yaska’s days that enabled most of Yāska’s readers to connect the two even when no explication was offered.13 This surmise is supported by what the pre-Yaska texts, Samhitās and Brāhmaṇas, say about vāc and by how they use the notion of vāc (cf. the passages referred to in Padoux 1990: 1-29, Sāstrī

1972: 1-23, Tripāthī 1976: 167-175 and Holdrege 1994).

$2.4 Another question arising out of the bipartite nature of the TKV 1.5 passage is more specifically hermeneutical. Several different interpretations have been offered of the Nirukta statement. Even if one were to assume that Yaska had a well-thought view of Veda revelation, is it not likely that the relationship of that view with BH’s view would depend on which interpretation one accepts? In other words, will out reconstruction of BH’s view not be uncertain to the extent we use the evidence of Yaska’s statement? (If we, in return, use the reconstructed BH view to understand Yaska, since we commonly use later direct and indirect commentators to interpret an older author, will the extent of uncertainty not double?)

Mutual dependence of sources is not a real problem in the present context, which is one of historical reconstruction, as it would be in an inference (understood in a technical sense of “inference” found in the system of logic, not just as a synonym of general words like “logical reasoning,” “argument,” “conjecture,” “hypothesis” etc.). Frequently, a source produced at time T2 has a more reliable interpretation of or has more clues helpful in the understanding of a source produced at time T1 than we, living at time T3, are likely to have. We have to take both the T1 and T2 sources and the clues they contain into consideration and arrive at the most probable interpretations of the T1 source as well as the T2 source, verifying them against each other.

13 No reader of Yaska is known to have accused him of trying to pull wool over his/her eyes. The readers include Yāska’s commentators, who showed a questioning attitude in countless areas over many centuries. While this consideration, being in the form of an absence, does not prove in itself that Yāska had a theory, it makes the absence of a theory very unlikely.

In determining how the very first stage of Veda revelation was understood, the utility of Nirukta 1.20 is limited. That text is principally concerned with the Veda transmission stage. Only the words sākṣāt-krta dharmana sayaḥ…mantran samprāduh have a bearing on the revelation process in that they tell us who the agents were, what qualification they had and what was revealed (the last by implication: since the agents imparted mantras, they must have come to possess them, which in turn, implies that what was revealed to them must be either the mantras or something that can result in mantras.) The implications of avarebhyo ‘sāksāt-kyta-dharmabhya upadesena would largely be deducible from sākṣāt-krta-dharmāņa rşayaḥ… mantrān samprāduḥand hence of no great help.

Of the truly relevant words, (a) sākṣāt-kyta, (b) dharma/dharman, (c) rsi and (d) mantra are found also in the TKV. While different paraphrases and translations of the first two have been offered, none, as we will see below, is likely to affect our process of determination. In the case of (a), all translations would work at the literal level (albeit specific connotations may have to be coaxed out of their contexts) and, in the case of (b), none would fit literally. There is no real disagreement in the translations of rși (those who translate the word with “sage” do not and will not deny that in the present context sagehood comes from being a seer), and mantra has mostly been left untranslated.

$2.5 Several interesting questions have been raised, explicitly or implicitly, in the case of Nirukta 1.20 that pertain to the words in the first two sentences, to which I just referred, as well as to the words in the remaining two sentences. As far as I am aware, they are: Does s-k-d belong to the subject part (“The s-k-d became/were rșis’) or the predicate part (‘rșis became/ were s-k-d’; see appendix 4, point 2)? Is the ‘avara : apara’ difference of reading significant? Does avara mean only ’later’ or must it have a connotation of inferiority in the present instance? Are the avaras also rșis? If so, are the rşis only in some such sense as ‘śrutarșis? What is the significance, if any, of the tense variation seen in babhūvuh (perfect), on the one hand, and samprāduh and samāmnāsişuh (aorist) on the other? Which meanings of sākṣāt + krand dharma should we accept? What was the nature of the upadeśa? Was it only oral? Did it include meaning explanation or interpretation? Does the text talk about two groups or generations or three? Toward which upadesa was the avaras’ fatigue or despondency directed to the one they were receiving or to the one they were providing? What is the meaning of bilma? Can the meaning one prefers be reconciled with the meanings given in the case of other occurrences of bilma? How is the compound bilma-grahaņa to be analyzed — as an instrumental tat-puruṣa or as a genitive tat-puruṣa? How many texts or text-bodies does imam grantham … vedam ca vedangāni ca refer to - one, two or three?Veda Revelation according to Bharty-Hari

As if this list is not long enough for a passage consisting of only three short sentences (four sentences if vedam ca vedangāni ca is counted as a separate sentence), I will add: What was the precise reason for the fatigue or despondency of the avaras, lack of success in learning the entire set of mantras or lack of success in learning the mantras accurately? How should

sam + ā + mnā be understood, especially in contradistinction to ā + mnā? Does imam grantham refer only to the direct or most explicitly acknowledged commentandum of Yaska, that is, the Nighantu, or does it include in its reference also the inherited lists such as those of the upasargas and the nipātas on which Yaska comments (compare Aklujkar 1999: fn 9)?

From among these questions, I will address the ones having a bearing on my present concern in the main body of the paper. Many of the rest will be taken up in appendices 3 and 4.

$2.6 To turn to preparatory observations not arising from the presence of the Nirukta citation, we should not expect to understand the process involved in coming to know the subtle, permanent and sense-transcending form of vāc (= the original form of the Veda; cf. appendix 1) that BH speaks of or the process making certain rșis possessors of mantras that Yaska presupposes in the sense of “Oh, yes, X/I experienced it, and here is how it went.” A number of passages (e.g., TK/VP 2.139) make it clear that in BH’s view only certain individuals with special qualifications may be said to have the experience and that they too cannot convey it exactly as it is — in its event or process aspect. The original unitary insight cannot be transferred as a single unit. Its transmission or instruction must take place through speech, and speech, as we commonly understand it to be, is necessarily sequential. If those who, in some sense, receive the experience of the rşis receive it in a sequential, divided form (see appendix 1, point 5), our understanding or reconstruction can only be theoretical

$2.7 The assumption of a subtle or more fundamental form that becomes manifest, communicable or accessible should not surprise us. Theories of a primal cause that contain movement from the unperceived to the perceived (or perceptible) exist in practically all accounts of creation of the universe. In fact, in the very admission of the possibility of such a creation, whether unique or repeatable, there may lie the postulation of a subtle, unperceived pre-existence. Also, as I indicated in a 1982 paper on the recovery of vyākaraṇāgama, almost all accounts of getting back a lost fundamental teaching have reference to that teaching’s survival somewhere in a hidden or unidentified form. The very logic of the situation can be said to demand the assumption of a lost yet not-completely-lost original. The vertical double reference of the Veda (appendix 1, point 1) is similarly structured and hence the presence in it of a subtle form of the Veda should not come as a surprise.

$2.8 The process that would emerge from our theoretical understanding or model-building should be such as would fit what we know about dharma/dharman (see $2.11-13) and mantras from other sources. Seeing the highest form of vāc should have some plausible connection with seeing the mantras (cf. mantradrśaḥ in TKV 1.5 that is under consideration). Similarly, witnessing dharma/dharman should mean furnishing the cause necessary for seeing the highest form of vāc and/or seeing or coming to possess the mantras.14 BH had a choice to write a sentence like yam/yām/yad … rsayah sāksāt krta-sūksma-vācah paśyanti tam/ tām/tad asākṣāt-kyta-sūksma-vāgbhyaḥ pravedayisyamāņā bilmam samāmananti. The fact that he takes over sākṣāt-kyta-dharmāṇaḥ from the Nirukta and makes sūkṣmām nityām atīndriyām vācam … paśyanti a predicate (of the relative clause) indicates that in his view sāksāt-krta-dharmānah belongs to a deeper, causal level on which the assertion that the subtle and eternal vāc is witnessed can rest.

$2.9 On the other hand, since ordinary persons do not witness the highest form of vāc, the qualification sākṣāt-kyta-dharmanah must express the difference between them and the ysis. The difference should be such as to be consistent with what we learn from other passages in which comparable persons who exhibit extraordinary powers of cognition (e.g., the sistas ‘spiritual elite who are capable of giving informed judgements free from vested interest’) are mentioned by authors sharing essentially the same world view.

Several relatively early authors like Caraka (Sūtra-sthāna 25.3), Patañjali (Paspasāhnika, Kielhorn’s edn p. I.11), and BH (TKV 1.23) speak of pratyakşa-dharman rșis or śistas in contexts that could be considered to have a logical relation with the statements we are studying. Vātsyāyana Paksila svāmin, on the other hand, in a similarly relatable statement, uses s-k-d.15 We should at least ensure that the sense we eventually attach to s-k-d does not clash with the sense of pratyaksa-dharman, which, given the synonymy

14 (a) Here, my use of ‘seeing’ is meant to cover TKV 1.5 and of ‘coming to possess’ to cover Nirukta 1.20. In the latter, as the s-k-d rșis impart the mantras, they must have come to possess them.

(b) The point stated here does not depend on whether sk-dis taken as a predicate in Yaska’s sentence, for there is a suggestion of the priority of sākṣāt-ksta-dharmatva to mantra in Yaska and BH and there is a suggestion of the priority of sākṣāt-krta-dharmatva to the perception of s-n-a vāc in BH.

15 Under Nyāya-sūtra 2.1.68 (mantrāyur-veda …), we find āptāḥ khalu sākṣāt-kşta-dharmāņa idam hātavyam, idam asya hāni-hetur, idam asyādhigantavyam, idam asyādhigama-hetur iti bhūtāny anukampante. “The reliable persons are s-k-d. They show sympathy for (other) living beings (thinking:) ’this is to be abandoned; this is the cause for abandoning; this is to be obtained by this person; this is the cause for obtaining.” Perhaps this passage is studied in Srinivas Sastri

1976, which publication has remained inaccesible to me.

Veda Revelation according to Bhartr-Hari

of sākṣāt and pratyakşa, means that the meaning of dharma/dharman in one compound must be compatible, if not identical, with that in the other.

$2.10 sākṣāt and sākṣātkāra are still used in Indian languages. This is true also of some other derivates (e.g., sākşya and sāksin) from the root word sākṣa (= sa + akşa/aksi) underlying them (and attested in Sabara’s commentary on Mīmāṁsā-sūtra 1.2.31). Wezler (2001: 226) informs us that sāksāt by itself is attested in the Atharva-veda and Taittiriya Samhitā, that Pāṇini 1.4.74 (sākṣāt-prabhịtīni ca) presupposes the possibility of its composition with forms of krand that sākṣāt-kyta in the Nirukta passage we are discussing is the first known realization of that possibility.

Given such an impressive continuity of use, one would expect the derivates of sākṣāt + krto be transparent in their meaning. To some extent they are. The meaning element of being a winess, being a perceiver is found in all of them. A natural extension of this element would be being a direct witness, perceiving without intervention.‘16 This semantic extension is a constant of the sākṣāt + kr derivates, 17 along with its understandable contextual variations (a) ‘obviously, evidently,’ (b) ‘clearly, openly,’ (c) ‘in bodily form, as something incarnate,’ and (d) ‘intuitive’ in the sense ’natural, inborn, not preceded by any training or practice.‘18

16 Compare a usage like “I saw X with my eyes” which implies that the speaker’s knowledge is not second-hand.

17 (a) Cf. the use of “direct,” and “actually” in the following translations of s-k-d: Halbfass 1988: 328: “having attained a direct experience of dharma.” Falk 1990: 109 and 1993: 241: “persons who had direct insight into dharma.” Ruegg 1994: 308-309: “having directly witnessed/perceived dharma(s) and “who directly perceived dharma(s).” Oliver 1997: 59 (on the basis of Falk 1990: 109): “The rșis had direct insight into dharma.” Kahrs 1998: 28: “[those who have] direct access to Dharma (ritual and social duty).” Rajavade 1940: 289: “by whom religion was actually seen.”

(b) In the context of s-k-dand sāksāt-kyta-dharmatā occurring, respectively, in Nyāya-bhāsya 1.1.7 and 2.1.68., Ruegg (1994: 307 fn 15) suggests an interpretation that has the merit of being original: “having direct perception for [their] nature/quality,” in which “[their]” stands for “[the aptas’ = the reliable persons’].” Since in the specified context Ruegg is very much aware of Nirukta 1.20, the same interpretation can be extended to Nirukta 1.20, the passage we are discussing. However, Ruegg helpfully adds: “… Indian commentators seem to have actually understood dharma(n) as the object of the direct perception in question. See e.g. Uddyotakara, Nyāya-vārttika, II.1.68: sāksāt-krta-dharmatā yam te padārtham upadisanti sa taiḥ sākrāt-krto bhavatīti; and Vācaspati-miśra, Nyāya-vārttika-tātparya-tīkā I.i.7: sudydhena pramāṇenāvadhāritāḥ sāksāt krtāḥ dharmāh padārthāh (and the same commentary on II.1.68: pratyakṣī-krta-heyopādeyatā).”

Further, in his fn 17, referring to Roth’s rendering “Recht” of dharma occurring in Durga’s dissolution of s-k-d and Monier Williams’ and Sarup’s translations “one who has an intuitive perception of duty” and “had direct intuitive insight into duty,” Ruegg rightly registers a mild protest: “But it is not established that dharma(n) here has the meaning of duty or virtue.”

Of the two explanations given by Vācaspati that Ruegg cites above, the first agrees in spirit with Durga’s cited in note 67(b) and the second with Vātsyāyana’s given in note 15.

18 Cf. Sarup 1921: 20 “Seers had direct intuitive insight into duty.” Kane 1973: 889: “the (ancient) sages had an intuitive perception of dharma.”

What intrigues a researcher are the issues of whether the ablative ending of the first member contributes something specific to the meaning of the composition and, if it does, what that contribution is. That there are not many other formations in which kris joined to a preceding member in the ablative makes the researcher’s work difficult. balātkāra ‘doing out of ( with) force, acting on the basis of physical strength, compulsion’ comes to mind, but balātin it is not a compound like sākṣa in sākșat. It is not given the designation gati by Pāṇini, sākșa, being a bahu-vrīhi, has the capability to function like an adjective and, when transformed into the neuter accusative singular (sākşam) not qualifying any noun, to function as an avyayi bhāva conveying an adverbial meaning. However, what we have here before a form of křis neither sākṣa nor sākşam but sākṣāt. Could there be an aspect of meaning present in this grammatical feature that is eluding us?

Could sākṣātkāra, originally, have had a meaning such as doing/acting out of the sense-equipped one, that is, with the mind/worldly self/soul as the basis, attention-done, concentration-accomplished? The implication of such a meaning, then, would be either that ’the usual senses are not required, the ‘mind-eye’ is used, a transcendence of the ordinary senses takes place’ or ‘what others do not or cannot see/sense is seen’? After all, specificities of meaning do frequently come from the context in which an expression is initially used, not only from a logical or plausibly logical extension of the etymological meaning. The two connotations specified just now are present in some modern usages of sākṣāt-kr; e.g., Marathi sākṣātkāra hone becoming (= coming into existence) of sākṣātkār,’ includes the meaning element that which was not seen or known before, now became seen or known.‘19 Several other modern Indian languages use forms of sāksāt-kr similarly. As many modern Indian languages form continuities of usage with Sanskrit and can be shown to preserve some word connotations that have so far not been recognized as existing in Sanskrit,20 our seeking a clue in them for the connotation of sākṣāt-kr should not be objectionable.

19 When used seriously in ādhyātmika etc. contexts, sākṣātkār hoạe usually implies either that a great effort preceded the happening or that the person in whose case the event took place no longer remained an ordinary individual — his very way of experiencing the world changed. When the word is used sarcastically, the same meaning acquires the tinge ’taking a long time to grasp something that was already there — that was not difficult to grasp in the first place, what appears like a sudden realization should have taken place much earlier.’ The suggestion is of a lapse or of going unnoticed before being seen. It would fit the context of the passages we are studying, especially if they presume recurrent creation. The seers are spoken of as recalling what they learned or knew in the preceding creation.

20 Such shared connotations could have originated in Sanskrit or in the regional languages. Historically, a give-and-take has occurred in both directions. In the present case, origin in Sanskrit and its continuation in regional languages seems more likely, although the regional languages help us in suspecting the presence of the connotation in Sanskrit in the first place.

Veda Revelation according to Bharty-Hari

Further, BH’s explanation of pratyakşa-dharmāṇaḥ, occurring in the Mahābhāşya (Kielhorn edn p. I.13), runs thus: dharmā ye parokşā lokasya te pratyaksās tesām, “The properties which are beyond the senses for (humanity at large or the rest of) the world are accessible to their senses” (Mahābhāşya tīkā, Abhyankar-Limaye edn p. 38.7). This explanation supports the existence of a meaning element like ’that which was not seen or known before became seen or known’ in sākṣāt-kyta, for the convergence of meaning between pratyakşa-dharman and s-k-d is undeniable.

$2.11 For reasons that do not need to be spelled out to most Sanskritists and Indologists, dharma is a word that has several related meanings and is frequently difficult to translate in a particular context (Fitzgerald 2004: 671-685; Aklujkar 2004: 693-694). The difficulty increases (as is commonly the case with words) as one reaches back in the past to the more ancient texts, of which the Nirukta certainly is one. Historical studies of the meaning of dharma, that is, of its semantic development over time, have naturally been attempted, for example, in Horsch 1967: 31-61 (English tr. 2004: 423-448) and Brereton 2004: 449-489, which, along with Mayrhofer 1963: 94-95, contain references to earlier discussions. The change from dharman, ending in -an and neuter, to dharma, ending in -a and masculine, at as early a time as that of the Atharva-veda and the accentual variation, dhárman and dharmán, seen at an even earlier time make the historical study of dharma complicated. In addition, the Buddhist usage in the sense ‘an abiding entity, a more fundamental existent’ and the Jain usage in the sense that which offers freedom of movement, space’ come in the way of giving a linear account. There is also the knotty question of how the modern Indian understanding of dharma as ‘religion’ came about. It is evident that this understanding is only partially justified; dharma does not have some important connotations that “religion” has. Yet there must be something in its traditional meaning (cf. Aklujkar 2004: 694) that suggested to India’s intellectuals after the arrival of Islam and European powers that dharma be used as an equivalent of the Arabic majahaband English “religion." It is not surprising, given the preceding facts, that in several scholarly writings dharma is left untranslated? or a single Western word like “law” (“loi” in French, “Gesetz” in German), which carries many of the likely meanings, takes the place of dharma and approximates the intended meaning on the strength of context.

$2.12 In the context of the present concern, the alternatives of sticking (a) to dharma and (b) to a single term like “law” are not open to me, for if I follow either one of them, without first trying to determine

21 In those languages like Hindi, Bengali etc. which have inherited the word from Sanskrit, there is no alternative to begin with, unless the translators decide to coin new words or phrases for each meaning of dharma they consider contextually likely.

what dharma in s-k-d is likely to mean, I will, in all likelihood, sacrifice an important indicator of the Veda revelation theory. Secondly, as the Nirukta text in which s-k-d occurs for the first time is relatively close to the time during which the available Vedic texts acquired their present form, I cannot escape the responsibility of specifying which parts of the semantic history of dharma reconstructed by scholars, if any, hold good in its case.

There is also an additional consideration. Yudhisthira Mīmāmsaka (samvat 2020: 344-345, samvat 2030: 366, samvat 2041: 392-393) noticed several decades ago that a dharmabhivyakti view — a view according to which dharma is manifested, not created — was associated with BH.22 But not much attention was given, until recently, to determining precisely what this observation entails and what the implications of dharmabhivyakti as a religio-philosophical concept are 23 As abhivyakti ‘manifestation, coming to notice,’ sākṣāt-kr and “revelation" are obviously connected through their literal or ordinary language meanings, it is very likely that there is a connection between the dharmābhivyakti view and the ideas or details associated with Veda revelation. In any case, we should not understand Veda revelation in such a way as to put it on a collision course with what emerges from the passages mentioning dharmābhivyakti.

$2.13 A detailed demonstration of why I accept what I accept in the semantic histories of dharma offered so far will take us too far afield and obscure my argumentation in the present essay. I will, therefore, attempt such an explanation in a separate publication. To summarize my views:

I accept that the older form of dharma was dharman, with accent on the first or root syllable and neuter gender, or, with accent on the last or suffix syllable and masculine gender.

The basic meaning of dhárman was ’that which is held/ supported/ possessed,’ contextually adjusting to such meanings as (a) that which a person, thing, or groups thereof has, i.e., ‘quality, attribute, property,’ (b) ‘distinctive, unique or defining quality, strength or force,’ and (c) ’essential or foundational nature.’ Contra Horsch and his predecessors, the basic meaning was not ‘(the action of) supporting, (the act of) holding’; dhárman was not an action noun (as distinct from an object noun). Nor was its basic meaning ‘foundation,’ unless by “foundation” a meaning like (b) or (c) is meant.

22 Although the immediate context of the dharmābhivyakti view in most of the references is that of yāga or ritual worship, that is, the same as of apūrva or adrsta in the Mīmāmsā tradition, the view needs to be understood broadly. Its application to the ritual worship context is only a part of the domain in which it applies.

23 I give references to earlier discussions and offer a different explanation of the dharmābhivyakti phenomenon in Aklujkar 2004: S$3.7-9.

Veda Revelation according to Bharty-Hari

On the other hand, dharmán, the stem with accent on the final syllable, was an agentive noun conveying the meaning ‘holder, supporter, ordainer? as earlier researchers referred to above have determined or accepted.

Accordingly, I take dharman in s-k-d as meaning quality, attribute, property and the compound as meaning those who had directly seen (= uncovered or discovered) the properties (of things24 not perceived by the average human beings).25

True, we do not know if the ‘dharma : dharman’ difference was only a matter of compositional conditioning for Yāska (i.e., a bahu-vrīhi was viewed simply as requiring dharman at its end, not dharma, regardless of the meaning and accent intended). Procedural caution, however, suggests that it is better not to rule out the possibility that the stem at the end of sākṣāt-kyta-dharmāṇaḥ was meant to be dharman, not dharma.

That accentual distinction seem to have existed in Yaska’s own Sanskrit would further suggest that it would be prudent to leave room for the possibility that Yāska was aware of the ‘dhárman: dharmán’ distinction. Since he is unlikely to have meant ’those who had seen the supporters, ‘26 the alternative we should accept is that the last member of the s-k-d compound was dhárman in his view and that the compound meant (minimally) ’those who had seen properties’ to him.

The discussion below in $3.1 of the second crucial passage will confirm that in reaching such a conclusion we are on the right track. The sāksāt kệta-dharmatva of the seers means direct, undistorted and extraordinary knowledge of those properties and actions or processes which make the world what it is. The following renderings of s-k-d, therefore, cannot be considered accurate or satisfactory, albeit some of them have greater potential for reconciliation with the evidence (i.e., the basis of my understanding of dharma) than the rest:27

24 Although in most instances below I will speak of properties of things or objects (inclusive of fleeting physical things such as sound continua or speech forms), the properties or effects of actions or states are also to be understood wherever the context is suitable and my interpretation of s-k-d or pratyaksa-dharman is meant.

25 This meaning, which I arrived at by studying the evidence in BH and the passages collected in papers discussing the semantic history of dharma, has support in the ‘padārtha’ rendering found in the comments of Uddyotakara and Vācaspati-misra cited in note 17b and in Durga’s remark amuşmāt karmana evam-arthavatā mantrena samyuktād amunā prakārenaivam laksanah phala-vipariņāmo bhavati. Indirect support can be read in S-M’s gloss mantra-brāhmaṇa and Vrsabha’s gloss abhyudaya-nihśreyasa-sādhana for dharma, since the brāhmana and sādhana, speaking of certain things and actions leading to certain results, must presuppose the presence of result-conducive properties in the things and actions concerned. see $2.15.

26 dharmán, meaning ‘supporter, holder’ is not a synonym for ‘god’, albeit the RV speaks of gods as supporters.

27 (a) Since the expression occurs also in Yaska’s statement and since there is no reason to

Muir 1874: 118: “The rishis, who had an intuitive perception of duty."

Sarup 1921: 20: “Seers had direct intuitive insight into duty.“28 Rajavade 1940: 289: “by whom religion was actually seen.“29

Biardeau 1964: 33: “qui ont l’intuition directe de la loi religieuse/ qui avaient l’intuition directe de la loi religieuse.”

Subramania Iyer 1965: 7: “who have realised the truth/realised that truth.”

Carpenter 1995: 44: “who have directly seen the ritual ordinances.”

Kahrs 1998: 28: “[those who have] direct access to … (ritual and social duty).”

Wezler 2001: 227: “[men] by whom (the) dharma [= this universal ‘law,’ this order of the world, of all its inhabitants, but especially men] was directly and wholly perceived.”

In like vein, we can rule out the meanings (a) ‘religio-spiritual merit, punya’ and (b) ‘property of the mind, intellect, consciousness or self (=property of whatever it is that a given philosophy views as the experiencing subject).’ Such a dharma, referring to a state of refinement or elevation of the cognitive apparatus which enables that apparatus to transcend its limitations, may make some persons capable of sāksātkāra of dharma. However, it can ill fit as an object of their sākṣātkāra. If it does, it does so indirectly, only after the assumption of extraordinary persons who can see

assume a wide gulph in the thinking of Yaska and BH as far as Veda revelation is concerned (cf. $2.3), I include below the translations of s-k-d given while dealing with Nirukta 1.20 as well.

(b) Joshi-Roodbergen 1986: 156 n. 632 (as reported in Wezler 2001: 227) suggest ‘what is right’ and ‘constituent element of reality in the context of pratyaksa-dharman. The latter meaning of dharma/dharman can be connected with what I am suggesting but is not exactly the same, since it is given by presupposing philosophies like Samkhya engaged in listing tattvas ‘constituents of universe models.’

28 Muir’s and Sarup’s translation is close to a meaning of dharma that is definitely and frequently attested in the pre-modern Indian tradition, but it suffers from the fact that no pre modern text is known until now which speaks of duty and mantra in such a way that the former could be seen as a cause of the latter. Duty is usually spoken of as coming from the Veda (= mantras in the present context), not the Veda from duty. Persevering in the performance of duty may, through cleansing of the mind, eventually make one capable of mantra acquisition, but in the available evidence we do not see any direct or necessary link between adherence to duty and receipt of mantras through revelation.

29 Rajavade’s rendering is anachronistic and misleading (cf. Wezler 2001: 228 fn 64). Religion is not known as the object of some sort of extraordinary seeing, although its constituent scriptures, commandments etc. are. Even if we were to assume that Rajavade’s phrase makes sense, there is no reason why the seers of religion would necessarily come to possess mantras (or the s-n-a vāc), since religion is not co-extensive with the mantras, Veda or Sruti.

Veda Revelation according to Bhartr-Hari

qualities imperceptible to others and who can advise regarding what is wholesome is made.

$2.14 Coming to dharmabhivyakti (“manifestation, not creation afresh or acquisition, of dharma’), I have tried in Aklujkar 2004 to sketch the conception of the world, universe or cosmos in which the view would fit. One constituent of the view is the assumption that dharma, a force implicit in brahman (the ultimate or first cause), is distributed over the objects to which brahman gives rise. In other words, it is assumed that there is, in

every evolute, a property or potential which persons of extraordinary insight can identify and relate to actions in such a way as to bring it out and harness for the prosperity and/or continuation of the world. The meaning of dharma in s-k-d that I am suggesting agrees with such an understanding. True, in dharmābhivyakti, the word dharma probably stands for a part of a universal force or energy that is typically revealed (brought to the forefront, made active) by the rites such as Agni-hotra, and in s-k-d (or pratyakşa-dharman), as interpreted by me, the same word stands for something already existing in individual things in a scattered, delimited way that may or may not be made noticeable by the appropriate actions. However, an entity can be the same whether it is viewed as universally present or as present distributively in the individuals populating the universe. The difference observed simply means that, like the terms brahman, sphota, pratibhā etc., the term dharman/ dharma has a layered meaning in BH’s philosophy (and probably also in Yaska’s philosophy). He strings together several conceptually or analytically different entities in one label if the same essence is thought to constitute them.

We can thus determine the meaning of dharma in s-k-d, a term positioned to convey the causal background of the acquisition of the subtle, eternal and transcendental form of language (or of the acquisition of the mantras), in such a way as to abide by the guiding observations made in

S$2.2-9 above.

$2.15 In Indology, there is, rightly, a tendency not to go against the interpretations of traditional commentators. These commentators had, in many cases, inherited an old understanding of the text on which they were commenting. The chain of this understanding can, in several cases, be reasonably presumed to go back all the way to the time of the commentandum author. Besides, some commentators seem to have spent long times, if not entire life spans, in studying the texts they elucidate. Therefore, some Indologist may, by way of objection, understandably draw my attention to the fact that the explanation I have offered is not found in the words of the traditional commentators of the Nirukta and TKV. S-M tell us that dharma in the present context means ‘mantra-brāhmana’ (dharmasyātīndriyatvāt sākṣāt-karaṇasyāsambhavāt dharma-sabdenātra tad

artham mantra-brāhmaṇam ucyate). Vrşabha (S. Iyer’s edn p. 24) paraphrases sākṣāt-kyta-dharmāṇaḥ with abhyudaya-niḥśreyasa-sādhano dharmaḥ yaiḥ sāksāt krtah prāptaḥ te dharmānugrhītāntah-karaņāḥ and restricts dharma to abhyudaya-nihśreyasa-sādhana ’the means of elevating oneself in the world(s) and of reaching the highest spiritual goal.’

As indicated in note 25, my interpretation, while not being identical with those of the ancient commentators, is not irreconcilable. The difference between SM, Vrsabha and me is not a definite difference of essence but probably a difference of using words with narrower or wider meanings. If brāhmaṇa stands for ritual procedure based on knowing the properties of things (bhāva, dravya) or actions (karman), as seems to be the case (recall the bandhutā, ‘connectivity,’ way of thinking writ large over the Brāhmaṇas), then S-M’s tad artha (= dharmartha) mantra-brāhmaṇa conveys what I intend but in a narrower way. It presupposes a ritual context, whereas I speak of what could be the broader philosophical basis of that ritual context.

At the other end, Vrsabha’s abhyudaya-niḥśreyasa-sādhana generalizes the thought involved, gives no hint of being confined to the ritual context and speaks of a further stage by incorporating the anticipated result: wordly elevation of all sorts (see Aklujkar 2004: 702 on abhyudaya) and moksa or nirvāņa. It reflects an understanding extending over the collectivity of things, actions and their properties, without resorting to any expression like ‘a thing/action with property P-1 leads to result R-1.’

In the gloss of Durga, we have the link between the generalized causality of Vrsabha and the particularistic idiom present in my understanding: [ye; see note 119] amuşmāt karmana evam-arthavatā mantrena samyuktād amunā prakāreņaivam-lakṣaṇaḥ phala-vipariņāmo bhavatīti paśyanti te rşayaḥ. “rșir darśanāt” [Yaska 2.11] iti vakşyati, tad etat karmaṇaḥ phala vipariņāma-darsanam aupacārikayā vrttyoktam sāksāt-krta-dharmāna iti. na hi dharmasya darśanam asti, atyantapūrvo [+ ontārupyo?] hi dharmaḥ. “Rsis (are) those [who] perceive that from action k, combined with a mantra having meaning m, a transformation (in the form of) result r, having such and such characteristic, comes about in such and such manner. It is this perception of the result-transformation of an action that has been

30 I am not sure about why Durga uses viparināma in addition to phala. His intended meaning could be transformation that is the outcome, the result itself is the transformation’ (phalam eva vipariņāmaḥ, a karma-dhāraya) or, what amounts to the same thing, ’transformation into the result’ (phale viparināmah, a locative tat-purusa). However, it is possible that Durga meant to convey that, because of its association with a mantra and a Veda-recommended process, the ordinarily expected outcome of an action undergoes a transformation — that what are outwardly or physically the same actions produce extraordinary results when performed as a part of a ritual. In that case, even a genitive tat-purusa dissolution (phalasya vipariņāmaḥ) could have been intended. The phrase karmaṇaḥ phala-vipariņāmaḥ would then have the same structure as deva-dattasya guru-kulam.Veda Revelation according to Bharty-Hari

expressed (by Yāska with the word) s-k-d through a metaphorical mode (of expression), for (really speaking) there can be no perception of dharma. Indeed dharma is something that cannot at all be given a form.“31

As the explanations of S-M, Vrsabha and Durga can thus be linked and as the explanation of Durga, the most ancient of the three commentators, is closest in spirit to my explanation, I consider my explanation to be free from conflict with their explanations. Even if it were to be thought of as conflicting, I would retain the freedom to recover an earlier understanding on the basis of evidence preserved in the works of a relatively early author like BH, who lived before S-M and Vrsabha and may not have been far removed in time from Durga.

$2.16 As stated in $2.8, we should accept as correct that interpretation of TKV 1.5 which agrees with the notion of mantra. It would be natural to maintain such an expectation or requirement because the Vedas have primarily been considered to be mantra.” Now, the invariant understanding of mantra, whether we speak of the Vedic/Brahmanical, Buddhist or Jain tradition, is that a mantra is a formulation in language that is assumed to have the power to affect the physical world or what, at the moment of the formulation’s application, is thought to be reality. This implies that a close connection between language and material objects or forces seen as material is assumed.* Statements such as Bhava-bhūti’s (Uttara-rāma-carita 1.10) capture this understanding: rșīņām punar ādyānām vācam artho ’nudhāvati. “In the case of the foremost (or most ancient) seers, content comes running behind the utterance (that is, in the case of such seers the meaning of

31 Since I cannot make a contextually appropriate sense of the reading found in the edns, which literally means ’exceedingly/absolutely unprecedented,‘I have followed the emendation that occurred to me as expressing a contextually suitable meaning. Besides, a corruption of ontārūpyo into ontāpūrvo is plausible.

32 In the thinking of the believers, the special power of the Vedic word must have something to do with its source or the way in which it is formed. Either way, there will be a close connection with the power or the process through which the Veda comes into being or is revealed.

33 (a) Alper (1989: 12) reports that Agehananda Bharati (1965: 102) saw, in the mantra concept, generation of a “somewhat complex feeling-tone” in the practitioner. Alper speaks of the same as “emotive numinous effect” on the practitioner. Such a generation or effect, being physical, supports my observation here. However, it will only be a part of what I mean by “physical effect.” Closer to what I have in mind is Hacker’s (1972: 118) remark (reported in Alper 1989: 14): “From ancient times there has been in India the conviction that mental representations, if reaching a high degree of intensity, are capable of bringing about a reality not only on the psychological level but even in the domain of material things.”

(b) The terms “material” and “non-material” are useful, but we should not forget that the distinction between the material and the non-material is not always made in the same way in the Sanskrit tradition as it is in the Western. For example, praksti, distinguished from the non material puruṣa, is viewed as the cause of mind etc., generally considered to be non-material, in the philosophy of the Samkhyas. There, materiality has gradation.

what they say is anxious to transform itself into reality, whereas, in the case of the ordinary good people, reality simply corresponds to what they say; such people are only truthful; they are not makers of truth).” The mantras are unlikely to have been assigned this power to transform material reality if an intimate connection between them and the properties of objects was not admitted.34 The situation should be analogous to the one noticed in the determination of merit-producing grammatical expressions. (sādhu śabdas). Just as the siștas are said to be able to determine which expressions have the capacity to lead to merit in a given period, the rşis of the creation stage should be able to determine which realizations of vāc are conducive to the good of men and hence should be transmitted or admitted in Veda formation.35 The connection of such realizations with the contents of creation itself should be, in early Brahmanical or Indian thinking, the basis of their efficacy at the physical level.

$2.17 One need not establish a one-to-one correspondence between the Veda statements, their correspondence with the real world, and their mantratva, ‘ability to affect the real world.’ That sākṣāt-krta-dharmatva is a cause of the vāc experience need not necessarily mean that a capturing or publicly accessible verbalizing of the experience has a mantric effect or an ability to recreate the experience or the reality. X leading to Y need not give Y’s resemblance an uplifting quality (a potential for causing spiritual growth) or an ability to affect the world, or to acquire X (sākṣāt-kệta dharmatva) itself. One, therefore, needs, also the persons who know or can know the real properties of things in terms of which properties are beneficial and which ones harmful and under what situations. In the ancient Indian view as preserved or presented by BH (and probably also by Yāska), the Veda the first seers make accessible has the potential to shape and affect36 reality, that is, it has mantratva, but, to be operative or useful, that potential of it needs seers who have the knowledge of properties of objects

34 Passages in BH which speak of mantras affecting objects are: TKV 1.33, 174; TK 2.323 and its Vrtti.

35 (a) The maharsis may also be able, in BH’s universe, to determine which particular benefits can result from a particular variation (ūha).

(b) Although I speak only of dharma and benefits in most contexts in order not to make any statements more complicated than they need to be, the opposites adharma, harm and absence or blocking of benefit should also be understood wherever appropriate.

36 TK 1.10 and its Vịtti speak of the Veda as a vidhātr (‘fashioner, maker’), through prakstitva (being the constituting material’) and through upadestr-rūpa (“teaching role’). The second attribution applies to vyavasthās (’establishments, arrangements, fixities, a series of regularities’) subsequent to creation. Under it, the programmatic or gene-like role of the Veda is replaced by a textbook-like role. The Veda then does not shape creation organically but by being a kind of manual or blue print that the rșis acquire or retrieve and interpret. A similar thought is expressed in Mānava-dharma-śāstra/Manu-smrti 1.21: sarveşām tu sa [= Ādi-brahmā] nāmāni karmāņi ca prthak prthak/veda-sabdebhya evādau prthak samsthāś ca nirmame//

Veda Revelation according to Bharty-Hari and who know which of the Veda’s linguistic units or sequences go with which object properties.

$2.18 We need to respect the ‘avara : apara’ difference of reading as appendix 4 points out. While the connotation of avara may clash with that of apara (Wezler 2001: 218-222), there can be no doubt that both words convey otherness. This common ground suffices for our understanding. The secondary differences will not affect our reconstruction, for the avara/apara belong to the Veda transmission stage and their qualitative difference from the s-k-d rsis or lack of relevance to revelation per se is made clear by the adjective a-s-k-d applied to them. For the same reasons, whether they are to be thought of as rşis, albeit of a different sort (śrutarsi), should not concern us. 37 Further, the issue of whether they are literally one group or generation of human beings or simply a logical category distinguished from the s-k-d rșis that, in actuality, consisted of several groups or generations need not come in our way. An abstraction or lumping together of all those who received the s-n-a vāc (or the mantras or the Veda text or the bilma) and who set down for transmission (note 7) the Veda and Vedāngas (including the Nighantu and its commentary, the Nirukta) will do for our purpose.

$2.19 Regardless of what the older or original meaning of bilma may be, it cannot be doubted that BH uses the word in the present context in some such sense as ‘representation, replica, image, resemblance, reflection.’ It is the only word in the Vștti that can correspond to anukāra ‘doing/ acting/fashioning after’ found in the karikā. The gloss bhāsanam, being/ becoming visible’ or ‘one that makes something visible,’ taken over from the then current (but not necessarily ungenuine) text of the Nirukta also points in the same direction, just as the fact that Vrsabha (pp. 22, 24-25) glosses both anukāra and bilma with praticchandaka ’likeness, picture, statue, image, substitute’.

As far as Yaska is concerned, the bilma must, minimally, be something related to the mantras. Otherwise, there will be no useful connection between his second sentence and third sentence. The two assertions, ‘The avara/apara got the mantras’ and “The avara/ apara set down for transmission certain texts,’ will have a common subject or agent, but why they are made one after the other will not be known. Something relating the mantras and the texts must occur in the latter assertion. Since upadeśa is too general a word and veda comes as one among the triad (imam grantham … vedam ca vedāngāni ca), claiming no exclusive connection with mantra, only bilma can provide the necessary link through bilma-grahaņāya. The entity it

37 Wezler (2001: 225 fn 51) informs us about a new division of rșis partly based on traditional considerations that Gurupada Sarma Hāldar (1955: 64) mentions in his Vrddha-trayī.

refers to must either consist of the mantras or be a means to the mantras. Under the first alternative, it would be a collection, recasting (rearrangement, redaction etc.) or reflection of the mantras (the entity received by or revealed to the s-k-d rşis); under the second, at least, something that follows the lead or determining status of the mantras. Neither meaning would be irreconcilable with that of anukāra ‘doing/ acting/fashioning after.’

$2.20 After this consideration of the crucial words that are common to what originates with BH and what he cites, we should briefly touch upon one final word from the first part in which some potential to influence our emerging interpretation may be seen. Although BH quotes a few passages from a text or class of texts he calls Purā-kalpa, the statement ending with ity esa purā-kalpaḥ is merely a summation of what BH learned from the Purā-kalpa text or texts, or their fragments surviving in his time, not a verbatim reproduction or quotation in the strict sense of the term “quotation.” The statement’s diction does not match that of the Purā-kalpa passage BH actually quotes in TKV 1.124-128. It has the stamp of BH’s own style as it is found in the TKV. 38 Vrsabha does not take BH’s purā-kalpaḥ as a reference to a specific text but as meaning artha-vāda, ‘a statement stating or implying recommendation or non-recommendation of a proposition in the Veda.’

$2.21 Maintaining as much awareness of the foregoing considerations as is possible to maintain and expecting my readers to do the same, I will now translate TKV 1.5 as follows:

“anukāra (etc. in the commentandum is meant to convey the following): About to reveals to those others who have not discovered the (ordinarily

38 (a) Note the expressions pravedayisyamāņāḥ, dysta-śrutānubhūtam and ācikhyāsantaḥwhich are unlikely to occur, especially so close to each other, in the Sanskrit exemplified by the surviving of Purā-kalpa passages. Also, distinctive is the sequence of adjectives sūkṣmām nityām atīndriyām and ysayaḥ sākṣāt-kyta-dharmāṇaḥ mantradyśaḥ which matches the question-anticipating style noted elsewhere in the TKV. For an explanation of this stylistic feature, see Aklujkar 1991b: $2.4h.

(b) At one point, I did lean toward the view that BH’s words yām … ācikhyāsantaḥ were a quotation from the Pura-kalpa as Wezler (2001: 240) reports.

39 It is unlikely that pra in pra+ vid is not significant. It could be an intensifier (as in pravarşa “heavy rain,’ praśānta ‘very calm’ etc.), carry the connotation of being ahead, being in the front, foremostly’ (as in pra + jñā etc.) or give the verbal root a sense of ‘reveal, make manifest’ (as in kāma-pravedana of Pāņini 3.3.153). A translation like ‘about to make (that vāc) well known, wishing to enable the recipients to grasp (that vāc) well,’ a translation like “about to convey for the first time,’ or a translation like ‘as those who will reveal (that vāc)’ will fit the context. I have preferred the last alternative because it would conform to Panini’s usage and Mahābhārata 7.52.1 (critical edn): śrutvā tu tam mahā-sabdam pandūnām putra-gyddhinām/cāraih prevedite tatra samutthāya jayad-rathah//. “When the informants communicated that thing, the news) after having heard the loud sound made by the son-loving Pandavas, Jayad-ratha, having got up (went

Veda Revelation according to Bhartr-Hari

imperceptible) properties of things that subtle, eternal and sense-transcending (form of) speech which they (themselves) behold, the seers who have discovered the (ordinarily imperceptible) properties of things (and) to whom materially effective speech formations appearł0 set down for transmission an image, 41 as they wish to convey, like something that happened in a dream, what they experienced through sighting and hearing. 12 This (i.e., the content of the foregoing sentence) is an ancient (or traditionally handed down) thought formulation (or systematized knowledge). Indeed, [another reliable or respectable source, the Nirukta] says: ‘There came about (or there were) (at a distant time) seers who had discovered the (ordinarily imperceptible) properties of things. Through instruction, they have entrusted materially effective speech formations

to the gathering of the allied kings).” One expects an association of revealing something that had remained previously unknown in the case of informants or spies in this context.

40 The qualification mantra-dyśaḥ, to judge from its position in the sentence, could have a sense approaching that of a present participle: ‘as the mantras appear to them, while they are in the act of seeing the mantras.’ In other words, there is no essential difference between seeing the mantras and seeing the subtle, permanent and sense-transcending form of vāc. However, the mantras do not exhaust the specified form of vāc; the domains covered by the two terms are overlapping but not coextensive.

41(a) As $2.19 indicates, I do not need to make a special effort to find a perfect translation (if there is such a thing as perfect translation) for anukāra in the present essay. At present, I think that “image” works best, because it can convey resemblance as well as the capability to function in the place of the original (= the s-n-a vāc) that are the contextually necessary meaning components of anukāra.

(b) The comments in Carpenter 1985: 194 are helpful in understanding the notion of anukāra: “To say that the Veda, as anukārah, is the “imitative resemblance” of Brahman should not be taken to mean, however, that the Veda somehow offers us a “description” of Brahman. For Bhartr-hari, the function of Vedic revelation is not to provide us with a representation of the “object” Brahman. Rather, the Veda functions as an imitative “presentation” or Darstellung of the unity of Brahman, mediating Brahman directly through the dynamic idiom of language and action in their inseparable interrelationship. The Veda is thus the outward linguistic form of the dynamic self-manifesting act of the Word-Principle itself. By virtue of this, it can itself be described as the “arranger” of the world, as a cosmogonic principle essentially identical with Brahman. In its concrete linguistic form, however, the Veda mediates the unity of Word Principle by manifesting the original order through which the world is related back to its unitary source.” Carpenter (1995: 41) further defends the rendering ‘imitative resemblance’ by drawing attention to the formation of the word and its uses by BH elsewhere.

42 Cf. Vrsabha pp. 24-25: drsta-śrutam iti. napumsake bhāve ktaḥ tena darśana-śravaņābhyām anubhūtam artham … dysta-śruta-grahaņam sarvendriya-jñānopalaksaņāya. “(The words) dysta and śruta (in the commentandum are derived by adding the past participial suffix) kta (=ta), so that a neuter (noun) in the sense ‘state, action’ is derived). Thus, (the compound dysta-śrutānubhūtam comes to mean) ‘an entity experienced through seeing and hearing. … The employment of dụsta and śruta is meant to stand for cognition through all senses.”

43 sam + pra + dā literally means ‘give forth/away in entirety (or as a collectivity).’ Such a meaning could imply, at least in certain contexts, ‘absence of holding back anything associated with the object being given, relinquishing of ownership or responsibility, making the recipient responsible’; cf. the use of the same prefix and root combination for the action of giving the daughter away in marriage that our standard dictionaries record. As mantras are also considered to be valuable, a sense of expecting the recipient to be responsible for their preservation is likely to be present in the transaction. Hence my translation with “entrust”.

to others who had not discovered the (ordinarily imperceptible) properties of things. The others experiencing fatigue toward instruction, have set down for transmission this corpus (i.e., the commentandum of the Nirukta, the Nighantus etc.; see Aklujkar 1999: fn 9) and the Veda and the Veda ancillaries in order to grasp the image.45 (The word) bilma is (to be thought of as) bhilma or (as) bhāsana.45

$2.22 The involved sentences of the foregoing literal translation can perhaps be made easier to follow by the following summary:

“The TKV 1.5 passage speaks of two groups. One group consists of seers who discover dharma(s)/dharman(s), sight mantras and behold a form of vāc, which is beyond the senses of ordinary people. They wish to reveal the vāc they have beheld to others who have not discovered dharma(s)/dharman(s). In the act or for the act of revealing, they do a samāmnāya of bilma ($2.19). The recipients of

the bilma constitute the second group.’ It will be noticed that the syntactic simplicity of these summary sentences has come at the cost of switching from normal English to ‘Indologese’ — of not translating the culturally and theoretically pregnant words of the original,

44 (a) I have attempted to indicate in my translation the change from the perfect babhūvuh of the first sentence to the aorists samprāduh and samāmnāsisuh in the second and third sentences. As Bhandarkar (1868: vi-x) pointed out, the aorist in the early period of Sanskrit functions the same way as the English present perfect (i.e., it denotes past in general and the recent past) and that this deduction of his based on a study of the attested usage is in keeping with Pāṇini 3.2.110-111, 3.2.115 and 3.3.135. In other words, the aorist forms indicate flowing of the past into the speaker’s present — the actions expressed are presumed to have a connection with the speaker’s present. Evidently, the gift of the mantras and the transmission of the Veda and the Vedāngas (including what the Nirukta comments on, the referent of imam grantham) were viewed by Yāska as activities related to his own time through their effects or products, while the s-k-d seers were viewed by him as coming into being or existing in a very distant past. He had no access to the s-k-ds (they were a cut-off fact), but what they gifted to the later generations and what the later generations composed to grasp the gift was within his reach.

(b) For other understandings of the difference between the meanings of the perfect and aorist forms, see Wezler 2001: 219.

45 As our concern here is with how BH understood Yaska, and not with Yaska himself, I am presupposing BH’s ‘anukāra’ interpretation of bilma and genitive tat-purușa dissolution of bilma-grahaņāya.

46 (a) In the ‘bilma : bhilma’ equation, one can read a suggestion to the effect that bhilma is easier to understand for Yaska’s reader than bilma or that bilma is a deviation from the familiar bhilma, a form current in a different time or region. The requisite phonetic similarity exists between the two forms. Such is not the case with bhāsana. So, if one insists that bhilmaand bhāsana must be thought of as prompted by exactly the same intent, one must translate the sentence as “The meaning of bilma is the same as that of bhilma or bhāsana.” Here, vā or “or” would be indicative of an alternative or of addition. There would be no commitment on the sentence author’s part as to whether the meanings of bhilma and bhāsana are different. It is, therefore, possible that BH thought of both bhilma and bhāsana as leading to his rendition by anukāra (S 2.19), although normally one would have thought only of bhāsana’an act/instrument of reflecting as capable of suggesting the idea of anukāra. Either such synonymy or the appearance of bhāsana in the second (i.e., final) place, which the Sanskrit authors usually reserve for stating their preferred view, could have prompted him to understand bilma as anukāra.

(b) Commentators have taken bhilma as a derivative of the root bhid to break.’ As bhās is not attested in the sense of ‘breaking,’ a consequence of this derivation can only be that vā should indicate an alternative.

Veda Revelation according to Bharty-Hari which required several paragraphs above for their clarification (and have been the subject of countless paragraphs in other publications).

$2.23 In the translation in 32.21, I have used “discover” for sākṣāt + kr, “appear, sight” for -drs (“catch sight of” may also do) and “behold” for pas (“notice” too would be acceptable to me). This is largely to highlight the fact that the original Sanskrit passage uses three historically different root words (one of them joined to a gati prefix) that are usually taken as practically synonymous. Among these, the translation of the first has received some justification from me in $2.10. It also fits the frame the other relevant theoretical concepts form. However, I would not claim that the semantic distinction I have made between drs and paś (found to be in complementary distribution at the formal level) is exclusively valid. I surmise on the basis of the early occurrences of drs and pas recorded in our standard dictionaries that drś was originally used for those situations in which the experience of seeing was thought of as initiated by the object. The root seems to have a sense closer to that of “appear.” Thus, not being subject-controlled, it could have the connotation of an experience that was occasional and time-limited. pas, on the other hand, seems to be closer to English “observe” or “spot” and to carry a suggestion of steadiness or intent on the part of the subject/ agent (cf. the historical relationship of paś with spaśand its Indo-European cognates including “spy”). Such a distinction between the meanings of the two roots may be said to be supported by the accounts in the Bșhad-devatā etc. that speak of mantra seeing as something happening sporadically and unexpectedly. Seeing of the s-n-a vāc, on the other hand, is more likely to be conceived as a result of intensive and sustained effort, since that vāc is the highest reality, and its experience is akin to a trance. The phenomena may essentially be the same, but the associations authors like BH have with them would determine word choice.

In his article “Justification for verb-root suppletion in Sanskrit,” Professor Madhav M. Deshpande (1992), suggests, on the basis of passages such as uta tvah paśyan na dadarśa vācam (Rg-veda 10.71.4a), that pas expresses seeing stretched out in time (cf. English “observe” and “gaze”) and drs expresses seeing as a conclusive event (cf. English “see”). This suggestion does not conflict with what I have surmised, but the basis of the aspect difference implicit in it is different from the one I have presumed. I also consider it possible that pas had a connotation of wishing to locate or to spot (a presumption of searching), whereas dró had a connotation of perceiving the object as it really is (cf. paśyantī as the name of a language level or phase in which a speaker is thought to be in search of the appropriate linguistic form to express the meaning he/she has in mind and darsana as standing for a view of what is reality or is believed to be reality). Any attempt to determine the shades of meaning, however, would not succeed if the

instances of use of pas and drs in close proximity come from a period in which the usage is prompted merely by a formal or grammatical convention (as, for example, is the case with forms like “be,” “am,” and “was” in English).

Another interesting feature of the original is that all the seemingly synonymous root words have been used with respect to objects — dharma, mantra and vāc — we would not normally think of as amenable to the act of seeing taken in a physical sense. Clearly, BH expects us to take ‘seeing’ in a metaphorical sense. This, in itself, is not problematic. It is quite common, probably in all languages but particularly in the Indo-European languages, to use “see” in the extended senses such as ‘observe,’ ‘perceive,’ find out,’ ‘understand’ and ‘visually imagine.’ What is problematic is that in the present case we do not know what the intended extensions could be.

The second TK passage having a

bearing on Veda revelation $3.1 Context: TK 1.173: avibhāgād vivịttānām abhikhyā svapnavac chrutau / bhāva tattvam tu vijñāya lingebhyo vihitā smrtih // “Those (rșis) who evolve from the (ultimate) unity (namely, brahman) come to know47 the Sruti as (ordinary persons come to know something) in a dream. As for+8 the Smrti, it is fashioned on the basis of the indications in the Sruti) after knowing the real nature of things.”

TKV 1.173: … yeşām tu svapna-prabodha-vrttyā nityam vibhakta-puruṣānukāritayā kāraṇam pravartate teşām — ?şayah kecit pratibhātmani vivartante. te stam] sattā lakṣaṇam mahāntam ātmānam avidyā-yonim paśyantah prabodhenābhisambhavanti.19 kecit tu vidyāyāmvivartante. te mano-granthim ātmānamākāśādişu bhūteşu, pratyekam samuditesu vā, visuddham anibaddha-parikalpam tathaivābhisambhavanti.50 teşām cāgantur avidyā-vyavahāraḥ sarva evaupacārikaḥ. vidyātmakatvam tu nityam anāgantukam mukhyam. te ca, svapna ivāśrotra-gamyam śabdam, prajñayaiva sarvam āmnāyam sarva-bheda-sakti-yuktam abhinna-sakti-yuktam ca paśyanti. kecit tu

47 From the way the Vrtti renders the idea of this kārikā half (svapna iva … āmnāyam … pasyanti), it is evident that the intended sense of abhikhyā must be something like ‘see’ or ‘grasp’. The root khyā contains elements of seeing’ and ’telling, conveying’ (cf. its use in khyāti, samprakhyāna, ākhyāna, khyāta, vikhyāta etc. and the forms resulting from its reduplication, namely caks, cakşus.). The prefix abhi indicates ‘facing, being in front of.’ The meanings the commonly used dictionaries record for abhi + khyā appropriately range in the same general area, with ‘impressiveness, beauty’ and ‘being well-known, celebrated as understandable extensions. Accordingly, I take the noun abhikhyā as basically meaning the process of coming to know, the state of encountering someone or something.’

48 The use of tu ‘however, on the other hand’ here is prompted by the intention to convey that the Smrti needs something more to come into existence, not to suggest that there is a stark contrast or absence of relationship between the Sruti and the Smrti.

49 I have followed here Vrşabha’s reading prabodhena instead of the reading pratibodhena of the currently available TKV mss and the published edns. Also, prabodhena is supported by TKV 2.152, quoted in note 54 below.

50 Unfortunately, Vrsabha’s commentary on this sentence is very poorly preserved.

Veda Revelation according to Bharty-Hari

puruşānugrahopaghāta-visayam teşām teşām arthānām sva-bhāvam upalabhyām nāyesu kvacit tad-vişayāni [/tat-tad-°) lingāni drstvā ca, drstādrstārtham smrtim upanibadhnanti. śrutim tu yathā-darsanam avyabhicarita-sabdām eva, prathamam avibhaktām punaḥ samgrhīta-caraṇa-vibhāgām, samāmanantī— ty āgamaḥ.

The inherited views of those who think that the (original) cause52 constantly (that is, again and again)53 proceeds forth (to create), in the manner of sleeping and waking up,34 fashioning itself after the individual persons (or the distinct puruṣas) is this: Some seers come about as a multiplicity in the unitary entity pratibhā (that is, at a stage which is just one step short of reaching brahman and to reach which most persons must go through a long process of spiritual cultivation). They, seeing that (pratibhātman which is the same as) mahat ātman, the one characterized by Being (alone, that is, the one which is the undifferentiated or highest-level existence), matrix of nescience, 56 join that

51 The relative clause beginning with yeşām finds its fulfilment in teşām … ity āgamaḥ. (I have separated, with dashes, the sentences between teşām and ity agamaḥ that specify the nature of agama). In the translation, I have brought this syntactic connection forward to facilitate comprehension.

52 The use of kārana, ‘one which makes something else active,’ is attested elsewhere in the context of the very first cause; cf. Samkhya-karikā 16: kāraṇam asty avyaktam. pravartate tri-gunataḥ …,: “There is (thus) the unmanifest cause. It becomes active on account of the three strands …”

53 Vrsabha p. 226: nityam iti śāśvatam kāraṇam iti sambandhah. However, I think that nityam here is an adverb connected with pravartate and has the sense ‘as a given’ or ‘recurringly, constantly’; cf. nitya-prahasita ‘frequently laughing,’ nitya-prajalpita chattering, talkative’ etc. cited by Patañjali (MB 1.7 and under P 8.1.4 III.364). Given the specification svapna-prabodha vrttyā (“as sleep is followed by awakening and awakening by sleep’) and the presence of vibhakta puruṣānukāritayā between nityam and kāraṇam, such an understanding of the sentence would be more natural than Vrsabha’s, unless by kāraṇam Vrsabha means the phrase kāraṇam pravartate.

54 TKV 2.152: kācit svābhāvikī pratibhā, tad yathā parasyāḥ (praksteh pra] thamam sattā-lakṣaṇam ātmānam mahāntam pratyānugunyam, sușuptāvasthasyeva prabodhānugunyam phala-sattā-mātram [ → phala-mātram?) nidrāyāḥ. “Some action-prone knowledge exists naturally (i.e., arises from its very context or state of being — does not need a conscious effort ). An example (of this is the) conducive stance of the highest nature (i.e., the most primary cause, the sabda-tattva brahman or prakệti) toward the mahat ātman, the first (evolute) characterized by (worldly) existence (which stance is) like a sleeping person becoming prone to waking up merely as a consequence (phala, or as an existence of a consequence) of the sleep (he/she had).”

This Vịtti passage which is related in content to TKV 1.173, suggests that the intention in TKV 1.173 too is more likely to be to speak of the fact that sleep and waking follow each other naturally (at least in the case of most persons considered to be normal).

55 BH has used the root vi+ vrt or its derivatives in speaking of the seers, a group of whom receives revelation. As vi+ vrt (taken in the sense change only apparently’) is later contrasted rather sharply with pari + nam (taken in the sense ’to change really, to undergo an actual transformation’), a question may arise as to the truth of the arisal of the seers and of the revelation experienced by them. BH’s use of the dream analogy may further be used to question the revelation’s truth. However, the primary import of vi+ vrt in BH’s writings is not illusory change but ‘arisal of many effects from a single source or cause.’ When these many effects are thought of as a collectivity, even pari + nam can be used. I have substantiated this semantic reconstruction in an unpublished article.

56 Nescience is the cause of the subsequent evolutes in the view being stated at this point. When these evolutes are looked at from the point of view of the original or highest reality, they do not

(pratibhātman),57 through awakening (that is, through advanced awareness, realizing the ultimate futility of multiplicity). Some (seers), on the other hand, come about as a multiplicity in vidyā.58 They, likewise,59 join the ātman that has the knot of mind (that is, the ātman equipped and delimited for engagement with the world and that, yet, remains) pure (and) conception-free with respect to the elements ether etc., taken jointly or severally. Their adventious, nescience-based interaction (with the world) is not literally so (that is, it can be predicated of them only through a transfer of ordinary persons attributes to them). What is constant, intrinstic and primary (to them) is (their) vidyā-nature. They see (our) traditionally handed down text in its entirety with insight alone as one would hear in sleep a word (or sound) inaccessible to the sense of hearing — (the text) having all the powers of differentiation and having the powers inseparable (from itself, i.e., the subtle

really exist. Yet at the level of ordinary experience they are very much real. Therefore, a force that would account for their not-true-to-reality apprehension needs to be presumed between them, on the one hand and either the mahān ātman or the ultimate cause, on the other.

57abhi + sam + bhū, which must literally mean ‘come into existence as a whole and as something facing (= appear next to, link up with),’ is appropriately attested in the senses ‘reach, arrive at’ and ‘obtain the shape of, be changed into’ (cf. Monier-Williams, p. 73) and glossed with prāpnuvanti and eki-bhavanti (Vrsabha, pp. 226-227).

58(a) A contrast between pratibhātmani vivartante and vidjayām vivartante is intended in the context. The essence in different uses of pratibhā by BH seems to be that of ‘action-centred knowledge, a state in which what will happen or is to be done presents itself to the mind.’ An element of passivity or neutrality (of which ’lack of conscious control’ and spontaneity’ would be other descriptions) may be said to be implicit in the present context. How pratibhā differs from vidyā is not entirely clear, but the contextual indications lead me to surmise that vidyā is knowledge in which informational content is prominent. Knowledge of itmes as well as how they work or are to be used, association with mind or worldly personality (mano-granthi) and engagement with the constituents of the world (bhūtas) seem to shape it in the translated passage. Those who come to exist in vidyā are not said to differ in wisdom, spirituality or capability from those who come to exist in pratibhātman. However, they are open to an engagement with the world and are capable of descending, without becoming ‘polluted,’ to a level that would be lower in the estimation of liberation-seekers. The ones coming to exist in pratibhātman, on the other hand, have no proneness to engagement with the world. They remain at what would be thought of as a higher stage on the path of liberation.

(b) The myths in which Brahmā or Prajāpati first gives birth to some sons who turn away from the world to asceticism and are utterly free from delusion and in which only the second batch of Brahma’s or Prajapati’s sons help him in continuing with the creation of the world

should be recalled, as also the passages speaking of the mind-born sons of Brahmā or giving one or both lists of Saptarsis; cf. Agni-purāņa ch. 17.15-16; Mahā-bhārata 12.160.15-16; Vāmama purāņa, Saro-māhātmya section, Adhyāya 22, A.S. Gupta’s critical edn., p. 247; O’Flaherty 1975; Dimmitt 1978: 155-156, which translates Kurma Purāņa 1.10.1-38, and p. 310; Mitchiner 1982, particularly pp. 233-248. The structure implicit in these myths etc. is reflected in the philosophical statement translated here. BH may have preserved for us the earliest Indian understanding of the specified complex of texts.

59The word tathaiva of the original could not have been meant to convey that the object of becoming one is exactly identical; mano-granthi is not mentioned in the case of the first merger. The intention behind the use of tathaiva, therefore, must be to convey sameness in the manner (‘without delusion, without any loss of purity’) or intensity of merger.Veda Revelation according to Bhartr-Hari

form of the authoritative Veda.).60 Some (of them), additionally,61 having ascertained the nature of specific entities as it concerns the helping or harming of humans and having seen indications to that effect in some parts of the traditionally received texts, compose the Smộti, meant for mundane and non mundane objects. As for the Sruti, they set (it) down for transmission (note 7) as it was seen in the experience described above), without a change of wording (or sound) whatsoever — initially, undivided i.e., as a single corpus), later incorporating the caraņa division (see appendix 1, note to point 5). “62

$3.2 A part of the second passage, TKV 1.173, is structurally parallel to the first passage, TKV 1.5. The phrases svapna ivāśrotra-gamyam sabdam (echoing abhikhyā svapnavac.chrutau of the kārikā) and prajñayaiva sarvam amnāyam sarva-bheda-sakti-yuktam abhinna-sakti-yuktam ca paśyanti establish its connection with the first passage, since they mean “as one would hear in sleep a word (or sound) inaccessible to the sense of hearing” and “They see (our) tranditionally handed down text in its entirety with insight/mind alone — (the text) having all the powers of differentiation and having the powers inseparable (from itself). “63 The additional details TKV 1.173 provides can, therefore, be used in pursuing our objective.

The seers who arise as a multiplicity ‘in vidyā’ are credited with the transmission of the Veda as Yaska’s s-k-ds are credited with the bestowing of mantras and BH’s s-k-ds are credited with the conveying of the s-n-a vāc.

60 (a) I have taken sarva as a qualification of sakti in light of TK 1.2 and its Vrtti. It is also possible to take the word as qualifying bheda and translate as ‘power (or potential) for all differentiation. But what such a translation conveys may not ultimately be different from “all powers of differentiation.”

(b) Vrsabha p. 228: sarva-bheda-sakti iti nānā-phala-janaka-karma-prakāśakatvāt. abhinna sakti iti, yad anekam apy ekam karma prakāśayati. yad utkam sarva-śākhā-pratyayam ekam karma iti. “The author speaks of the amnāya as sarva-bheda-sakti-yukta because the amnāya enlightens us about variously productive actions. He employs the adjective abhinna-sakti-yukta, because more than one Vedic text throws light upon the same action. As has been said, the various recensions of the amnāya jointly enlighten us about a single ritual action.” Although it is thus possible to make sense of the adjectives as informing us about the amnāya text, I think the adjectives were really meant for the basis of the amnāya text (the subtle form of language or brahman). They are too ontological to be applicable to a text literally. One would not normally ascribe a bheda sakti (all bheda-śaktis at that) to a text in the conventional sense. Nor is a query like ‘Are the powers separate?’ likely to be raised in the case of a text.

The function of tu at this place cannot be to suggest a contrast of the predicate of the preceding sentence with the predicate of the sentence underway. The particle is meant simply to indicate that the author is now turning to another ‘scene’ or adding a new thought.

62The stages spoken of in this passage remind one of the sabda-pūrva-yoga process mentioned in TKV 1.14-22 and of the process of paramātma-siddhi, attainment of the amrta brahma or kşema-prāpti mentioned in TK 1.144 and its Vrtti (which includes TK 1.145-48).

63 In TKV 1.5, we do not have a clear mention of the cosmogonical context of Veda revelation. Perhaps only a hint to the effect that it is a phenomenon belonging to the beginning of creation can be read in the assertion that division is absent in the beginning. However, in TK 1.173 and its Vrtti, it is made explicit that the revelation talked about takes place at the time of cosmic creation.

The same vidyā seers are also credited with the knowledge of what in a thing brings benefit and what causes harm, as well as with the composition of the Smrtis which preserve this knowledge. The sākṣāt-krta-dharmatva must, therefore, have a close connection with knowing the properties of things, and the meaning of dharma in its context must be ‘property, quality, attribute’(cf. $2.13-15).

$3.3 The fact that we are not required to propose any unusual meaning but can make do with one of the most common meanings (in my view, the oldest meaning) of dharma should suggest that we are on the right track.

Our surmise receives support from a few other TK passages. Just before our passage 2, BH brings up the topic of whether the śāstra (of which the Veda is the foremost in his view) invests actions with the capability for unseen, non-mundane results or whether the śāstra merely throws light on the specific capability (sakti) or nature (sva-bhāva) that an object (bhāva, dravya) employed in such an action already has. 61 His acceptance of the second view is indicated (a) by the subsequent placing of that view and leaving it unrefuted and (b) by what he conveys elsewhere, including TK 1.173 translated in $3.1 above:

TK 1.171: sva-bhāvajñais ca bhāvānām drśyante sabda-śaktayaḥ. “And the powers of the words are seen by those who know the natures of things.”

TKV 1.171: te [= sistāh] … dharmādharma-sādhana-bhāvena samanvitām śabda-saktim avyabhicāreņa pasyanti. “The sistas see, without fail, the power of words which can be an instrument of religio-spiritual merit or its opposite.”

In such passages, the sakti or sva-bhāva, which is substitutable with what an entity possesses or displays’ and hence is indistinguishable from “property,’ is spoken of as something bestowed on an object. To the extent it relates to dharma (or adharma), it is spoken of simply as detected or determined by persons capable of extraordinary cognition. The relation of dharma to object properties thus presupposed should hold also in the case of the sāksātkāra of dharma.

$3.4 The understanding toward which we are moving conforms to the role BH associates with the sistas, the learned spiritual elite acting without any vested interest.’ These individuals are similar to the rșis in some respects. In the following passages, they too are spoken of as having a type of acquaintance with the properties of objects that the ordinary persons cannot have:

TKV 1.37, which has the context of cause-and-effect relationship: atha ca tapasā nirdagdha-dosā nirāvarana-khyātayaḥ śistāh pratibimba-kalpena

64The text in question, tatra kecid ācāryāḥ … parāņudyate, a part of TKV 1.172 (pp. 224-225 of Subramania Iyer’s edn), is cited and translated in Aklujkar 2004: 695.

Veda Revelation according to Bharty-Hari

pratyakşam iva svāsu khyātișu samkrāntākāra-parigraham avyabhicaritam sarvam paśyanti. “Then too (i.e., even under such circumstances), the sistas, who have thoroughly burnt (any) detracting elements (that might have once existed in their personalities) by ascetic heat and whose cognitions are free from obscuration, see everything, without mistake — (everything) that has transferred (its) form to their cognitions, as if it is right before them, through the formation of a mental counterpart. “65

TKV 1.171: santi tu … śiştāh sarva-jñeyeşv apratibaddhāntah-prakāśah. “But there indeed) are sistas whose inner light (i.e., power of cognition, intuition including inductive capability) is not impeded with respect to any object.”

TK 3.13.21 in the context of accounting for the difference the grammatical genders constitute: bhāva-tattvadrśaḥ śiştāḥ śabdārthesu vyavasthitam / yad yad dharme ’ngatām eti lingam tat tat pracakşate // “The śistas see the real natures of things. They specify the grammatical genders existing in word meanings as they may be conducive to (the revelation or generation of ) religio-spiritual merit.“46

$3.5 The preceding discussion forms a link with the passages in which pre-BH authors like Caraka and Patañjali speak of pratyaksa-dharman persons. The context of these passages is most commonly that of establishing the properties of things. There probably was a difference of connotation between s-k-d and pratyaksa-dharman but, given the obvious overlapping of the core content direct perception of dharma,’ there must be a conceptual relation, too (S2.10).67 It makes sense, therefore, that the dharma cognized

65If pratibimba-kalpena is taken as an adjective having the sense ‘resembling a reflection, functioning in the manner of a reflection,’ as my first instinct would be and as Vrsabha (p. 93) seems to have done, the problem would be that the sentence contains no noun with which the adjective would connect. Taking kalpa as a noun and taking pratibimba-kalpena as a genitive tat purusa would remove this difficulty. However, then, pratibimba in its literal and usual sense ‘reflection, the image on the other side’ would not make a significant addition to what the sentence conveys with pratyaksam iva. The translation I have attempted is not entirely satisfactory either, as it overlaps with samkrāntākāra-parigraham.

66Cf. Helā-raja on this passage: teşam (=sistānām] ca vastu-paramartha-sākṣātkāritā laksanam. te hi nirāvarana-khyātayo ‘bhidheyeșu samavetam strītvādi lingam abhyudaye yad yad yasya śabdasya sādhanatām eti tat tad eva tasyācaksate.

67(a) From the contexts of the occurrences of s-k-d, the explanations of Durga and S-M reproduced in appendix 3, the analysis offered in $2.10-14, the glosses cited in notes 15 and 17 above and (b) below, and the considerations advanced in SS5.1-3, I conclude that s-k-d connoted knowledge of properties of things based on verification, whereas pratyaksa-dharman conveyed directness of knowledge of properties of things without the specified qualification. Additionaly, s-k-d may have carried the connotation of coming into being at the beginning of creation and (hence?) of being independent of instruction, which connotation pratyakşa-dharman did not have as a part of its meaning.

by the persons referred to by the two adjectives should be the same, namely properties of things.

$3.6 In both TK-TKV 1.5 and TK-TKV 1.173, having a direct bearing on Veda revelation, we learn that the experience of the seers who witness the subtle, eternal and supra-sensual form of language or the mantras is comparable to a dream (svapna-vrttam iva drsta-śrutānubhutam, svapna ivāśrotra-gamyam sabdam).68 A mixture of seeing and hearing is present in both the references. In one of them, the loss of distinction between seeing and hearing is explicitly acknowledged, and the point is underscored by addition of anubhūta ’experienced, which is not associated with any particular sense faculty.

The suggestion of the passages that the revelation experience cannot be ascribed to any particular indriya has been caught by Vrsabha (note 42) and is borne out by TK 3.1.46: jñānam asmad-višistānām tāsu sarvendriyam viduh. “The cognition with respect to those (that is, the universals), which persons superior to us have, is (traditionally) known to be an all-senses (phenomenon).” Interpreting the same remark of BH, Helā-rāja writes: tathā cāgamah “nedānīm indriyair eva paśyati ghrāmataḥ sabdam śrnoti, prsthato rūpāņi paśyati, apy anguly-agreņa sarvendriyārthan upalabhate” iti. “And, thus goes (a statement handed down in) the tradition: Now he (the seer or yogin) does not perceive only with the senses (or only with specific senses). He hears a sound with (his) nose as a means. He sees forms with (his) back as a means. Why, he can access all objects of senses (even) with the tip of (his) finger!”

(b) Durga, under Nirukta 7.23, while explaining Yaska’s statement asāv āditya iti purve yājnikāḥ “This (Vaišvānara Agni spoken of in RV 1.59.6) is Aditya” says: vidhi-manträrtha-vādebhyo yajña-satativam unniyainam yajñam prayogatah prathamam ye cakruh, te pūrve yājnikāḥ sākṣāt-krta dharmāna ity arthaḥ. “The first/earlier sacrificers (referred to by Yaska’s phrase pūrve yājñikāḥ) are those who, having figured out the true nature of sacrifice on the basis of injunctions, mantras and ancillary remarks, performed a sacrifice for the first time; (in other words) the first/earlier sacrificers are sāksāt-krta-dharman ( or the meaning of the phrase purve yājnikāḥ is the same as that of sāksāt-kyta-dharmanah).”

This gloss indicates that the s-k-ds are the ones who try to determine if what they have pieced together from the Veda is borne out by experience — if what they have understood from the Veda really works. The involvement in joining the account gleaned from the scripture with sacrificial performance and in seeking verification (a) of the nature of yajña on the basis of what the Veda actually states and (b) of true yajña on the basis of the outcome shows a spirit of empiricism.

68It might be suggested that the entire phrase svapna-vrttam iva drsta-śrutānubhūtam should be understood as expressing the analogy (“like what is seen, heard or experienced in a dream”). However, in that case vittam ‘happened, took place, occurred’ would become redundant and there would be no object left for ācikhyāsantaḥ. The perceptive commentator Vrsabha (pp. 24 25) is, therefore, right in taking only svapna-vrttam iva as expressive of the analogy.

Veda Revelation according to Bharty-Hari

One possible reaction to such passages may be that they contain nonsense or show the ’typical Indian/Asian’ tendency of being satisfied with mystic experience as an answer to all problems. However, we should not rule out the possibility that if something is said to be known by all senses simultaneously or by any sense whatsoever, the person making the statement may, in effect, be saying that the thing is not known by perception in the ordinary sense of “perception” but is as undeniable as a perceived object would be. A vivid experience inexplicable in the usual way is not necessarily invalid or mystic. It could be an instance of induction.

Theoretical reconstruction

of the Veda revelation process $4.1 With this much of accumulation of evidence and its analysis, we can now attempt a reconstruction of the Veda revelation process as understood by BH and also try to provide a straightforward account, presupposing that we have by now left behind the difficulties of readings, translations and interpretation. Sometimes it is helpful to read an ancient text in its entirety, translate its passages bearing on a concept as literally and precisely as possible and then forget about the problems of translation etc. and reconstruct the various statements in a sequence convenient to oneself and in one’s own words. For this reason, the shoes that have so far traversed the hard rocks of philology should now try to traverse the tilled soil of philosophy.

$4.2 According to BH, there are two kinds of seers at the beginning of creation. Some seers arise as distinct entities at the level of pratibhā and do not get involved with the creation process or the world to which the creation process gives rise. Other seers arise as a multiplicity at the level of vidyā, ‘wisdom’ with a wordly profile or use that does not become corrupt and lose its ability to lead to brahma-prāpti. The self of these latter is associated with a mind but really remains unsullied. It is they who, in unitary sweeps of cognition, perceive the Veda without the involvement of the usual extrovert senses (or, to state the same extraordinary character of the process differently, with the simultaneous involvement of all the senses or with the cognitive limitations of individual senses transcended). Some among them find out what qualities of things are beneficial and what are harmful and under what circumstances. In some cases, they find supporting indications for this in the Veda or āmnāya which they have perceived. With these two streams of information, they compose the Smrtis ’the texts of the recalled, that is, traditionally handed down, knowledge,’ which are useful, positively or negatively, to ordinary people.

Secondly, Veda revelation is a recurring process, just like the created world itself. In fact, it co-occurs with creation (presumably without losing the ablility to occur within creation). Just as creation is thought to move

from an age of absolute purity to increasing impurity and dissolution, the Veda, in its differentiated form, is suggested to be increasingly in danger of dissipation as the creation ages.

In its subtle form, as something identical with brahman itself, the Veda shapes the world. As an evolute in the early phase of creation, it provides a blueprint, readable by the qualified, for arranging social, cultural, religious and spiritual life. It contains the seeds or principles of guidance regarding how things should be and should not be done.

The continuity or link between the cycles of creation is provided by the seers. Some of them remain in direct contact with the first evolute, the pratibhātman (also expressible as sattā-laksana mahat ātman avidyā-yoni), not to be involved in differentiation and diversity, preserving the creative energy (probably for the initial phase of the next round of creation). Others, have or acquire and spread the specific knowledge of the properites of physical objects and actions and, being or having become s-k-ds, participate in the transmission of the Veda as a text in the subsequent creation.69

$4.3 TKV 1.5, which we studied earlier, mainly informs us about the later part of the process. The place where mantradrśah occurs in it indicates that the causal sequence is sāksāt-krta-dharmatva ‘being those who have directly discovered dharma/dharman,’ → sūksma-nityātīndriya-vāg.drktva being those who have seen the subtle, eternal and sense-transcendent speech’ → mantra.dệktva ‘being a seer of mantras.’ When read with TK TKV 1.173 and other less directly relevant passages, it suggests the concept sequence: properties → thing-action relationship (including mantra and Brāhmaṇa) → means of worldly elevation (abhyudaya) and of reaching the most beneficial state (niḥśreyasa).

The wording of TKV 1.5, further, gives the impression that the mantras are a relatively concrete part derived from the s-n-a vāc. The latter is limitless

69 (a) Recall the categorical statement TK 1.30: ysīņām api yaj jñānam tad apyāgama-pūrvakam (“The knowledge of even the seers is preceded by inherited knowledge”) and its justification in TKV 1.30: svābhāvike hi tasmin (=artha-jñāne) prayatnah phalad vyatiricyeta, sva-bhāvataś ca pratyavāyo ‘pi tathā-bhūtah prasajyeta. “If that (knowledge regarding what is good and bad) were natural, the effort (put in) would be delinked from the outcome and there would be the unwelcome outcome that an impediment too would present itself by nature (just as the desired outcome is thought to present itself in the possibility being considered).” Cf. the explanation of caraņa nimittā pratibhā given in TKV 2.152.

(b) A parallel offering indirect support: Samkara on Vedānta-sūtra 1.3.30: īśvarāņām hiranya-garbhādīnām vartamāna-kalpadau pradur-bhavatam parameśvarānugyhītānam supta prabuddhavat kalpāntara-vyavahārād anusamdhānopapattih. “It makes sense that the lords Hiranya garbha etc., appearing at the beginning of the present cycle of creation and helped by the Great Lord (the brahman), be able to reconnect (with the earlier creation cycle), for they deal with the other (earlier) creation cycle, as a man awakened from sleep (would deal with what happened before he went to sleep).”

Veda Revelation according to Bharty-Hari

and has no identifiable or expressible self. The mantras, however, have a specific sequence of phonemes. They are precious stones separated from a vast mine. Therefore, the relationship between them and the bilma is practically one of identity. The latter contains them as an anukāra, “likeness’ and ‘sample,’ of the s-n-a vāc. Consequently, the bilma is also practically the same as the Veda, although it would primarily refer to an undivided and unorganized collection or pile (rāśi) of mantras, while veda would primarily refer to the separated and arranged bodies of mantras (that are like anthologies and have association with practically useful knowledge).70 In this form, the Veda is a tool for the bilma, since its manageable size helps one in approaching the undivided bilma (just as the bilma facilitates one’s approach to the s-n-a vāc). The function it serves in this toolhood is the same as that of the commentandum of Yaska and other Vedāngas (although the latter in historical times would connect with the bilma through the Veda). The logic through which the concepts would then be linked for BH can be visually presented as follows: s-n-a vāc → individual) mantras = (as a collection) bilma → Veda → Vedānga (including the Nighantus and the Nirukta; see appendices 1 and 2 for other related considerations).

$4.4 If, as this essay has proceeded, it has been felt that there are related but different forms of the object of revelation, the feeling is justified. From one angle, it is the subtle or eternal form of the Veda that is revealed. As the revelation takes place, this form (one must assume) expands, acquires contours and gains in perceptibility as mantras or mantra-rāśi. It is not specified if it has a limit or if all of it is made the object of communication by the sk-d rşis, but we should assume that it, though unorganized, has a limit (albeit the individual șși may not be aware of it at the time of witnessing) and that the s-k-ds attempt to communicate all of it.

Whether the individual rsis each perceive the same entity or parts/ profiles of the entity is also not specified, but, again, given the separateness of the hymns and the accounts available in texts like the Bșhad-devatā (or Devatānukramani), we should assume that the latter is the case.

Whether all of the perceived s-n-a vāc, i.e., all of the mantras perceived by the various seers become the Veda is the next question that may arise in our minds. The answer would depend on how we interpret bilma and veda.

70 (a) Carpenter 1995: 41 does not entertain the intermediate bilma stage: “One presumes that the Veda, prior to its fourfold division by the rsi-s, is identical with that True Word itself.”

(b) At this point, I am ignoring the inclusion of texts such as the Brāhmaṇas in the Veda. See appendix 1, point 6 and notes to point 2, for the reason. Also, the practice of applying the same term (veda in this instance) to cover entities which were viewed to be practically or essentially the same needs to be borne in mind. Thus, even if our texts use only mantra, it is possible that their authors meant the inclusion of Brāhmaṇas etc. by implication; cf. comments of Durga and S-M reproduced in appendix 3.

In BH’s statement, it is clear that (a) the s-n-a vāc and (b) the mantras or bilma (as anukāra of the former) are overlapping in content but different. In Yāska’s statement, the mantras and bilma may be in a similar relationship, but there is no mention of s-n-a vāc, and the nature of the relationship between mantra and bilma would depend on the meaning one attaches to bilma in bilma-grahanāya. One could even attach such a meaning to bilma as would preclude the bilma’s inclusion or reflecting of

the mantras (see appendix 3).

At the other end, the relationship between the bilma and the Veda would also depend on how one interprets the compound bilma-grahaņāya. There, from BH’s side, we have a genitival dissolution (bilmasya grahanāya) and the possibilities (a) that the bilma is larger than the Veda (b) that the bilma, unlike the Veda, lacks a specific settled shape (S4.3) and (c) that the Vedāngas, although touching only parts of the Veda from various view points, are essentially its replication. The Veda is declared to be an anukāra in the kārikā, and the Vrtti would not make sense if the word bilma is not taken in a sense like ‘anukāra.’ While Yāska could have held a similar view, it is a historical reality that his commentators Durga and S-M do not present him as subscribing to it (see appendices 3 and 4). As the Nirukta part upadeśāya … vedāngāni ca has not found a paraphrase or echo in BH’s own words, that is, as it was not necessary for BH’s immediate purpose to comment on the specified Nirukta part, we do not know from him as well how he reconciled his understanding of bilma with the words actually employed by Yāska.

In any case, if it is felt that, as we discussed revelation, the object of that process has shifted somewhat, the feelling would be justified. What is said to be revealed has a range formed by the related concepts such as s-n-a vāc → mantras/bilma + Veda.71

$4.5 The Veda revelation process has parallels in what BH puts forward (a) as his concept of language and (b) as his explanation of the process of articulation. The former, in theory, must be viewed as if static entities are stacked up. In that kind of view, the subtle form of the Veda or the

71(a) Holdrege (1994: 52) correctly concludes that the Brāhmaṇas give the impression that the Veda is a closed canon which, in a specific way, is still open to expansion. The fringe area where I may disagree with her is the suggestion in her very last remark to the effect that the Brāhmaṇas were dropping hints of expansion in order to make room for themselves. I consider it more likely that the philosophy behind the Veda samhitā/anthology notion itself left scope for expansion along certain lines.

(b) The possibility that some slippage can occur between what the first group of seers wishes to impart and what the Veda contains is not explicitly denied by Yāska or BH. However, while noting this, we should also note that the two authors did not have any reason to consider the possibility. They did not claim that the Veda known to them was the entire text Veda.

Veda Revelation according to Bhartr-Hari

unrevealed Veda corresponds to the highest form of language, para paśyantī-rūpa or sabda-tattva-brahman, the language-principle itself. The Veda of the first revelation stage, that is, the total corpus of the mantras or the bilma (appendix 1) corresponds to the paśyantī stage, along with parts from that corpus which are memorized (the individual Samhitās etc.; see note 70). Both are reflections of the seer’s revelation experience, albeit a difference of extent and arrangement may exist and the total corpus ‘seen heard-experienced may not be preserved. Both, while being close to the revelation phenomenon, are associated also with the transmission process. In the former aspect, they parallel the steady or at-rest phase of the paśyanti and the active or extrovert phase of paśyantī, respectively.72 When the Veda preserved in the mind or memory is reflected over or is uttered, it appears in the madhyamā or vaikharī stages, but then, it is indistinguishable from other human-made realizations of language (except, of course, for the importance attached to it).

The process of articulation as theorized by BH follows the same model as the one he accepts for language. The only difference is that the constituents of the model are now viewed dynamically — as involved in a process. The Veda in its subtle form then corresponds to the sentience or pure formless consciousness. Its form made accessible to the ordinary persons has an analogue in the paśyantī. That form is a particular assemblage of what could come from a limitless, amorphous source, just as the pasyanti is a storehouse of specific realizations of the boundless language principle capable of taking the form of any language. From this storehouse can emerge

72 (a) TKV 1.159-170: pratisamhịta-kramā, saty apy abhede, samāvista-krama-sakit) pāsyantī. sā calācalā pratilabdha-samadhānā cāvrtā visuddhā ca, samnivista-jñeyākārā pratilinākārā nirākārā ca, paricchinnārtha-pratyavabhāsā samsystārtha-pratyavabhāsā praśānta-sarvārtha-pratyavabhāsā cety aparimāņa-bhedā. “Pasyanti has a limitless (internal) variation. It is an entity in which sequence is withdrawn (i.e., is not overtly present, but it is also) an entity of which the capability for sequence is ever present, although it is not different (from that capability). It is unsteady, steady and (also one from which the ‘steady: unsteady’ transformation is absent, namely) one which is entirely settled (having no need to move from one object to another). It is covered by impurities) and absolutely (vi-) pure. It is one in which the forms of objects to be known are placed, one in which forms merge and (also) one which has no form. It is one in which the reflections of (spatially or temporally) separated entities exist, one in which the reflections of mutually joined entities exist and (also) one in which the reflections of all entities utterly

subside.”

(b) In translating the last triplet in the preceding passage, one could, grammatically, take paricchinna, samsrsta and praśānta as adjectives of artha-pratyavabhāsa instead of just artha. However, in that case the triplet’s difference from the preceding triplet would diminish.

(c) The adjectives pratilabdha-samadhānā, visuddha, nirākārā and praśānta-sarvārtha pratyavabhāsā apply to the higher form of paśyantī, that is, to the language principle.

(d) Carpenter (1995: 47-48) equates the “crucial intermediary stage at which the unity of direct vision intersects with the multiplicity of spoken words” with the madhyamā stage of speech.

a bounded specific text that is the Veda accessible to us, very much like a sentence of the language one accesses in the madhyamā phase. When transformed into sound, the text steps down to the vaikharī level, as any sentence we utter for the purpose of communication does. 73

$4.6 BH’s model of language parallels the one he presupposes for his epistemological account. In fact, the two can be viewed as identical in reality but separated for the sake of convenience in talking about them. The entities presupposed in the Veda revelation process, therefore, have their counterparts also in the entities making up BH’s epistemological model (see apendix 1, point 3).

Beyond this, there is a suggestion in the Veda revelation account of parallelism with deep sleep and dreaming. As naturally as waking up from sleep is, the universe emanates from brahman.74 The Veda, as a part of this emanation, must also be understood as moving from its subtle form to the perceptible form, very much like the signless para paśyanti-rūpa making way for the paśyanti. When the form perceptible to the seers is perceived by them, the experience is very much like that of dreaming, and when the seers wish to transmit what they have perceived, the activity is similar to that of narrating a dream experience (svapna-vrttam iva; see note 62 for other parallels).75

73 Carpenter (1995: 45) has sensed the parallel I point out here. However, I part company with him when he confines the idea to ritualistic life, beginning with the phrase “as a form of dharmic activity.” Writes he: “… the transition from vision to actual utterance is conceived of as the manifestation of the essential form of speech itself rather than the use of speech to express what is beyond it. The primary function of the seers is to cause that essential Word to attain the sequential, temporal form of actual utterance. They do not ‘compose’ the Veda; they’enact’ it. They ’translate’ or ’transform’ it from its unitary visionary state to its temporal, manifested state as a form of dharmic activity, originally the sacred speech employed as an

integral component of the ritual action of the sacrifice. The function of a seer is thus first to see and then to act, to speak and thereby to ‘repeat’ (as is implied in the verb sam-ā-mnā-) or ‘imitate’ (as is implied in the term anukāra) the unitary Word in the medium of actually spoken sounds, in the activity of speech. They function merely to bring about a change in state in the Veda; they are not its authors.’ [Fn 24 at this point: Bhartr-hari says this explicitly at 1.148, where he contrasts the Veda with tradition by describing the former as being like consciousness itself (caitanyavat) and as apauruşeya, literally ‘non-human.’] They are rather its ‘agents.’ They ‘act-ualize’ or ’en-act’ the potencies immanent within the True Word itself.”

74See TKV 2.152 cited in note 54, in which too the references to pratibhā, mahat ātman, and sattā occur as they do in our second passage, TKV 1.173.

75 (a) From the preceding internal connections in BH’s philosophy and the seeds of his revelation account found in the Veda that are pointed out in S$5.11-12 below, it should be evident that the possibility entertained in Halbfass 1991: 48, namely that BH’s ārşa cakşus may be an implicit response to the Buddhist notion of divya cakṣus, need not be entertained. In fact, the employment of divya as an adjective in divya cakşus by the atheist Buddhists indicates that they took the notion from the Vedic tradition or from the popular tradition shared by the Brāhmaṇas and Sramaņas.Veda Revelation according to Bhartr-Hari

Larger significance of

the attempted reconstruction76 $5.1 Philologically, our exploration has led to a Veda revelation view or theory that is textually supported, does not take undue liberties with the meanings of words (especially the meanings of crucial words such as dharma, vāc, mantra and veda) and connects BH’s thinking in one part of his works with the thinking in other parts. Beyond these features which are required by the very theme of the essay, the exploration has, in the last few sections, suggested how BH achieves an impressive economy of theory and gives a very defensible explanation of Veda apauruşeyatva with his Veda revelation account. In my statements, there also has been a suggestion that the account may be philosophical, not just an expression of faith.

Given our present acculturation, our first reaction can expectedly be to see mythology or the realm of the irrational and the untestable in texts speaking of scriptures, persons with extraordinary capabilities, dream like experiences in which the distinction between seeing and hearing is effaced and the beginning of the world from a supra-mundane entity culminating in a merger with the same entity (such features are particularly present in passage 2, which reads like a creation myth). Such a reaction can even have a tone of conviction when we come across a text that speaks of all of these entities and ideas cumulatively, appealing to a source (Purā kalpa) similar to the Puranas and claiming for its scripture and scriptural language the status of the mother of all scriptures and languages.

However, our exploration has resulted in the finding that there is an empirical spirit too in what BH has written (note 67). He speaks of śāstra as simply informing us about the properties of things, not as investing things with properties. His seers parallel the physical scientists to the extent they uncover what is hidden and unknown to ordinary people. The activity that primarily sets them apart is that they connect their discoveries with words in certain received texts and they see in some of these words (when uttered properly by qualified persons) the potential to activate the properties. As the words are sounds or vibrations, the distinctive activity becomes one in which the focus is on the relationship between two physical entities: produced vibrations and the properties of things. There is nothing exclusively or primarily appealing to faith in this feature either. Could it, therefore, be the case that Veda apauruşeyatva and the revelation account associated with it are, at bottom, inspired by a desire to solve or block some

(b) The role of the rsi that Mitchiner (1982: 246-248) establishes with a different purpose in mind is indirectly helpful in understanding some aspects of the revelation process I have reconstructed.

76 The appendices following this section contain some incidental conclusions that hold larger implications.

basic problem in epistemology? Could the myths, nurtured by a tradition that enjoyed riddles and put different ‘spins’ on the essentially same message (e.g., as pointed out in note 58b), be disguised philosophy?

95.2 What is involved in the further consideration suggested by the question just asked is (not conviction or faith but only) willingness to concede the possibility that what we call rationality could have preceded our reconstructed frame of Veda revelation — in particular the possibility that the passages under consideration could be indicative of a point in theorization (carried out by the predecessors of BH and, perhaps, of Yāska) at which a reference to something beyond the terms and axioms accepted in the investigation was felt to be necessary (in words unknown to us, in a way unfamiliar to us and, probably, dimly and vaguely). The interpretation of the passages, justified above on independent grounds, suggests an investigative spirit in the midst of a mysterious-sounding or myth-like talk. The passages make a clear reference to things and properties, express concern with the questions of how the properties can be known reliably and how we can be sure that they are known reliably, and allude to the importance of understanding the nature and role of language properly. The texts we have studied cannot be viewed as totally lacking interest in analyzing the material or empirical world, which we associate with science, or as being entirely innocent of the problem of understanding the human thought process, with which is associated the issue of rational thinking. One is free to consider the texts’ engagement with science or rational thinking as elementary or primitive until evidence to the contrary emerges, but one can certainly not be justified in brushing them aside as nothing but the writings of credulous persons brought up in a tradition dominated by mystical religious teachings.

95.3 The plausibility of a non-mystic interpretation can be argued for in one more way. The word dharma conveys three meanings in BH’s writings:

(a) ’norm, what an individual person is expected to do’ or ‘what the śāstra or āgama advises one to do,

(b) a positive and unseen, that is, non-mundane effect generated by sticking to a norm,’ and

One may view the relationship between meanings (a) and (b) in one of the following two ways: Because the śāstra knows that the effect spoken of in (b) is good for human beings, it advises the way it does. Or,

77In order to be able to avoid a tiresome repetition of “sāstra or agama,” I will assume here that “śāstra” stands for both. For the same reason, I have left out the opposites of “good,” dharma etc. from the following statement. The larger question that is said to arise in the case of dharma should be understood as applicable to adharma, dharmābhāva etc.

Veda Revelation according to Bharty-Hari

because the sāstra advises something, one’s following of that advice must produce a positive effect for oneself ($3.3, note 64).

Either way, there will be appeal to trust or faith and arbitrary blocking of further logical inquiry. The second kind of relationship, namely the investing of an action with capability to produce a beneficial effect because of a śāstra statement, also presupposes that the sāstra knows what is good.78 In other words, both alternatives end up in an unquestioned or uncritical acceptance of the śāstra. They stand in need of giving people the tools that can help in determining which śāstra is good for them (in its entirety or in parts). The discussant faces the question, ‘Why should one empower the śāstra in the first place? What need is there of śāstra? Why should one not stop just with the advice of rationality, empiricism etc.?’ But experience and logical considerations themselves establish (as BH points out in TK 1.31-40) that one cannot stop at rationality, empiricism etc. and that it is a misconception that there is something called pure rationalism or total objectivity.

So, one cannot do without the sāstra, and one cannot accept it simply as an article of faith. What can one do when faced with this dilemma? BH’s solution (and perhaps that of the Vedic tradition in general) can be said to consist of four steps:

(a) Reject the propositions implicit in the preceding statements that knowledge has a beginning at some point and that all circularities can be avoided. Instead of postulating that either the human beings get their knowledge from the śāstras or the śāstras are invested with knowledge by human beings, postulate that an interplay, a give-and-take, between the śāstras and human beings has been going on all the time, that is, wihout any temporal beginning as such. The śāstras have been advising human beings about what is good for them from time immemorial, and human beings, following the śāstra advice, have been discovering truths to add to the śāstras also from time immemorial.

(b) Imagine a theoretical source or beginning for each for ease of statement: Veda for the śāstras and rșis for the human beings.

(c) Think of these sources as co-existing right from the beginning of creation, regardless of whether any real creation took place or not.

(d) Go beyond the language of ‘investing with properties.’ Neither the śāstras nor human beings do that the latter may do so at a later stage in the development of the physical world when they compound the items

78If one takes the position that the śāstra acquires this capability, there arise questions such as ‘From whom did the śāstra acquire this capability?,’ ‘When did it acquire the capability?’ and ‘Can śāstra function like an agent as human beings etc. do?’

that have been given to them, but that is not the level we have in mind here.). Accept the physical world, in itself or in the way it comes to us, as consisting of things that have certain properties.

Here, a thought system’s incompleteness or ultimately not being absolutely autonomous comes to the surface. By the very nature of things, such a system needs unquestioned acceptance of something. The contradiction between starting with insistence on proof for everything and ending up with unproved acceptance of something comes across as inevitable.

$5.4 At the same time, a logic-transcending yet logical explanation of why BH and certain other Brahmanical thinkers could have situated the Veda where they have begins to emerge as Halbfass (1991: 39) senses:

“They [= BH and the great thinkers of the Pūrva- and Uttara-mīmāmsā] invoke this idea [= the idea of the Veda] as a response to epistemological problems, and to the dangers of religious and ethical pluralism and relativism.”

While anticipating and excellently expressing the point toward which I was moving in developing the present essay over the years (note * above), Halbfass has not elaborated upon the point as I would prefer. There is another possible epistemological consideration behind the kind of conception of the Veda BH has accepted and the kind of revelation process he seems to have accepted in its case. To indicate the importance of that consideration, I should, however, first refer to two related questions that Halbfass (1991: 39) eloquently asks:

(a) “Why did they [= BH and the great thinkers of the Purva- and Uttara mīmāṁsā] not face and articulate these problems as such, instead of relegating the answer to a particular text, the Veda? Their reliance on the Veda may be associated with a genuine sense of the limits of human thought and understanding, an awareness of the confusions, the aporias, and the existential and spiritual vacuum human reasoning may produce. Yet the question remains: why did they rely on the Veda, and only on the Veda? Why not on any other kind of “revelation”? Why did they not simply recognize the need for “revelation,” or “objective epiphany,” as such and in general? Are there any truly philosophical reasons, apart from cultural, psychological and ideological motivations?”

(b) “Is the “Veda of the philosophers essentially a fiction and projection?” The first reaction of some specialists of Indian philoshophy who are used to thinking along the lines that have become standard among academics may be to reject the very possibilities implicit in Halbfass’ questions. They may point to the frequently asserted conflation of religion and philosophy and to sectarian affiliations of the philosophical traditions in pre-modern India and declare pre-modern Indian philosophers to be incapable of articulating problems and solutions in non-sectarian terms and/or of entertaining logically or theoretically necessitated fictions. However, just as India has abundance of both religion and philosophy without having a word for either,

Veda Revelation according to Bhartr-Hari

it could have theoretical fictions without having a class name for them.79 The fact that linguists have given credit to India for the zero morpheme fiction and the Indian grammarians have declared the verbal roots etc. to be non-existent in the grammarian-postulated forms) in the object language makes the existence of theoretical fictions in philosophy quite probable. The idiom of ’theoretical/philosophical fiction’ may be missing, but the incidence of ‘articles of faith’ that, upon inspection, turn out to be articles accepted only for the sake of logical necessity or system completion may still be there.

$5.5 In our effort to determine if the ‘fiction possibility can be claimed for the Indian philosophical tradition, let us first note two instances: (a) acceptance of sāmānya-lakṣaṇā pratyāsatti under yogaja pratyaksa to account for the first perception of universals — to meet a part of the need for which induction is accepted in Western philosophy;80 (b) taking the position that universals are not destroyed in the dissolution of the world prior to its re-creation. Religion or spiritual life will not suffer (at least in any direct or convincing way) if the problem of the genesis of universals is not solved or if the problem of the fate of the universals in a mahā-pralaya is left out there ’to dry. Both the problems are primarily and ultimately philosophical problems. Yet, when the philosophers deemed it necessary

791 do not wish to leave the impression that not having distinctive names for religion and philosophy is a failure. My intention is simply to convey (a) that doing X should not be mistaken for being conscious of how (with which label) or why one does X and (b) that absence of conscious doing of X does not imply absence of doing of X.

80 (a) We need universals (real or fictitious) to account for the use of the same word with respect to all members of a class; e.g., we do not use a new word for each chair we encounter. But we cannot be certain about how many instances of the ‘same’ object we need to see before we come up with the notion that the instances are related and form a class. Even if we were to come up with a statistical average, we would not have a ‘philosophical’ explanation of how the inclusion under one name took place. In such a situation, there is no real difference between (a) admitting it to be inexplicable and (b) saying that it is an extraordinary phenomenon — that, at some point in our encountering of (what in the future will be) a chair, suddenly a link between all chair instances or tokens is established and all chairs become members of a class amenable to a single name. Western logic accounts for the phenomenon by admitting induction as a process in addition to deduction in its logical theory. The Indian tradition, particularly that of Nyāya, accounts for the same phenomenon by making room for yogaja pratyakşa, specifically its sāmānya-lakṣaṇā pratyāsatti variety.

(b) Although the following remark of Kaiyața is made in the context of Patañjali’s use of pratyakşa-dharmāṇaḥ, it corroborates the point made here: yogaja-pratyaksena sarvam vidilavantah.

(c) Awareness of universals has been ascribed to asmad-visistas (TK 3.1.46), the cognitively extraordinary persons, by BH, very much in the manner of ascription of Veda revelation to the sșis.

(d) Every use of the word yoga should not be thought of as landing us in the realm of the irrational or of mysticism. An element of what Indian philosophers like BH would call antah prakāśa, ‘internal light, a (sudden flash-like) realization from within,’ (TKV 1.135, quoted in $3.4) is present in the notion of induction, too.

and economical to adopt the positions they could neither prove nor disprove, they did adopt them in a manner hardly distinguishable from a believer’s manner of accepting certain ‘religious’ truths.

$5.6 It will be appropriate at this point to note the following statements showing awareness of the limit of philosophical thinking or of the unavoidability of acceptance of (what we would call) a priories:

(a) TKV 1.30: sarve ‘pi hi vādino dūram api gatvā na sva-bhāvam viativartante. “No party to a discussion can go beyond the own nature of a thing, no matter how far they go (in pursuing an issue; i.e., all parties to a discussion are forced at some point in the discussion to take the position “things are what they are’).”

(b) TK 3.1.95: sva-bhāvo ‘uyapadesyo vā sāmarthyam vāvatisthate / sarvasyānte yatas, tasmād vyavahāro na kalpate // “At the end of all (that a philosopher proposes or an exploration of cause leads to) an unnamable nature or capability (of a thing or postulate) remains (i.e., the discussion or investigation boils down to ’this is the way things are; we cannot identify anything more fundamental or enlightening’). Since (what the preceding sentence states is a fact of life), therefore, there is (ultimately) no (purely logical) enabling for communication (i.e., it takes place, but we cannot explain how it takes place without accepting some notions as unquestionable givens). 81

(c) TK 3.6.18: caitanyavat sthitā loke dik-kāla-parikalpanā / prakrtim prāṇinām tām hi ko ’nyathā sthāpayisyati // “In the world, the conception of time and space is as deep-rooted as that of one’s own) being alive. Who can change that conception, which is the very) nature of living beings!”

(d) From the Kevalādvaita tradition (source not specified in the Nyāya kośa or by the teacher from whom I heard the verse): jīva iso viśuddhā cit tathā jīveśayor bhidā/ avidyā tac-citor yogah şad asmākam anādayah // “For us (i.e., in our philosophical school or system), six things have no beginning (do not ask when they came into existence; that would be an inappropriate question; these six things are:) (i) individual self, (ii) god (postulated as

81(a) Helā-rāja takes tasmātas standing for avyapadeśyāt (qualifying) sva-bhāvāt / samarthyāt). This forces him to ignore yataḥ. In any natural construing of the verse yatah and tasmāt should be related, even if English may not allow us to include the literal translations of both the words in the sentence. There is no masculine or neuter noun in the preceding verse to which tasmāt can refer. Therefore, I have taken it as referring to the content of the proposition sarvasya ante avyapadesyaḥ sva-bhāvaḥ (avyapadesyam) sāmarthyam vā avatişthate.

(b) The precise meaning of the seemingly simple words vyavahāro na kalpate is not easy to determine. It would depend on which of the related but different senses of vyavahāra and klp one chooses and what one expects BH to say in the context. I have taken klp to be enabled, to become able’as meant in the sense of logical accounting or justification and vyavahāra’exchange, transaction’ as applicable to the arena of language.

Veda Revelation according to Bhartr-Hari

brahman’s first evolute that manages the creation to follow as note 69b bears out), (iii) pure consciousness, (iv) distinctness of the individual self and the (managing) god, (v) nescience and (vi) the association of nescience and pure consciousness.”

$5.7 That BH was aware of the need, felt by the different philosophical traditions, to allow unfalsifiable or logically untestable entities at the highest or most fundamental level is corroborated also by the following kārikā-vrtti continuum:

TK 1.38: atīndriyān asamvedyān paśyanty ārşeņa cakşuşā/ye bhāvān vacanam teşām nānumānena bādhyate // “A statement of those who, with a seer’s eye, see things that are beyond the senses and cannot be felt is not invalidated by inference.”

TKV 1.38: antar-yāmiņam, aņu-grāmam abhijāti-nibandhanam, anabhivyaktam sabda-brahma śakty-adhisthānam, 82 devatāḥ, karmaņām anubandha-pariņāma-sakti-vaikalyāni, sūksmam ātivāhikam śarīram, pythag anyamś ca tirtha-pravādesu prasiddhān arthān, rūpādivad indriyair agrāhyan, sukhādivac ca pratyātmam asamvedyān ye śistā vyāvahārikād anyenaiva cakşusā mukta-samśayam upalabhante, teşām annumāna-visayātītam vacanam vyabhicāribhir anumānair apākartum asakyam.” “A statement of those sistas, who, through an extraordinary eye, grasp entities such as the ones specified below) with total certainty cannot be set aside by using inferences (because what the sista statement pertains to) falls beyond the domain of inferences (and the inferences) are prone to deviation. (The entities meant as examples bearing out the preceding observation are:) the inner controller; atom groups which form the basis of creation; unmanifest Word Principle that is the seat of (various) powers; deities; impressing, maturation and dissipation of actions (saṁskāra formation etc. in the case of individual selves), the subtle transmigratory body, and (similar) other entities well known in various philosophical exchanges — entities which cannot be (perceptually) grasped as form (or color) etc. can be and which cannot be felt inside like happiness etc.”

Here, the fundamental entities admitted in several different schools or the final causes or supreme truths as advanced by different schools are said to have been perceived through an extraordinary eye, functionally parallel to intuition, insight and induction. It is noted that they are not accessible to sense experience like the features of external objects or to

82I have followed Vrşabha’s commentary in taking certain nouns as qualifiers of the nouns next to them. However, I do not rule out the possibility that abhijāti-nibandhanam and sakly adhisthānam were meant to be independent nouns standing for entities at higher or deeper logical levels in schools such as Jainism and Mīmāmsā. We cannot be certain about our understanding until similar passages or the sources utilized by BH become accessible.

feelings like happiness etc. and that one cannot expunge them on the strength of logic alone.

$5.8 On the strength of the evidence given so far, we may view BH as a philosopher capable of opting for theoretical fictions when necessary. He might have accepted such fictions for purely logical reasons accompanied by “cultural, psychological and ideological motivations,” if I may repeat Halbfass’ phrase.83 Two questions then present themselves: Would BH do so in the case of the Veda? If he were to do so in the case of the Veda, why would he do so?

There is no indication in the words BH actually uses to the effect that his acceptance of the Veda as authority or source is anything less than heartfelt and genuine or that the Veda is something to be invoked just to make up for a desideratum in one’s philosophy. This justified impression, however, does not rule out two possibilities: (a) a tradition of thinking of the Veda as useful beyond its historical context and literal form had become established before BH’s time, making it easier for thinkers to lean on the Veda without having to defend their leaning or without realizing every time that this is what they were doing. (b) There was something in BH’s philosophy that made it particularly natural for that philosophy to seek a complementation in the Veda and to feel no need to state explicitly that a complementation was being sought.

$5.9 The sākhādi-bheda-bhinna apauruşeyatva of the Mīmāṁsā was known to BH (appendix 1, points 4-6). It has the flavor of fiction. Further, a dehistoricized view of Vedic texts, particularly of the RV, is likely to have existed for a long time. Something viewed as appearing at the dawn of creation or in an inaccessible past even by an ancient author like Yaska (cf. note 44) is unlikely to have been gathered and preserved exclusively or mainly because it, in some sense, contained history, even if that history pertained to religio-spiritual life. The activities of collection and preservation of mantra or hymns are, in fact, more likely to have been preceded by a philosophy or theory of religio-spiritual life that delinked such life from (what we would call) history or at least de-emphasized the importance of history to that life.84 The reliability of a religio-spiritual teaching was probably

83The space for ideological motivations, however, seems small in BH’s case. An evenly accommodative philosophical temperament is writ large over his works.

84 The observations made in this section hold implications for the efforts that are made to recover the most ancient history of India and of the speakers of Indo-European languages on the basis of the Vedic texts. While such efforts should continue to be made, they need to be made with the awareness that we may be trying to recover history from texts that were(a) either chosen because they contained no or few historical clues or (b) were edited, as far as possible, to remove historical clues. Also, we need to be aware that whatever history can reasonably be said to be reflected in the texts may pertain to a period which had ceased to be historical —

Veda Revelation according to Bharty-Hari

not to be made dependent on whether its articulator actually lived or how he lived. Such a surmise would accord with at least two curious absences or seeming incongruities in the Vedic tradition:

(a) This tradition has kept a record of the names of persons to whom the hymns belong. It has attached importance to remembering the names. In several cases it has also preserved the traditional stories about how the hymns came into being. It has not hidden the details that might indicate that the canon grew over time (either in the form of genres such as the Samhitā, Brāhmaṇa etc. or in the form of sākhās). Yet it does not accord a greater value to a seer’s teaching because that teaching was imparted in an older period or was closer to the time of the original revelation. Nor does it make efforts to resolve the differences in teaching by comparing the available variant readings and deciding which one of them could be older and hence more likely to be genuine. On the contrary, accommodation of the variation in detail is sanctioned in statements such as sarva-śākhā pratyayam ekam karma (TKV 1.6, probably quoting from Sabara 2.4.8, 9, 30 or 32) and sarva-vedānta-pratyayam (Vedānta-sūtra/Brahma-sūtra 3.3.1).85

(b) Yaska gives us a ‘history of how the mantras or the Veda came down to him: seerhood qualified by unmediated knowledge of dharma → mantras → upadeśa → Veda and Vedāngas. But he refuses to go along with the Aitihāsikas, the Legendarians or Mythologists, who may occasionally give us some history because legends and myths do in some cases arise out of historical facts. He also offers nirvacanas or niruktis for which being historically justifiable is evidently not the concern. They are in fact context

which was distant enough and had no political, economic etc, implications left — for the gatherers of the Vedic hymns.

85This apparent incongruity was pointed out by Muir as early as 1874 (= reprint 1967, p. III. 210), albeit with a different purpose in mind. His attempted refutation of the Brahmanical/ Hindu acceptance of the authority of the Veda is still heard in different words from the proponents of Christianity. The refutation rests on discounting the possibility that a tradition could deliberately decide to de-emphasize history. It reads thus: “The same (as in the case of the ancient view of the origin of the river Nile] might be said of the Indian speculators, who argue that the Veda must have had a supernatural origin, because it was never observed to have had a human author like other books;—that by thus removing the negative grounds on which they rest their case into the unknown depths of antiquity, they do their utmost to place themselves beyond the reach of direct refutation. But it is to be observed (1) that, even if it were to be admitted that no human authors of the Vedas were remembered in later ages, this would prove nothing more than their antiquity, and it would still be incumbent on their apologists to show that this circumstance necessarily involved their supernatural character; and (2) that, in point of fact, Indian tradition does point to certain rishis or bards as the authors of the Vedic hymns. It is true, indeed … that these rishis are said to have only “seen” the hymns, which it is alleged) were eternally pre-existent, and that they were not their authors. But as tradition declares that the hymns were uttered by such and such rishis, how is it proved that the rishis to whom they were ascribed, or those, whoever they were, from whom they actually proceeded, were not uttering the mere productions of their minds?”

fitting meaning determinations, made or recorded by Yaska’s predecessors in many/most instances, which are to be explained by following the technique of the grammarians (not necessarily Pāṇinian grammarians). They may in some instances coincide with what we call (historical or true) etymologies, but whether they do so or not would not have mattered to Yāska.

$5.10 Additionally, the surmise in $5.9 is indirectly but substantially supported by the following facts or independently reached conclusions:

(a) Halbfass 1991: 41 “The Veda itself exhibits a paradigmatic commitment to an absolute origin and foundation, and seems to provide clues for its own later role in Hindu thought. It has its own retrospective and reflexive dimension and refers back to the Rg-veda as its center and source.”

Particularly relevant in the present context is Halbfass’ first sentence. A text conscious of its role to come is unlikely to bind itself tightly or obviously with historical strings. 86

(b) The tradition of interpreting the Veda at different levels — adhi yajña “ritualistic,’ adhi-daiva ‘mytho-theological and adhyātma ‘spiritual, metaphysical, philosophical — is at least as old as Yāska’s time (the terms themselves are attested in the oldest Upanisads Brhad-aranyaka and Chandogya). As I hope to point out in a separate publication, the Parisista part of the Nirukta, in which this three-pronged approach is evident, has not really been proved to be late, although claims to that effect have been made.87 To judge from its language, the Nirukta belongs to a period that was an extension of the period of Brāhmaṇa composition. Its awareness that the Veda can be amenable to adhi-yajña, adhi-daiva or adhyātma interpretation is a fairly early awareness, justified to a considerable extent by the contents of the Veda anthologies. 88 The wide scope thus given at an early date to varied interpretations strengthens the probability that the Veda has been used in innovative, as distinct from literal or historical, ways for a long time before BH.89

86As religions usually gain much by emphasizing history (they can then arrange more rites, ceremonies, festivities etc. and remain in greater contact with the followers), it is quite unusual, especially for a culture unabashedly having a large number of rituals, worships etc. in its immediately following periods, to attempt to place its premier scripture beyond history. Such a move is unlikely to have been made without much thought and planning.

87In a forthcoming paper dealing with the semantic history of the term vedānta, I have observed how, in the ancient Indian perspective, the functions expected of the Veda must have included a metaphysical or spiritual-philosophical function from an early time (in fact, right from the start). See also my other forthcoming paper “Unity of the Mimāṁsās.”

88 The implication of the awareness is not that each mantra can be or has been interpreted in three ways.

89An incidental but important implication of how I have so far specified the larger significance of the present study is this: Available research has much to say about historicalVeda Revelation according to Bharty-Hari

85.11 It would thus be not only an inoffensive but also a familiar move in BH’s days to utilize the Veda as a theoretical fiction. In addition, BH has something in his philosophy that would make it natural for his philosophy to be very close to the Veda. It is a philosophy mainly concerned with speech and thought. There are references to speech and thought in parts of the earliest Veda.90 The effectiveness of speech, whether it be of the composer’s community or of the ‘other’ community, is presupposed several times (a wish for the greater efficacy of one’s own speech and non efficacy of the speech of the other community is expressed or implied). The Veda shows its fascination with language also in the language of its poetry, which, besides figures of speech such as simile, includes riddles, puns and shocking statements reminiscent of what the later Hindi tradition has called ulațavāṁsī style.

Secondly, word and meaning relationship and hence language has no beginning in BH’s thought. The argument behind this position is that even to establish a relationship between the first word in history and its meaning a community would need language. 9 A text, therefore, must always exist at least in the form of a single sentence (which, in turn, must be preceded by at least the nebulous text of community living or culture, for sentence formation without some concept linking is inconceivable). The Veda is also a text that has always existed (according to the view or belief of

layers within the Vedic anthologies (particularly the RV books), but it seems to have given very little thought to how the anthologies themselves were made. As far as I could ascertain, not many attempts to answer the following questions have been made: How were the various parts collected? What were the criteria for their selection? What ends beyond the conveying of certain linguistic meanings were the texts or anthologies intended to serve? Existing research, as far as I am aware, leaves the impression that the anthologization was a largely haphazard process mainly guided by practical convenience and shaped by inevitable accretions, generally going through the same stages as are found in the histories of other religious canons. Without denying that the stages could have been broadly similar, I would like to observe that research in the future needs to explore also the possibility that the Vedic anthologization had a specific, perhaps distinctive and sophisticated, philosophy of religious and socio-political life behind it. Remarks like Halbfass’ (1991: 19) “The Veda itself frequently presents itself as a cosmic or cosmogonic reality” need to be pursued seriously and studied in depth.

90 (a) Cf. Padoux 1990: x-29, in which the existence of a speech philosophy from the time of the earliest Veda is asserted with clear evidence several times, the continuity of that philosophy until the time of the later Upanișads is perceptively established and references to earlier allied explorations by K. Madhava Krishna Sarma, Louis Renou, Otto Strauss etc. are provided.

(b) Tripāțhi 1976 provides a collection of Vedic pronouncements on speech.

91 The implications are: (a) Even if there were to be a biological beginning for language, a philosopher would not be able to make use of it; he must proceed by assuming language to be beginningless. (b) Individual languages should be thought of as forming a continuum, not replacing each other completely. (c) If a ‘Tower of Bable’ type account were to be written by an Indian philospher like BH, it would not mention dispersion in such stark terms as loss of all contact between languages.

certain theoreticians or systematizers). Therefore, there is economy in identifying the theoretically needed, conceptual or abstract, ever-existing text with the Veda. The language of the Veda then can perform the same role as a postulated first language would, and the ultimate or subtle form of the language of the Veda can be equated with the language principle itself.92

$5.12 Obviously, something special must have been seen in the Vedic hymns before their preservation was deemed necessary. That this something special had an aural or sound aspect is evident from the fact that so much effort was made to preserve even the accents of the words in the hymns. Further, that the nature of the special thing was close to what BH says can be deduced in a somewhat indirect way: The elements of BH’s thinking that play a special role in the present context (s-n-a vāc, dharma, mantra/ Veda, and creation or revelation in a very distant past) are found to an extent in Yāska’s Nirukta. But they are not confined to that text. They appear with essentially synonymous expressions and different configurations in a number of diverse sources in a scattered or incidental way such as the Brāhmaṇas, Aranyakas, Upanişads (note 90), Epics and Purānas, suggesting a broad base, which in turn, increases the probability of their being present in the more ancient period of Indian history.

As a particularly informative passage Mahābhārata 12.224.55 and 12.671.1 may be noted:93 anādi-nidhanā nityā vāg utsystā svayambhuvā // ādau vedamayī divyā yataḥ sarvāh pravịttayah // “At the beginning was issued, by the self-born (brahman/Brahmā), the eternal (or latently existing) speech, which consists of the Veda, which is divine, which has neither a beginning nor an end (and) from which all proceedings (initiatives or actions) come about.” Here, the connection between creation, Veda and speech is evident.

Since the rșis are intimately connected with mantra and dharma, the following passages which support the detail that certain seers are born at the beginning of creation should also be taken into account: Madhyandina Sata-patha Brāhmaṇa 6.1.1.1 (quoted in TKV 1.2; cf. Vrsabha p. 31): asad vā idam agra āsīt. kim tad asad asīt. rsayo vāva te ‘gre tad asad asīt, ya rşayah

92 I cannot cite passages from BH’s works that explicitly state the content of “there is economy… the language principle itself.” However, the identifications in appendix 1, point 3, are certainly there in BH’s works. The reasoning they would require cannot but be in terms of theoretical economy: ‘If an entity with a different name cannot be proved to be bodily different, then why should one not think of it as the same entity?’ Lāghava (’economy of postulates and/ or reasoning steps’) was a recognized criterion among India’s śāstra authors.

93 Although, as indicated by the numbers here, the original is found in the critical edn with its halves separated from each other, there can be no doubt that the halves need to be read together as in some ms traditions and citations by pre-modern authors. In their present places in the critical edn, the halves do not connect well with what follows and precedes.

Veda Revelation according to Bharty-Hari prāņā iti. Svetāśvatara Upanişad 5.2:94 rșim prasūtam kapilam … tam agre … Gauda-pāda on Sāmkhya-kārikā 43: bhagavataḥ kapilasyādi-sarga utpadyamānasya catvāro bhāvāḥ sahotpannāḥ dharmo jñānam vairāgyam aišvaryam iti. Yukti-dipikā on Sāmkhya-kārikā 69 (Wezler-Motegi edn p. 267): bhavāgrotpannair api sanaka-sanātana-sanandana-sanat-kumāra prabhytibhir … . Vrsabha p. 225.21, explanation of abhikhyā occurring in TK 1.173: brahmaiva ?și-rūpeṇa vivartate iti khyātam/vyaktam; also, the passages collected and translated in Muir 1868, pp. I.36, 64, 65.

$5.13 In the initial section of this essay I wondered about the extent to which Renou’s remark about the ’tipping of the hat’ treatment of the Veda was applicable (and suggested the comparability of the position accorded to the Veda with the position assigned to a matha head in Kulkarni’s profoundly disturbing story). It was not my intention to determine the extent of applicability. I leave that difficult work for the community of scholars to undertake if it so wishes. However, I hope that I have succeeded in establishing at least the possibility that the seeds of the phenomenon so succinctly captured by Renou were sown long before the phenomenon is thought to have come into existence — that the sowing may go back to the time the Samhitās were put together. It is likely to have been preceded by considerable contemplation on the human condition, particularly on the type of authoritative text societies need for proper functioning. I have also put forward evidence and reasoning to the effect that, in authors like BH, who were well-versed in the Veda, the acceptance of the Veda could have been a consciously adopted and logically justified fiction that co-existed with a believer’s reverence for the Veda — that the situation was more complex than the dichotomy Renou’s remark presupposes. If and how the ’tipping of the hat’ phenomenon influenced the post-BH periods of Indian thinking (e.g., that of Kashmir Saivism or of the modern mystics such as Ramana and Aurobindo) is a determination to be made by other researchers (such as the ones whose writings appear in Patton 1994).

94 Larson 1987: 109: “When this reference is compared with other Svetāśvatara references, namely IV.12, VI.1-2, VI.18, and III.4, it becomes clear that kapila is to be construed with reference to Hiranya-garbha and Rudra.”

APPENDIX 1 Summary and supplementation of Aklujkar 1991a (“Bhartr-hari’s concept of the Veda”)

Vertical double reference of veda; (a) the subtle, original form of the texts we know as Veda, which form is also called pranava (probably when it is in the mode of creating or engaging with the world; cp. Sāṁkhya use of pradhāna). (b) the texts themselves, either in the bilma/rāśi form or as the Samhitā(s) etc.

The Veda referred to in (a) is, as it were, at the top or the source node, with the second as its descendent. The order is primarily logical, although, when we speak of it, it may come across as chronological.95

NOTES TO Point 1: (a) Here, the word “text” does not primarily stand for a written composition. Nor does it necessarily stand for a text in a particular book chapter-section etc. order that the designations like Rg-veda, Yajurveda, Sāma-veda and Atharva-veda have come to connote or the designations like rc, yajus and sāman, taken to stand, respectively, for the verse, prose and ‘sing-mode’ compositions, may convey. Being in a specific word order and/ or a sentence order is all that texthood implies in the present context. One may think of such ordered linguistic matter as leading to the distinct text-bodies or collections like the RV right from the start, but one is not logically obliged to do so in the present context.

(b) As evidence of BH’s acceptance of the subtle form, note the phrases under (d) in point 3 below. Note also that in TKV 1.5 (reproduced in point 5 below) the darśanātmani sthita drsya artha, that is ’the entity in the vision-unit,’ comes before the labdha-krama vāg-ātma-rūpa ‘sequential form consisting of language’ and samhitā-pada-krama-vibhāga division into the text modes Samhitā-pātha, Pada-pātha and Krama-patha.’96

(c) The subtle single Veda meant here is different from the unity it may have as a rāśi ‘pile, mass, collection’ before Krsna Dvaipāyana divided

95See point 7 below for some consequences of Halbfass’ not identifying the two related but distinct referents of veda in the way I have identified them.

96 In a remark echoing eko ‘yam vedākhyo darśanātmani sthito dršyo ‘rthah, Helā-rāja (TK 3.1.46) observes: visista-racanăvata eva vedasyeśvara-buddhau darśanātmani sadāvasthitatvam. However, the elements visista-racanăvatah, īśvara-and sadā of the remark are not supported by the relevant passages of BH. In other words, Helā-rāja’s attempt cannot be said to be historically valid in the present state of our sources. BH does not accept īśvara in the sense of the supreme lord’ or ‘personal godhead.’ The eternal Veda is subtle, not a particular sequential text, in his thinking. He leans toward thinking of the text Veda (in any Sākhā form) as a later or second stage development.

Veda Revelation according to Bhartr-Hari, appendix 1 it into the four major Vedas (as a traditional account or legend says) and also from the unity the Mīmāmsakas presuppose for their hermeneutical endeavor in remarks such as kytsno veda ekam vākyam “All the (texts in the) Veda form one sentence (with no discordant constituents) " or sarva-śākhā pratyayam ekam karma “All Veda-branches inform us about one (and the same) ritual act (although there may be differences of detail in them when they speak of that particular ritual act; i.e., the branch differences pertaining to what is essentially the same ritual act are not to be taken as calling for performance of separate ritual acts; the differences are to be reconciled).”

Horizontal double reference of veda: (a) a narrow or specific meaning: revealed or heard literature (Sruti) only; more probably, the mantras only and, less probably, the mantras and Brāhmaṇas, with the Upanişads either being viewed as parts of the Brāhmaṇas and Mantras (and thus as Sruti) or not being viewed as Sruti at all (that is, viewed only as recasts of or comments on the Sruti — as Trayyanta, which may not be exactly synonymous with Vedānta).

(b) a wide or general meaning:

Variety (i): the Upanişads are a distinct part of the Veda. Variety (ii): all vidyā-bhedas or lores are Veda, because they

consist of words and meanings, which, in turn, are

reducible to praņava. That in the case of some early Upanişads the division into Brāhmana and Upanişad is not clear is well-known.

We may speak of variety (ii) as assigning the widest sense to veda. However, while doing so, we should not overlook that it is based on the subtle form of the Veda, that is, on the non-text Veda. It can potentially include any number of branches of knowledge and not be subject to any question presupposing a specific extent. In this respect, Aklujkar 1991a; note 42 and the following remark of Eliot (1921: 1.76), extendable to the Samhitās, Brāhmaṇas and Aranyakas, are relevant: “According to Indian ideas there is no a priori objection to the appearance now or in the future of new Upanishads. All revelation is eternal and self-existent but it can manifest itself at its own good time.”

NOTES TO POINT 2: (a) The evidence for the inclusion of Brāhmaṇas in the reference range of veda or śruti is indirect and scanty in BH’s works as Aklujkar 1991a: $2.3-6 indicates.

(b) We usually use the words brāhmaṇa and upanişad for books or titled next continua. However, there is also another usage in which these

words denote passages or content serving specific functions within the Vedic thought complex. These passages or contents may not, in all cases, be found in the existing texts having the same genre title, but would probably have been thought to be eligible for inclusion; cf. the usage of brāhmana in the following quotation made by Vrṣabha (1.148 p. 203) that, in turn, quotes a passage which agrees with Chāndogya Upanişad (also called Chāndogya Brāhmana in the tradition) 8.12.1 entirely in thought and almost entirely in wording: tasya sāngāmś caturo vedān āvartayataḥ kramenedam brāhmanam ājagāma “To him, who was repeating the four Vedas along with (their) ancillary texts, this brahman-associated text (= Chāndogya Brāhmana/ Upanişad 8.12.1 quoted next) occurred in due course”;97 also, the usage of upanişad in Brhad-aranyaka 2.1.20, Chāndogya 1.13.4, Kena 4.2, Svetāśvatara 1.16, Taittirīya 1.3.1 etc.98

There seems to be only one occurrence of brāhmana in the surviving part of BH’s scholarly bequest in a sense other than ‘a specific varņa or social class’ or ‘a member of a specific varna’. This occurrence is Mahābhāşya tīkā, Abhyankar-Limaye edn p. 30: yathā brāhmana-sabdā abhyāsa-kāle ’narthakāḥ. para-pratipādane sva-rūpa-padārthakāḥ “evam pathaivam pathe’ti. viniyoga-kāle ‘rthavantah. “This is) like (the following): The Brāhmaņa expressions do not convey meanings when one is reviewing (i.e., is reciting them for memorization); at the time of imparting (them) to others, the meaning they have is their own form (i.e., they are mentioned, not used, as is borne out by the instruction) ‘recite thus, recite thus’; (however) at the time of employment (in a ritual act), they have (the usual denotative,

970n 02 May 1997, Dr. Elliot Stern drew my attention to the fact that a passage strikingly similar to what Vrşabha quotes with evam hy āha occurs in Vyoma-siva’s tīkā on Padārtha dharma-samgraha (Chaukhamba Sanskrit Series edn. p. 19.23-26). In my response of 06 May to him, I observed (a) that we cannot be certain that Vyoma-siva is Vrşabha’s source, although the possibility of his being the source is not to be discounted, and (b) that the word brāhmaṇam in the poorly preserved Vyoma-siva passage stands for the text cited or content reproduced next, which has the apearance of explicating the hidden significance of a brahman ‘an earlier Samhitā statement, a Vedic mantra/prayer,’ not for a Brahmin person. The texts we call Brāhmaṇas could easily have been first thought of as collections of brāhmaṇas in such a sense (very much like sūtra standing for an individual aphoristic statement as well as a work consisting of such statements). In his 2004 publication, Professor Johannes Bronkhorst incidentally notes in fn 63 that a similar meaning of brāhmana (“single fomulations rather than whole texts’) has been suggested by Professor Walter Slaje in the context of yājñavalkāni brāhmaṇāni occurring in Patañjali’s comment on the Vārttika purāņa-proktesu brāhmana-kalpesu (Pāṇini 4.3.105.). Compare the usage of Durga and S-M in the passages from their commentaries cited in appendix 3 below.

98 The passages referred to here establish that even within the text bodies we call Upanişads there is usage of upanişad in some such sense as ’truly effective/transforming knowledge, precious proposition,’ implying thought/information not to be indiscriminately spread.’ This usage continues even after the word comes to be associated with a particular body of texts as the Artha-śāstra, Kāma-sūtra etc. indicate.

Veda Revelation according to Bharty-Hari, appendix 1

non-self-referential) meanings.” Here a text which is employed in ritual worship as something settled and around which a tradition of learning has grown is obviously meant, but the passage is opaque as to whether textual wholes are meant or shorter brahman-associated pronouncements and revelations that could be incorporated in the textual wholes are meant.

As for upanişad, the word does not occur in BH’s available writings. However, his most ancient commentator accessible to us writes in a way in which there is a suggestion that upanişad is a function-based term like artha vāda. In explaining the TKV phrase vedākhyasya prasiddhasya brahmanah “Of the well-known/ established brahman called Veda,” Vrsbha (p. 39) writes: loke sārthavādakasya sopanișatkasya brāhmaṇasya mantrāņām ca vedākhyā. “In the world, the word veda is used to refer to the Brāhmaṇa (a text/text-complex), inclusive of artha-vādas and upanişads, and the Mantras.”

(c) If the text bodies we now call Brāhmaṇas and Upanişads are included, the bodies should be understood as not extending beyond what we, as historians, generally consider to be the older Brāhmaṇas and Upanişads. BH himself may not have made any distinction in terms of old and new, but it is justifiable to presume that some of the works that now pass as Brāhmaṇas or Upanişads either did not exist in his time, were not known to him or were not known to him as Brāhmaṇas or Upanişads.

(d) Many Indologists state or presume that the Samhitās first came into existence, then the Brāhmaṇas, and so on. The better ones among them are aware that this sort of temporal reconstruction is only generally true; it does not mean that every part of a Samhitā is older than every part of a Brāhmaṇa, and so on. The better ones among the better Indologists, further, do not make the mistake of taking the lateness of language as proof of the lateness of content or thought. It is important to be aware, however, that, for a thinker coming from a milieu like that of BH, the text bodies the Indologists (and, because of them, most modern scholars concerned with the issue) see as developing over time, were a synchronic or atemporal text complex. If this thinker privately thought of one text as coming before or after another, his theoretical thinking was rarely, if ever, based on that possibility.

Place of the subtle form of the

Veda in BH’s philosophy: (a) vaikharī language-1 = speech. (b) madhyamā language-2 = jñāna cognition. (c) paśyantī language-3 = buddhi mind or intellect. (d) param paśyantyāh rūpam, = citi sentience = chandasām yonih,

sūkşmā nityā atīndriyā vāc,

chandomayī tanuh, vācah uttamam rūpam,

vedākhyah darśanātmani sabda-tattva-brahman

sthitah dysyah arthah, language-principle, language-4

vedah … prakstitvena … vivartesu … vidhātā, pranavah, subtle form of the Veda

NOTE TO POINT 3: The remembered Veda or the Veda performing the function of a teacher (vedah… upadestr-rūpatvena … vyavasthāsu … vidhātā; note 36) could be a part of the paśyantī in the case of those who know the Veda text. However, it would not be the totality of paśyanti.

Different views regarding how Veda as a

specific textual body is revealed: (a) Vrsabha on TKV 1.6 p. 27: 23-24: ayam vedo brahmanah sakāśāt sākhādi-bheda-bhinna eva vivartate. “This Veda appears from brahman in no form other than eva) the one which has the division based on the mutual difference of the branches etc.”

(b) eka-rupa eva vivartate. (anantaram] pravibhaktah. “(The Veda) appears as having one (undivided) form. It is divided (later). “99

NOTE TO POINT 4: As BH speaks later in the same Vrtti of a view according to which the division of the Veda into branches is made again and again, with interruptions in between, (yeşām tv ayam sākhā-pravibhāgo vicchedena punah punar bhavati …), it follows that the views stated in the earlier part of Vrtti 1.6 must be of those who think of the Veda divisions, not just of the Veda, as eternal.

This view should perhaps be distinguished from the view stated in TK 1.172 and at the beginning of V1.173 (the view of the Mīmāṁsakas according to Vrsabha p. 226: 9-10). In that view too, the Sruti with its divisions and variations is eternal (and so is the Smrti as far as its content is concerned). However, there is no revelation, at least as a cosmogonical event.

99To provide exact contrast with the preceding, I have made up the two sentences in (b) utilizing BH’s own diction. BH’s acceptance of the view they express is established in point 6 below.

Veda Revelation according to Bhartr-Hari, appendix 1

Ways in which the subsequent

division comes about:

TKV 1.5: eko ‘yam vedākhyo darśanātmani sthito drsyo ‘rthaḥ, sa maharşibhir bhedenābhedasya100 pratipādayitum aśakyatvād, abhivyakti-nimittāl labdha krame vāg-ātma-rūpe prāpitaḥ, ekatvānatikramena samhitā-pada-krama vibhāgena pravibhakta-mārgo ‘dhyayana-nimittām adhyetīņām carana samākhyām vyavasthāpayadbhiḥ samāmnātah.

apara āha. yathā vāg deśa[di-]bhedena bhinnā, saty api sva-rūpa-bheda, ekābhidheya-nibandhanatvam avyatikrāntā, saiva ca deśādi-bheda-prakalpana vyavasthā-hetuh, evam caranabhede ‘py ekārtha-nibandhanatvam avyatikrāntāni śruti-vākyāni, sva-rūpa-bheda eva carana-bheda-prakalpana-vyavasthā-heturiti.

apare manyante. yathāstānga āyur-vedah purā-kalpa eka evāsit, sa eva hi kalau sakti-vaikalyān nịnām pravibhaktāngo drsyate, tathāyam apy aparimāņa mārga-sakti-bhedo brahma-rāśir iti.

A translation of the preceding, segmented to show the different ways stated by BH, would be as follows:101

(al) “This visible entity called Veda, as (an entity) existing in the vision unit (of the great seers), is one. It was, through division, set down for transmission (note 7) by the great seers, for (the) unity (of vision they have in the revelation experience) cannot be transmitted (as it is, i.e., the great seers must follow a sequential form if they are to transmit the Veda). It (= the Veda entity) is (first) given a linguistic form that has sequence due to the causal factors which bring about (its) manifestation (that is, due to the organs of articulation etc.).

(a2) “(then) its ways are differentiated due to the division into samhitā-pāțha, pada-patha and krama-pātha.”

(a3) “(finally) they establish the caraña designations of the (Veda) students which are based on the form of the Veda text they study.”

(b) “Speech becomes different due to the difference of region etc., despite that difference of (phonetic) form, it continues to express the same meaning (i.e., the dialectal variations do not imply variation in what is conveyed). (In turn,) it itself becomes the basis in setting up the differences of regions etc. In the same way the Sruti sentences express the same (pre-variation) meaning (that is, have a unity through meaning), despite the difference of caraņas (Veda-learning

100 One should imagine a comma here after bhedena (“through division’) and take that word as syntactically connected to samāmnātah. It is possible to connect bhedena to vāg-ātma rūpe prăpitaḥ, but, in that case, the great seers need not be invoked as agents. Whether they give the linguistic form or someone else does, the non-unity or sequential nature in the linguistic form will be inevitable. In other words, the phrase abhedasya pratipadayitum asakyatvāt should be taken as a separate unit explaining why what was unitary in the vision of the great seers must assume a sequential form in transmission.

101 As the main concern of the present essay is not what happens after the initial revelation of the great seers and as the differences to be noticed in the translations by Biardeau, Subramania Iyer etc. are minor, I will not note them here.

traditions), for the caraņa difference is set up on the basis of difference of speech form itself.“102

(c) “Just as the Ayur-veda, consisting of eight specialities, was only a single (field of learning) at an earlier age (and) the same (Ayur-veda) is seen divided into specialities in the (present) Kali age, because of the decline in the capabilities of human beings, in the same way is seen this Veda (textual) mass also, as one having the difference of powers of countless paths (or as one having countless path power differences).”

Thus, TKV 1.5 points out three basic ways in which the unitary Veda revealed to certain seers becomes diverse without really losing its unity: (a) Division is unavoidable in speech. (b) Speech changes from region to region. (c) The inability of the recipients to grasp the original in its enirety means that they will grasp only parts of it and thus introduce division into

the unity.

Although (b) and (c) are prefaced with apara āhaand apare manyante, they are a part of the thinking of the author himself. BH’s commentators have rightly abided by this convention behind the use of apara in their interpretation of his thought (cf. Helā-rāja 3.14.615).

NOTE TO POINT 5: A suggestion of the sequence in which things happen must be seen in the word sequence vāg-ātma-rūpe prāpitaḥ … samhitā-pada-krama-vibhāgena pravibhakta-mārgah, for the talk of division into Samhitā etc. would not make sense unless the (subtle) Veda is assumed to have gained a linguistic form. However, we should not infer from the sequence ‘(a2) → (a3)’ that, in BH’s view, the generation of pada-pāțha, krama-pāțha etc. and the generation of the caraṇas of the Veda necessarily takes place in that order. True, we normally think of the four Vedas, the division of each of them into samhitā-pāțha, pada-pāțha etc. or the division of each of them into sākhās, and the expansion of each Veda through Brāhmaṇa, Aranyaka etc. as forming a top-to-bottom or earlier-to-later diversity, but a specification of this type is not to be found in BH’s works. In the relevant passages of his works, in which TKV 1.5 analysed above is the most detailed, the emphasis is on the ways (i.e., the various manners) in which the diversity comes about, not on how the noticed or recognized constituents of the diversity are related to each other. BH does not tell us what the sequence of occurrence is between (a) the division into four Vedas, (b) the division into rc, yajus, and sāman, (c) the division into samhitā-pāțha, pada-pāțha, krama-pātha etc., (d) the division into sākhās, and (e) the division into Brāhmaṇas, Aranyakas etc. The absence of an exhaustive and explicit statement on the divisions could simply be due to absence of need in the relevant contexts,

102What BH states here is valuable in that it anticipates the most commonly given modern explanation of the origin of Veda sākhās: regional speech peculiarities were bound to affect Veda recitation despite the effort made to ensure its uniformity.Veda Revelation according to Bhartr-Hari, appendix 1

but it could also be a reflection of the time. Variety on all the indicated fronts and in all the indicated ways existed, but either there was no interest in determining how it historically came about or there was no possibility of being able to make a historical determination. Although, it is frequently taken for granted in modern statements that, first, the four major Vedic traditions (RV, Yajur-veda, Sāma-veda and Atharva-veda) took shape and then each began to show internal variation due to geographic spread, dialect influence etc., leading to the recognition of sākhās, it is possible that at least some of the sākhās came into existence, say, before the RV and Atharva veda, having the verse mode, were separated or before the RV and Sāma veda were set apart to do justice to the ‘verse mode: sing mode’ distinction. The only justifiable course open to BH was to write in such a way as to reflect synchronically the scene of his time. Further, we should also be aware of the fact that, for a believer, the validity of the different parts of a scripture does not depend on which part is more recent (although it can depend on the purpose of the part).

In accordance with this surmise, BH’s usage of carana is neutral as to the difference between a whole Veda and a Veda-śākhā. The ground reality was that the Vedic community known to him was primarily associated with specific sākhās. The etymological meaning of carana suggested that the word should cover what the groups in the community lived by or practiced. Thus, the coverage could be in a wider sense (by the Veda disivion) or in a narrower sense (by the sākhā division).103

Depending on the context, carana could also refer to the accepted ways of non-Vedic communities.

Establishment of what the first revelation

should include in BH’s view: As Vrşabha, in 4(a) above, notes for us, BH was aware of a view which considered even the division of the Veda into śākhās etc. as part of the revelation. In this view, the unfolding of the Veda diversity would not be gradual. Even the sākās would be eternally present in the Veda whole (although the texts of some of them could be lost on the human level). But BH does not seem to prefer this view. In his thinking, the Veda coming from the unitary vision of the great seers is intially undivided; it is a (brahma) rāśi. Its division into four major or nodal traditions, the branches associated

103 Carpenter (1995: 41), while commenting on TK 1.5 writes: “The apparent multiplicity involved here would seem to refer to the differentiation of the Rg, Yajur, Sāma and Atharva Vedas rather than the various branches (sākha-s) of each of these, since these branches are dealt with separately in [TK] 1.6. The unity that Bhartr-hari has in mind here thus precedes or underlies the fourfold division with which we are familiar.” As my discussion here indicates, the first sentence in this statement may not be accurate.

with these traditions etc. is something that happens predominantly on the level of ordinary (non-maharsi) humans, in what we would call ‘historical’ time and not always with planning or deliberation.

We can determine this to be the case on the basis of the following direct and indirect evidence:

(a) The kārikās form the main text of the TK. The main text in the traditional Indian genre consisting of ‘main text + author’s own commentary is primarily given to stating the author’s own views. Unless a kārikā is followed by a kārikā stating a different view, it can generally be taken to express an author’s own thinking. In TK 1.5, the wording is aneka-vartmeva samamnātah, ’transmitted as if it has more than one way,’ indicating that the vartmans, ‘ways,’ are not the ultimate state — the Veda continues to remain one even when the vartmans come into existence. Such a statement indicates that unity is the starting point.

(b) Explicit support for the inference made in (a) is furnished by TKV 1.5 and TKV 1.173: eko ‘yam vedākhyo darśanātmani sthito drsyo’rthah “This entity which abides in the vision unit (or unitary vision) is one” and te ca … sarvam āmnāyam sarva-bheda-sakti-yuktam abhinna-sakti-yuktam ca paśyanti “And they (= the seers in the second group) … see the entire traditionally transmitted (Veda) text body, which possesses all powers of differentiation and which possesses (those) powers in such a way that they are not different (from it).“104 As (c) below will establish, these two statements appear in contexts that state BH’s own views. They speak of the vision as unitary. Logically, the first phase for the object of such a vision, when it appears on the ordinary level of human experience, must be one in which it is free from labeled or labelable segmentation and phonetic deviation.

(c) In TKV 1.6, an explicit statement (or refutation) of the view that the śākhās are always there (and hence they must be there even in the first revelation) is missing. However, it is obviously implied by tu in the text that is available: yeşām tu ayam sākhā-pravibhāgo vicchedena punaḥ punar bhavatīty āgamah, teșām, prāk pravibhāgād, avyabhicāra eva samhịta-kramāyā vāca, ity etad darśanam. “But, the view of those whose inherited position is that this (Veda) division into branches comes about again and again, with loss (of the division in between), is that, prior to the division, language as an entity with withdrawn sequence invariably exists.” Such a statement clearly presupposes the presence of yeşām tv ayam sākhā-pravibhāgo vicchedena punaḥ punar na bhavatīty āgamah … . This latter kind of statement seems to have been available in the mss on which Vrsabha (p. 27) based his commentary. His wording indicates that the lost statement began with yesam tu, which

104If BH had thought of the seers as seeing the amnāya in its diversity or as a divided object, he would have used the expression sarva-bheda-yuktam, not sarva-bheda-sakti-yuktam.

Veda Revelation according to Bharty-Hari, appendix 1

made the later scribes lose it through haplography. These considerations make the preserved statement a second (i.e., the final) statement in the present case. Normally, when a śāstra author states alternative views, the view in the final position is the siddhānta.

A similar situation is noticed in TK 1.172-173 and their Vrtti. The view that the sākhās come and go is stated after the view that the sākhās are permanently there.

(b) Statements elsewhere in the TK establish that if the possibility of world (universe or cosmos) creation must be entertained BH would prefer to entertain it as something happening cyclically (Aklujkar 1991a: $3.2 fn. 40). The facts in (a)-(c) are consistent with this conclusion. Thus, we should view BH as a proponent of the view mentioned in point 4(b). At the same time we should note that BH does not have anything of fundamental importance in his philosophy to lose by leaving room for the possibility in 4(a). Such a gesture would agree with his perspectivistic way of doing philosophy, that is, with his philosophy of philosophy. The ultimate truth of BH’s philosophy does not depend on accepting a creation of the physical world. There are two versions of his sabda-tattva brahman thesis. According to the weak one of the two, the sabda-tattva would be the ultimate ‘given’ or truth even if the world is not thought of as actually coming into being from it. It can be thought of as a cause of the world only in the sense that the world exists for someone only as long as the sabda-tattva exists in that someone (cf. Aklujkar 2001).

Further, it is reasonable to hold that BH thought of the object or content of genuine Vedic revelation to be Mantra as distinct from Brāhmaṇa (see point 2, note (b) above) etc. At least that seems to be his subconscious association from the employment of phrases like mantradrśaḥ paśyanti in TKV 1.5. Such an association would agree with the evidence Bhagavad datta (1978: 94-116), Yudhisthira Mīmāṁsaka (1974: 139-78, 1977: 68-86, 1980: 17-18) and Holdrege (1994) have collected.

Halbfass’ understanding of

BH’s Veda concept: Wilhelm Halbfass (1991) addresses the topic of the place of the Veda in BH’s thinking with much learning, insight, and sensitivity. His discussion is broader in its concern than mine. It is also spread over several intervened pages and is frequently comparative in spirit (most of the comparison takes place with respect to Kumārila, Sankara, and Jayanta). Also, he pays greater attention to role of the Veda than to the concept and revelation process of the Veda.

In what follows, I point out how some pertinent remarks of Halbfass would have been more defensible had he clearly stated that there are (at

least) two senses in which BH uses the term veda. In other words, I offer further justification of the explication strategy adopted in Aklujkar 1991a, as well as seek to ensure that the wording of Halbfass’ remarks does not mislead future researchers. As I see it, by not telling his readers that BH uses the word veda in more than one sense — by presenting as one complex the senses that BH indicates to be different — Halbfass unintentionally leaves the impression that BH has some mystical or mysterious and possibly inconsistent concoction of elements in his Veda concept105_ that BH lumps together ideas we would normally keep apart.

My procedure will be to cite Halbfass’s remarks and indicate how they support the critique offered just now. I give, in square brackets, the numbers Halbfass assigns to his notes (printed toward the end of the chapters in his book) and the textual references the notes contain.

(a) Halbfass 1991:4-5: “Understanding the role of the Veda in Indian thought involves more than textual hermeneutics. It also involves what we may call the hermeneutics of an event. The different approaches to the Veda are not just different interpretations of a text, and commitment to the Veda is not only, and not even primarily, acceptance of a doctrine. In another and perhaps more fundamental sense, it means recognition of a primeval event, and a response to a fundamental reality. In the understanding of those who accept it, the Veda itself is beginning and opening par excellence. It not only speaks, in its own elusive fashion, about the origin and structure of the world and the foundations of society; it is also their real and normative manifestation and representation.” The observation is very good. The questions that remain are:

(i) How do thinkers that otherwise show great sophistication in rational thinking accept the position that a text could contain nothing but what is really there or what ought to be there always — a comprehensive actual or potential matching with the present or would-be worldly reality? They were obviously aware that not all texts are reliable. Are they engaging in only dogmatic assertions? Are they deliberately refusing to question faith in some areas, or do they have simply ‘human’ weaknesses and understandable blind spots? Is their position not different, in some essential way, from that of those who place their faith in a person (e.g., the Buddhists, the Jainas) or from that of those who declare a certain text to be the unquestionable guide in order to ensure that followers in the subsequent generations do not put their trust in anther human being (e.g., the Sikhs)?

(ii) How did the thinkers concerned reconcile the Veda being an event with the Veda being a text? Were event and text not different categories in their thinking?

105The inconsistency would be with the distinction BH clearly makes eleswhere between language and reality or between a text and the world.

Veda Revelation according to Bharty-Hari, appendix 1 I think Halbfass would have agreed with me in holding that BH’s credulousness — if that is what we are noticing here — would not go to the extent of confusing event and text. BH may be a man of faith and he may even have his favorites chosen non-rationally at some level (that is, in effect chosen for him by the situation in which he was born and brought up), but he is not so naive as not to realize when his statements would appear opaque, confusing or contradictory.

(b) Halbfass 1991: 5: “The language of the Veda is primeval reality. Bhartr hari says that the Veda is the “organizing principle” (vidhātr) of the world, that is, not only its “teacher” or principle of instruction (upadesty), but also its underlying cause and essence (prakrti). [12: TK 1.10 and V; ch. 3 $8f.] This may be an extreme and somewhat unusual form of expression, but the basic viewpoint it articulates is by no means isolated. The Manu-smrti, as well as the other dharma texts, characterize the Veda as an organizing and sustaining principle, and even as the real basis of the social and natural world. [13: Manu 2.76ff; 3.75; 12.99; 2.7 with Medhātithi’s expl. of Veda as sarva-jñānamaya] It would be wrong to view such statements as merely metaphorical. The Veda is the foundation of language, of the fundamental distinctions and classifications in the world, and of those rituals which are meant to sustain the social and natural order. It is itself the primeval manifestation of those cosmogonic occurrences which establish the dharma. [14: India and Europe ch. 17] Text and world, language and reality, are inseparable in this world-view and self-understanding. [15: The Veda itself frequently presents itself as a cosmic or cosmogonic reality. See, for instance, Rg veda 10.90.9; Muir III.3ff.] The text itself opens and sustains the “world” in which it appears, to which it speaks, and by which its own authority has to be recognized and sustained.”

I am not aware of anyone else who has so perceptively noted the notions contained in the relevant Sanskrit texts and articulated them so succinctly. Still the following question occurs to me: How could the individuals who knew that all language is not reality — that the text and the world cannot be identical if commonsense is to be respected — subscribe to a position of the described kind in the case of the Veda? Is it really likely that they did not see the contradiction? And, if they did not see it themselves, would their contemporary Buddhist and Jaina philosophers not bring it to their attention? If they made an exception, what was special about the Veda that distinguished the Veda as language or text from ordinary language(s) or texts? Could it not be the case that BH, possibly following his predecessors, explicates what is implicit in the statements found in works like the Manu-smrti? In other words, there must be a notion in addition to the ones so nicely recorded by Halbfass which helped the thinkers involved to reconcile their statements — to satisfy the rational man and responding individual in them. This notion must be that of levels and two different senses of veda relative to the levels as the Vrtti sentence vedo hi lokānām prakstitvena copadestįtvena ca vivarteșu [ca] vyavasthāsu ca vidhată confirms. Here, the role as fashioner (vidhāts) through being praksti and through

being a teacher are distinguished as obtaining in the contexts of vivarta and vyavasthā, respectively (note 36). BH tends to use two ca-s when two notions that should be kept apart as belonging to different categories are paired (cf. Aklujkar 1991b). Thus, the repetition of ca seen in prakrtitvena copadestytvena ca is significant. Vrşabha (p. 38), too, states that two different explications of vidhātus tasya lokānām in the kārikā are offered in the Vrtti sentence. He is careful to point out the ‘respectively’ or yathā-samkhya construction implicit in it and thus to convey that two functionally different Vedas, that is, two meanings of veda, are intended: yasmād ayam vedo brahmākhyo jagat srjati varnāśramāmś ca sveșu karmasu vyavasthāpayati.106 yato ‘yam prakrtitvena vivartesūtpattisu lokasya vidhātā, ādye vyākhyāne. dvitīye tu varņāśrama-vyavasthopadeśakatvād upadestrtvena vidhātā. Later, in the same V 1.10, BH himself suggests that his preceding remarks were based on a wider, functionally bipartite, concept of the Veda (and brahman) when he uses the phrase: vedākhyasya prasiddhasya brahmaņo ’ngebhyaḥ.107

(c) Halbfass 1991: 35: “Unlike Bhartr-hari [51: see below $12], they [= Kumārila and his followers; AA] do not recognize a dynamic extension of the Veda into the world of human speech and thought.”

Here Veda must be what I call the subtle form of the Veda. The Veda as text cannot so extend. Any talk of it doing so will puzzle a reader. What Halbfass’ point, therefore, boils down to is this: BH and his followers are willing to admit the language principle as the ultimate reality and to accept its equation with the Veda, in one sense. Kumārila’s side rejects both the steps.

(d) Halbfass (1991: 37-38: “Unlike Kumārila, Bhartr-hari does not draw a strict border between the uncreated Veda and the traditions of human thought and exegesis. And unlike Sankara, he does not postulate a radical dichotomy between absolute and relative, empirical-practical truth (i.e. paramārtha and vyavahāra). Bhartr-hari’s Veda is brahman’s unfolding into the world; it extends into the social and natural world as its underlying structure and basis. The Veda itself is a dynamic process, initiating its own divisions into different parts, branches and recensions, this process of differentiation and expansion is continued and extrapolated in the work of human exegesis. Not only the “seers” (rsi) who manifest the Vedic texts, but also their exegetes and interpreters, are agents and instruments of the self-manifestation, self-differentiation and self-explication (vivarta) of the absolute “word-brahman” (sabda-brahman). … they are not only speakers about, but agents and representatives of the reality of the Vedic word and they are participants in cosmic and cosmogonic processes. The Veda, in

106Up to this point Vrsabha exploits the two meanings of veda (a particular textual cor pus’ and ‘sabdatattva/ sūksmā vāc’) and brahman (‘sabda-tattva/sūkṣmā vāc’ and ‘Vedic cor pus’) as BH himself exploits them (but only in the case of veda at the beginning of his remark).

107Vrsabha rightly points out here that veda is now meant in the narrow sense of a specific body of texts’: laukikām prasiddhim āsthāya veda-śabdārthasya vyākhyām karoti, loke sārthavādakasya

sopanisatkasya brāhmaṇasya mantrāņam ca vedākhyā.

Veda Revelation according to Bhartr-Hari, appendix 1

whose manifestation they participate, is not just a text about brahman, but its actual “imitation” and representation (anukāra).”

Here too, for most of the passage, Veda must be what I call the subtle form of the Veda. Halbfass seems to realize this when he states “The Veda … is not just text,” but he still leaves the impression that the non-text Veda is simply an addition to or an extension of the text Veda without a difference of levels and that the functioning of one takes place, in BH’s perception, at the same time as that of the other in the same realm. Actually, the subtle, formless creation-prone (pranava) Veda pervading human reasoning, exegesis etc. as the language principle or creator source of rşis, is distinguished from the specific text body that appears as diverse and is bestowed by the rșis.

This is not to say that BH is a Kevaladvaita Vedāntin or that he does not differ from Kumārila. I am only pointing out that a clear awareness of the meaning of veda would put us in a better position to locate the differences of thought accurately. Halbfass’ way of differentiating between the philosophies of BH and Sankara raises the question of how he would account for the acceptance of avidyā by BH. On the other side, his way of differentiating between the philosophies of BH and Kumārila makes the reader curious about how he would explain the convergence of BH and Kumārila in locating the source of even (what we consider to be) non Vedic philosophies in the Veda.

(e) Halbfass 1991: 38: “Thinking and reasoning (tarka) have to be supported and upheld by the Vedic tradition. They are “permeated” (anuviddha) [67: TK 1.131] by the Vedic words; legitimate human reasoning is ultimately nothing but the “power” and manifestation of these words (sabdānām eva să śaktis tarko yah puruşāśrayah). [68: TK 1.153; cf. Bhartr-hari’s citation of Pāraskara Gșhya-sūtra 3.6.5 vidhir vidheyas tarkas ca vedah in V 1.10)”

The qualification “Vedic” employed by Halbfass is not supported by the contexts in which the cited remarks of BH occur. The remarks are general in nature: thinking and reasoning are not language-independent (even though most persons are not aware even of the possibility that language could be determining how they think or reason). The reference to Pāraskara’s definition is made under the alternative ‘pranava = sarva sabdārtha-praksti (= language principle) = Veda,’ that is, with the subtle form of the Veda or the wider sense of the term veda in mind. The tradition and the words Halbfass has in mind here ultimately go back to the Veda but in a non-textual sense of veda. While maintaining that all language or the language principle and the Veda meet and/or merge at some level of his theory or that all traditions (āgamas) are ultimately derived from the Veda, BH is not claiming that thinking and reasoning, to be legitimate, must have a basis in a particular realization of language, namely the Veda. Words coming from the subtle Veda can even be used to articulate positions

that run counter to the Veda text and the Vedic tradition. TKV 1.8 (asad asato … avastukād avastukam jāyate) speaks of what we would take to be a Buddhist view as arising from an artha-vāda or artha-vāda-like sentence in the Veda, which sentence, by definition, would be a part of the Veda text and the Vedic tradition.

Thus, as a result of not specifying which of the two senses of veda (my “vertical double reference” in point 1 above) he has in mind while offering a particular observation, Halbfass’ account becomes misleading at points. BH’s view of the process and extent of Vedic revelation comes out as more faith-based, that is, as less philosophical and more dogmatic or blindly assertive, than it actually is.

APPENDIX 2 Other translations and interpretations of TK and TKV 1.5 and 1.173108

TK 1.5 and its Vịtti: Biardeau 1964; 33:

TK 1.5: “C’est de Lui que le Veda est moyen d’accès et figure; quoiqu’ Il soit un, les grands voyante l’ont transmis comme comportant de multiples voies séparées les unes des autres.

TKV 1.5: «Figure» : cetter Parole subtile, éternelle, au-delà des sens, les voyants qui ont l’intuition directe de la loi religeuse, qui voient les formules, la voient, et désirant la faire connaitre à d’autres qui n’ont pas l’intuition directe de la loi religieuse, ils la transmettent fragment par fragment, comme on essaie de décrire ce que l’on a vu, entendu, éprouvé en rêve, C’est, du moins ce que l’on raconte de l’origine. En effet, on l’a dit : «Il y eut des voyants qui avaient l’intuition directe de la loi religieuse; aux autres qui n’avaient pas l’intuition directe de la loi religieuse, ils

108 In the case of the Vrtti, I will reproduce translations only of the parts relevant to Veda revelation, Secondly, I will generally refrain from commenting on the differences from my translations and interpretation. In most cases, those with good reading ability in Sanskrti should be able to decide which translation is preferable.

Only Carpenter has made an attempt to provide separate elucidations and interpreta tions, some of which have already been acknowledged in the main body of this essay. From Subramania Iyer and Biardeau we do not get much more than literal translations (generally the latter is closer to the original Sanskrit). These translations are, in some cases, unclear in their intent even to those who are at home in Sanskrit and Indian philosophy, for no effort has been made by the translators to unravel the reasoning that may lie behind the propositions in the original Sanskrit. The lack of effort of the indicated kind is probably due to the fact that Biardeau and Iyer were translating the Vștti for the first time (the latter mainly as an aid to his edition).

Veda Revelation according to Bhartr-Hari, appendix 2 transmirent les formules par l’enseignement; ceux-là, trop faibles pour (transmettre cet) enseignement, a fin d’en donner une connaissance par fragments cherchèrent à transmettre cet ouvrage, le Veda et ses annexes. Bilma signifie «> ou «explication» [Nir. 1.20].»

Subramania Iyer 1965: 7:

TK 1.5: “A means of attainment and a symbol of that One is the Veda, which though one, has been handed down as though in many recensions by the sages.”

TKV 1.5: “By the word symbol (anukāra) the idea contained in the following ancient saying (purā-kalpa) is meant: The Rşis[,] the seers of the mantras, those who have realised the truth (dharma)[,] see that subtle, inaudible Word and, wishing to communicate it to those who have not realised the truth, teach the symbol of it which is like a dream 109 in their desire to tell what they have seen, heard and experienced. It has indeed been said:— The Rșis realised that truth (dharma); they taught the mantras to those who had not realised the truth; these others, also anxious to teach, proclaimed the Vedas and the Vedāngas, in order that the symbol of Brahman may be understood (bilma-grahaņāya). Bilma is bhilma which means something which illuminates (bhāsanam).”

Carpenter 1995: 44-45110

TKV 1.5: “The seers who have directly seen the ritual ordinances, who see the mantras, see the subtle, eternal Word which is beyond the senses. Desiring to make it known to others who have not directly seen the ritual ordinances, they proclaim (literally, repeat from memory) an image [bilma] of it, desiring to relate what they have directly seen, heard or experienced, as if in a dream. [AA: translation of the quoted Nirukta 1.20:] The seers saw the ritual ordinances directly. To others who had not seen the ritual ordinances directly they proclaimed the mantras by way of instruction. What is called the Veda is a single object of vision established in vision itself (darsanātmani sthito drsyo ‘rthah). Because it is impossible to

109Iyer’s note on which is like a dream” reads: “Just as one’s experience in a dream is a kind of reflection of our experience in the wakeful state, in the same way, the Vedas are a kind of reflection of what the rșis saw in their vision.” I agree with the Vedas are a kind of reflection of what the rsis saw in their vision," but think that Iyer has misunderstood the intent behind svapna-vittam iva, just as his “also anxious to teach” is a mistranslation of upadeśāya glāyantaḥ. He should have accepted the guidance of traditional commentators (Vrsabha, Durga and S-M) in both the instances. The omission of nityām and imam grantham from his translation does not constitute a crucial loss, but it should nevertheless have been avoided.

110 In the parts copied from Carpenter, the brackets contain his additions, except where “AA:” appears.

explain that which is [thus] undifferentiated, the great seers caused the essential form of speech to acquire sequence, for the sake of [its] manifestation in a differentiated form. Then, without overstepping the unity, [the Veda] was proclaimed in different paths because of its different [modes of recitation]: samhitā, pada, and krama, by the great seers who established schools for the Vedic study of the students."

TK 1.173 and its Vrtti: Biardeau 1964: 185:

TK 1.173: “La Révélation fait connaitre les choses qui se déploient à partir de l’indivision (primitive) comme en un rêve; tandis que la tradition est prescrite à partir de signes indicatifs, une fois que l’essence des choses est connue."

TKV 1.173: “… Mais il y en a pour qui la cause (du monde) passe perpétuellement par des phrases de sommeil et de veille et prend la forme d’hommes distincts lorsqu’elle entre en activité; selon ces derniers, des «voyants» apparaissent au plain de l’intuition elle-même, qui ont la vision du grand Soi, être pur et matrice de l’ignorance, et s’identifient à Lui par connaissance directe. Mais d’autres (voyants) apparaissent au plan de la connaissance; ils (voient) leur soi avec les noeuds du manas, (mais) pur par rapport aux éléments, éther, etc.; soit pris éparément soit pris tous ensemble, et non encháiné par l’illusion, et ils s’identifient à lui de la même manière. Pour eux, toute activité au plan de l’ignorance est adventice et fictive, tandis qu’être fait de connaissance est pour eux perpétuellement essentiel et premier. Ce sont eux qui, ayant la connaissance de la parole non audible par l’ouïe, voient comme en rêve toute la Révélation à la fois douée du pouvoir de toutes les différenciations et douée d’un pouvoir indifférencié. [#] D’autres encore, voyant la nature des choses en tant qu’elle est un bien ou un dommage pour l’homme, et voyant aussi dans les texes révélés çà et là des signes indicatifs qui s’y rapportent, composent la tradition, celle qui a des résultats visibles et celle qui a des résultats invisibles. Tandis que la Révélation, ils ne font que la transmettre telle qu’ils la voient sans changer sa formulation, mais tout d’abord à l’état indivis, puis divisée en recensions. Telle est la Tradition.”

[The relevant part of Biardeau’s fn 1 at this point]: La suite [= what follows the allusion to the Mimāṁsā view in the preceding paragraph] est beaucoup plus obscure quoiqu’elle ait probablement sa racine dans le texte de Yāska, trad. p. 33 (Nir. 1-20), mais on peut évoquer à son sujet la littérature āgamique plutôt que le Vedānta shankarien; une expression telle que avidyā-yoni par example, appliquée à l’ātman universel, rend un son bien peu vedāntique. De même, il est question de «voyants» qui se manifestent au plan de la vidyā en un second stade de l’émanation cosmique; le terme vidyā doit se distinguer duVeda Revelation according to Bhart?-Hari, appendix 2

pratibodha qui précède : pratibodha — que l’on peut comparer à pratyakșa — connote l’idée de connaissance directe, immédiate et sans distance qui correspond à pratibhā; ce’st l’intuition par identification à la Parole même. Vidyā serait au contraire une connaissance d’ordre plus naturel et discursive?

Subramania Iyer 1965: 131:

TK 1.173: “In those who evolved out of the undifferentiated, there is knowledge of the Sruti (revealed Scripture) as in a dream. The written Tradition, on the other hand, is composed by the sages, after understanding the nature of things and following the indication (found in the Vedas).111

TKV 1.173: “… According to those who hold that the ultimate cause works in the manner of sleep and wakefulness and the differentiated individuals, 12 some sages manifest themselves as identical with Intuition; they see it, the great Self in the form of Being, the source of Nescience and endowed with all knowledge and they become one with it. Some sages manifest themselves together with (the means of) knowledge. They identify themselves with their Self in the form of mind-knot, free from the elements ether etc., either severally or collectively, i.e., devoid of any sense of ’l’ in regard to them. All the activity of those sages is the product of Nescience and, therefore, adventitious and secondary. Their being essentially knowledge is eternal, non-adventitious and primary.113 They see the whole Scripture, endowed with all power of differentiation and all power of unity, as one hears sound in a dream, inaudible to the ear.

Some other sages, after perceiving the nature of objects, conducive either to the welfare or to the harm of man and after seeing in the Scripture indications thereof, compose the tradition, (the observance of which) leads to visible and invisible results. At first, they hand down the Scripture in an undivided manner, without any deviation in the words, as they saw it and later, they hand it down, divided into branches. Such is the tradition.”

Carpenter 1995: 46-48 (see note 110):

TK 1.173: “Insight into śruti (or Veda] belongs to those who become manifest out of the undifferentiated. The smrti is composed on the basis of indications [found in the śruti] after the nature of existent things has been examined.”

111 This translation does not differ in intent from mine. It would have been better if Subramania Iyer had not used “written” and made “nature’ reflect the force of tattva in the original kārikā.

112 Here, Iyer renders the two words vrttyā and anukārtayā with a single phrase ‘in the manner.

113 I do not understand the syntax or intended meaning of this sentence.

TKV 1.173: “… But those who believe that the eternal cause develops by an activity of sleeping and waking that imitates (the sleeping and waking] of an individual man [understand śruti and smrti this way]: Some seers become manifest in intuition (pratibhā] itself. Seeing the great Self that is defined as Being and that is the source of Nescience, they become identified with it through direct insight. Some, however, become manifest in knowledge [vidyā). They become identified with the Self [which has acquired the form of the) mind-knot, which is free from the elements such as the ether, whether individually or collectively, and which has an unbounded power of imaginative construction. In their case, whatever is adventitious, being the activity of Nescience, is all secondary. But whatever is knowledge per se, eternal and non-adventitious, is primary. Through their intuitive knowledge [prajñā] they see the entire Veda [āmnāyam], joined with the capacity for every differentiation and with the capacity for indifferentiation, like a word that cannot be approached through hearing. Some, however, after examining the nature of those objects that relate to the welfare and harm of man, and having seen indications relating to them in the Vedas, composed the tradition with its visible and invisible purposes. Sruti, however, is proclaimed first in an undivided state, according to vision, with words that are without deviation, and then again, divided into schools.”

[Carpenter’s elucidation and interpretation:] “Bhartr-hari describes three types of seers here, which can be ranked according to the intensity of their vision and their stage in cosmic evolution. The first type, being the first to become manifest at the beginning of a day of Brahmā,” is identified with pure vision. Here knowledge and being are identified. Though they see the source of the Nescience (avidyā) that will soon cast a veil over Being’s luminosity, there is no indication that this Nescience itself is as yet actual. The picture is one of perfect self-conscious unity.

“At the next stage, and among the seers of the second type, the power of Nescience becomes active and we now have a less perfect vision, mixed with what is adventitious, but still inward, independent of the subtle elements from which the material world will evolve. These seers see the Veda that is still one, but that already displays its capacity for differentiation, for becoming the Veda as actually proclaimed in human speech.

“Finally, Bhartr-hari describes the third type of seer, which is really to be identified with the sista-s, the learned author of the various traditional texts and derive their authority from the Veda, but are also based on the experience of their authors.

“It is the second type of seers described here that helps us understand the manner in which the Veda as a unitary, visionary Word takes on the form of actually uttered words, for it is clearly this type that actually “proclaims” the Veda. 114 … here the context is not purely individual or psychological but cosmogonic, and the role of the imagination is assimilated to the creative activity

114 In contrast to Carpenter, I see evidence for only two types of seers in TKV 1.173. I think of them as appearing in a logical sequence, not in a chrolological sequence.

Veda Revelation according to Bharty-Hari, appendix 3

of the True Word in its self-manifestation. It is the True Word itself that introduces the factor of diversification, the power of Nescience or self-veiling (avidyā-sakti) that makes possible the manifestation of the world of multiplicity. … For Bhartr hari, this fundamental cosmic mystery is likewise the fundamental mystery of language and consciousness.

“… even in their limited role of proclaiming or “enacting” the Veda, “translating” it from its visionary to its uttered state, the seers function not purely as individuals but also as vehicles of the True Word’s own intrinsic dynamism. In introducing diversity and temporal differentiation into the essential form of Speech, they are merely imitating the True Word itself in its move to self-manifestation. Indeed, they are themselves perhaps best understood as symbols of this process. 115 It is a process that ends not with the vikāra-s, or manifold forms of the world of ordinary experience and action (vyavahāra), but with the Veda as the anukāra, the True Word, with the world revealed as language and the order that that language reveals, the dharma.”

APPENDIX 3 Explanations and translations of Nirukta 1.20116

The passage sākṣāt-… vedāngāni ca has its difficulties of interpretation, but it is not one of the unusually difficult passages of ancient literature. The difficulties it presents cannot, except for bilma, be said to arise out of use of obscure words. The central meanings of sākṣāt, kr, dharma, ?și, bhū, avara, upadeśa, mantra, sam + pra + dā, glai, grahaņa, grantha, sam + a + mnā, veda and vedanga are well-attested. The difficulties of interpretation are, thus almost entirely due to ambiguity of relations between the notions that appear in the sentences. Their links must have been immediately clear to Yaska’s hearers or readers but are not so to us as $2.5 indicates.

The passage is preceded by a discussion of the importance of knowing the meaning of the mantras, which discussion, in turn, is prompted by a desire to establish the usefulness of the Nirukta. Yaska cites three verses (two of them identified as Rg-veda 10.71.4-5) 117 which state or indicate

115The agency I attribute to the seers in my interpretation is much greater. Also, since I view the discussion also as belonging to a philosophical mode, I do not personify or deify the true word or language principle as much as Carpenter does.

116 The immediately relevant parts of the texts covered here have been referred to or cited at the appropriate points in the main essay. The continuous presentation of the texts made in this appendix is intended to provide the larger context necessary to ensure that the views of other scholars are not misunderstood and to facilitate precise comparison. More importantly, the reader should get some sense of how much careful textual work is still to be done in the case of the valuable commentaries of Durga and S-M.

117 Wezler (2001: 216) speaks of the source of the verses with this specification and also as Samhitopanișad Brāhmaṇa 3, following Bhadkamkar 1918: 139.

that the meaning of the Veda should be known or that one should go beyond the perceptible form of vāc. However, there is no compelling reason why the passage sākṣāt- … should be connected to the explanation of the last verse or, as Wezler (2001: 216) prefers, to the common point of jñāna praśamsā and ajñāna-nindā, ‘praise of knowing and censure of not knowing, that the three verses jointly make. If the reference to imam grantham and vedāngāni in the passage is taken as a reference to the tools of making sense of the Veda, the passage could be connected to the common point. But then there are several details in the part of the passage preceding imam grantham and vedangāni which have no obvious or easy connection with jñāna-prasamsā and ajñāna-nindā. Besides, Yaska, who begins the adhyāya with samāmnāyaḥ samāmnātah, can justifiably be thought of as indicating to his hearers or readers through samāmnāsiṣuḥ that he is now returning to the delineation of the nature and historical background of his commentandum (82.5, paragraph 2) after commenting on some general concepts and issues involved in his undertaking. Therefore, the traditional commentators seem right to me in taking the passage as marking the conclusion of the first adhyāya of the Nirukta and, in effect, tying the end of that adhyāya to its beginning by telling us how the samāmnāya and the Nirukta based on the samāmnāya came into being.

The text of the passage does not vary significantly as far as the details crucial to our inquiry are concerned. It appears as follows in the edns commonly used: sākṣāt-krta-dharmāņa rsayo babhūvuh. te ‘varebhyo ‘sākṣāt-kyta-dharmabhya upadesena mantran samprāduh. upadeśaya glāyanto ‘vare bilma-grahaņāyemam grantham samāmnāsişur vedam ca vedāngāni ca. bilmam bhilmam, bhāsanam iti vā.

The Yukti-dīpikā (Wezler-Motegi edn p. 251-252), citing this text, uses the stem apara in the place of avara. In this, it agrees with the TKV, Helā-rāja 3.1.46, Nārāyana’s commentary on Kaiyața’s Mahābhāsya-pradīpa cited in appendix 4. The ‘avara : apara’ difference will, therefore, be considered in appendix 4. Despite the difference of connotation and possible significance in terms of attitudes of the authors concerned, the meaning of ‘other’ remains constant. We should also be open to the possibility that this ‘other’ may refer to a single historical generation but is more likely to stand for a conceptual group comprising several generations.

Rajavade (1940: 290-91, 686) has written: “… I think vedam ca vedangāni ca is an interpolation. samāmnāyaḥ samāmnātaḥ is said [at the beginning of the Nirukta] about the Nighantu; to use samāmnāta [ samāmnāsiṣuḥ?] about the Veda would be something like an insult to the seers who must have seen the Veda in all its several branches. Yāska would not include his

Veda Revelation according to Bharty-Hari, appendix 3

Nirukta which is a Vedānga among helps composed by the old teachers to facilitate Vedic study; he would never say niruktam samāmnātam

[→ samāmnāsiṣuḥ?]. Besides what follows is about the Nighantu only and not about the Veda and its helps.”

This comment is only one among the many whimsical, overconfident and logically loose comments by Rajavade. He would have been better off, albeit not correct, if he had given a reason like ‘vedam ca vedāngāni ca’is an interpolation, because it occurs after the verb samāmnāsișuh, which already has an object in imam grantham.

Three old sources, Durga, BH and S-M, attest to the existence of vedam ca vedāngāni ca. If the phrase is an interpolation, it must be such an old interpolation as to leave no room for its determination as such. Further, does sam + ā + mnā become a dirty root word inapplicable to the Veda simply because it is employed for the Nighantu? Would Veda, which is called āmnāya, be insulted if it was called samāmnāya? Are there not first person occurrences sam + ā + mnā with which the authors of certain sūtras refer to what they wish to plant in the tradition? Thus, none of Rajavade’s arguments has any substance.

Some discussion has taken place as to the referent of imam grantham (cf. Wezler 2001: 238). Does it refer to the text on which Yaska is commenting, namely the Nighanţu, or to the Veda that is mentioned after samāmnāsiṣuḥ (i.e., as a phrase standing in apposition to vedam)? In other words, are there two grammatical objects of samāmnāsisuh or three?118 There may be the fault of a little indirect) overlapping in taking the Nighanțu as a referent of imam grantham. The Nighantu, being a part of the Nirukta tradition, can be viewed as covered by vedangāni. However, it would not amount to accepting a strained or arbitrary interpretation if one pointed out that the very existence of the phrasing imam … ca commonsensically suggests that the vedangas mentioned later should be understood as referring to the usual territory of the Vedāngas minus the territory covered by imam grantham, i.e., the Nighantu (on the analogy of vasistha āyātah. brāhmaṇā apy āyātāh). Cf. Durga (p. 147) who supplies itarāņi and S-M who supply anyāny api after vedangāni ca. Besides, the part of the Nirukta which follows the sākṣāt- … passage mainly explains the division of the Nighantu. It would have no organic or a particularly strong logical relationship with the sākṣāt … passage if imam grantham were not employed as standing for the Nighantu (as a part of the Vedānga called Nirukta) as the traditional commentators do (cf. Durga p. 147, S-M p. 116: gav-ādi deva-patny-antam). Also, since Yaska is not commenting on the whole Veda,

18Vrsabha (p. 25) offers a variation in the appositional or ’two objects’ understanding. He takes imam grantham as equal in extent to vedam ca vedangāni ca collectively.

he is unlikely to use imam, a form of idam ’this,’to refer to the Veda. Thus, three grammatical objects (Nighanţu, Veda and Vedāngas) standing for two physical or conceptually joined objects (Veda and Vedāngas, with the latter including the Nighantu and pre-Yaska Nirukta parts; $2.5), seem right for the sentence, even if constructions giving the appearance of three grammatical objects and using a demonstrative pronoun for the first grammatical object, identical in fact with the second or third grammatical object, were to be found elsewhere.

Traditional explanations in their

text-critically improved form: Durga:

āha. kutah punar idam āyātam nirukta-śāstram pradhānam, itarāni cāngānīti?

uchate. sākṣāt-krta-dharmāna rşayo babhūvuh. sākṣāt-krto yair dharmah sākṣād dystaḥ prativisistena tapasā ta ime sākṣāt-kyta-dharmāṇaḥ ke punas ta iti. ucyate. rşayah. rsyanty [… ye ….]19 amuşmāt karmaņa evam-arthavatā mantrena samyuktād amunā prakārenaivam-laksanah phala-vipariņāmo bhavatīti pasyanti te qşayah. “rșir darśanāt” [Yaska 2.11] iti vakşyati, tad etat karmaṇaḥ phala vipariņāma-darśanam aupacārikyā vịttyoktam sākṣāt-krta-dharmāņa iti. na hi dharmasya darśanam asti. atyantapūrvo →ntārūpyo?) hi dharmaḥ. (see $2.15]

āha, kim tešām iti.

ucyate. te ‘varebhyo ‘sākṣāt-krta-dharmabhya upadesena mantrān samprāduh. te ye sākṣāt-krta-dharmāņas te ‘verebhyo ‘vara-kālinebhyaḥ śakti hinebhyaḥ śrutarsibhyah. tesām hi śrutvā tataḥ paścād rşitvam upajāyate, na yathā pūrveșām sākṣāt-kȚta-dharmaņām śravaṇam antareņaiva.

āha. kim tebhya iti.

ucyate. te ‘varebhya upadešena śisyopadhyāyikayā vịttyā mantrān granthato ‘rthataś cal20 samprāduh sampradattavantah G oprattava”?). te ‘pi copadeśenaiva jagrhuh.

119I do not see how the meanings of root is that the dictionaries give could fit the context of Durga’s statements. Durga is unlikely to have stated the derivation of rși from ?s without providing some specifics bringing the meaning of the noun close to that of the root. Hence my assumption that a textual loss has occurred before ye, the addition of which is warranted by the pronominal form te in the following part of the sentence. As the presence of ye immediately before amusmāt would have resulted in the absorption of a through sandhi and as such an absorption has not taken place in the mss of Durga’s commentary used by Sāmaśramī, Bhadkamkar and Rajavade, we need to postulate a text loss also after ye. See $2.15.

120Yaska’s mantrān samprāduḥ becomes mantran arthamś ca samprāduh in Durga’s hands. Given the probability that brāhmaṇa stood for comment (including meaning, application, purpose etc.) on the mantras in (Yaska’s and) Durga’s perspective, the paraphrase mantran arthamś ca samprāduh is close in essence to mantrān brāhmaṇānīti samprāduḥ of S-M.

Veda Revelation according to Bharty-Hari, appendix 3

atha te ‘pi upadeśāya glāyanto ‘vare bilma-grahaņāyemam grantham samāmnāsişur vedam ca vedāngāni ca iti. upadeśāya upadeśārtham. katham nāmopadiśyamānam artham ete saknuyur grahītum ity-etam artham adhikrtya glāyantah khidyamānās, teşv agyhnatsu tad-anukampayā teşām āyusah samkocam avekşya kālānurūpām ca grahaņa-saktim, 12 bilma-grahaņāyemam grantham gav ādi-deva-patny-antam samāmnātavantah, kim etam eva? nety ucyate, vedam ca vedāngāni cetarāṇīti.

katham punaḥ samāmnāsişur iti. āha. (→ āha. katham punaḥ samāmnāsişur iti].122

srnu, vedam tāvad ekam santam atimahattvād duradhyeyam aneka-sakhā bhedena samāmnāsiṣuḥ sukha-grahaņāya vyāsena samāmnātavantah. tad yathā. eka-vimsatidhā bāhvrcyam. eka-satadhādhvaryavam. sahasradhā sāma-vedam. navadhātharvaņam.123 vedangāny api, tad yathā. Vyākaraṇam aştadhā, niruktam

121 (a) Scholars have understood Durga as speaking of two generations of seers and S-Mas speaking of three generations. Note, however, here that the referents of Durga’s avare, glāyantah (taken over in the pratīka part from the Nirukta), and khidyamanāḥ, on the one hand, and of ete and tesu, on the other, must be distinct. This means that, in Durga’s understanding of the Yaska passage, (i) the s-k-ds teach the avaras as upādhyāyas would teach śişyas, (ii) the avaras become despondent and compose the helpful tools out of compassion (iii) for their recipi ents/students who are showing diminishing capabilities.

(b) Vrsabha (p. 25) too hints at the presence of the third group, namely the students of the avaras/paras/aparas (see appendix 4, point 1), in his mind: svayam viditam tādyk katham sākṣād upadeksyāma iti khidyamanāḥ. upadeśārthah kheda iti tādarthye caturthi. The part up to khidyamanāḥ can go only with pare/apare in Vrsabha’s preceding text. His use of the future form upadeksyāmah and his next sentence presume that the paras/aparas are concerned with instructing students (the third generation).

(c) The avaras are a-s-k-d but good enough to put together for transmission not only the Nighantu (-Nirukta) and the other Vedāngas but also the Veda. Their ranking is only moder ately lower. Wezler (2001: 218-222) comes to a similar conclusion on the basis of other consid erations.

(d) Durga’s statement confirms the ‘settled text’ connotation in the meaning of veda that I have deduced from other considerations in $4.3.

(e) As confirmation of the thought in (a) and (b), note Durga, introductory section, p. 29: sā [=Pancadhyāyī = Nighanțu] ca punar iyam sāksāt-kyta-dharmabhyo maharsibhya upadešena mantrārtham upaśrutya śrutarsibhir avara-sakti-daurbalyam avekşya … samāmnātā.

122 One expects this āha to be before the immediately preceding sentence (katham punah… iti). In its present place, āha would have Durga (the Uttara-paksin or Siddhāntin) as the speaker. But then the following srnu also would suggest a speaker change to the same effect and come across as unnecessary or odd. In the commentary style adopted by Durga, the appropriate role for āha in most places is to indicate that a questioner or pūrva-pakṣin is speaking (cf. āha. kim tebhya iti, which occurs earlier). When a certain short expression is repeated many times in a work, its occasional misplacement in the mss of that work is but to be expected. The Yukti-dīpīkā, which too assigns the same function to āha, has also suffered similarly in transmission.

123In Durga’s understanding, Veda refers to the text divided in sākhās etc. as in the ‘other’ view known to BH; cf. appendix 1, points 4 and 6. He does not clarify how the sākhās contrib

catur-dasadhety-evam-ādi. evam samāmnāsişur bhedena grahaņārtham. katham nāma bhinnāny etāni śākhāntarāņi laghūni sukham gyhṇīyur ete sakti-hīnā alpāyuso manusyā ity-evam-artham samāmnāsişur iti.

bilma-grahaņam bhāsya-vākya-prasaktam nirbravīti. yad etat etad] bilmam ity uktam etad bhilmam vedānām bhedanam. bhedo vyāsa ity arthah.124 bhāsanam iti vā. athavā bhāsanam eva bilma-sabdenocyate, vedānga-vijnanena bhāsate prakāśate vedārtha iti ata H ity-ata?) idam uktam bilmam iti. evam bhider bhāsater vā bilma-sabdah.125

evam idam rsibhyo nirukta-śāstram āyātam itarāni cāngānīti parisodhita āgamah.

Skanda-Maheśvara (pp. 113-117):126 evam ukta-prayojanasya niruktasya parenāgamaḥ kathyate.

sākṣāt-krta-dharmāna rşayo babhūvuḥ. dharmasyātīndriyatvāt sākṣāt karanasyāsambhavat vad] dharma-sabdenatra tad-artham mantra brāhmaṇam ucyate. tat sākṣāt-kytol27 dharmo yais te sākṣāt-kyta-dharmāņa rşayah.

katham punas taiḥ sākṣāt-kytam?

ucyate. smrtikārair aitihāsikaiś cābhyupagatatvāc.chrutyā cāvirodhād antarāla-pralayaḥ punah-systiś cāsti. tatra s?sty-ādau ya rsayas te ’tīta-systāv

ute to sukha-grahaņa. If they all are approximately equal in length, their creation or coming into existence would not be of much help when the life span is shrinking. Therefore, Durga could be suggesting that specialization (assigning responsibility for the preservation of only certain Veda versions or realizations) made the situation manageable.

124 Durga could have known the legend that Vyāsa divided the Veda-rāsi into the four Vedas. Bhagavad-datta (1978: 159-172) is useful for attestations of this legend in the surviving sources.

125Durga’s explication of bilma leaves something to be desired. From what he says in the paragraph beginning with śrnu and from his phrase bhedena grahaņārtham, it is evident that he takes ‘bheda’ as the meaning of bilma. But then when he comes to the second nirvacana, bhāsana, of bhilma available in his commentandum, he does not go beyond saying that bhāsana stands for ‘illumination of the meaning of the Veda through the knowledge of the Vedangas.’ He does not expand on how this fits the Nirukta sentence in which bilma-grahaņāya and imam grantham (= Nighantum) as well as vedam occur. A sentence like vedanga-vijnana-prakasita vedārtha-grahanāyemam grantham samāmnāsisur vedam ca vedangāni ca is logically possible, but it would be awkward.

126 The syntax of several sentences is problematic in Sarup’s edn, despite much dedicated work on his part. Where a mere change in Sarup’s punctuation removes the problem, I have silently changed the punctuation, as in dealing with Bhadkamkar’s and Rajavade’s edns of Durga’s commentary.

1271 am not certain about how this word is related to the preceding; tat could refer to mantra-brāhmaṇa of the earlier sentence and give us the compound mantra-brāhmaṇa-sāksāt kytah, meaning ‘seen with the help of mantra-brāhmaṇa, (implying) not directly seen because dharma is imperceptible.’ However, such a compound sounds unnatural in the company of the following words. On the other hand, tat taken in the sense of tasmāt’therefore seems superfluous. The next sentence would be natural if confined to dissolution only of the com pound s-k-d.

Veda Revelation according to Bharty-Hari, appendix 3

adhītam supta-pratibuddha-nyāyena mantra-brāhmaṇam smaranti. kaścit kimcid yo yat smarati tat ’tena dystam,‘’tena sāksāt-kytam,‘’tena proktam,’ ’tasyārşam’ iti cocyate. yasya yāvad ārşam tena tāvad eva sākṣāt-kytam. anyat tu tenāpi yat sākṣān na kytam tad upadesenaivādhigatam.128

na ca janmāntarānubhūtam niyamena na smaryate. drsyante hy adyatve ‘pi jāti-smarāh kimcit smarantah.

te carsayo yady api prati-systy anye ’nya utpadyante tathāpy atīta-systi-kyta punya-visesa-vaśāt tat-karmāņas tan-nāmānas cotpadyante. tenaikasyām systau viśvāmitra-nāmnā yat smrtam systy-antare ‘pi viśvāmitra-nāmaiva tat smarati. ato nityatve ‘pi vedasya nārşa-vyapadeśasya nāpi sākṣāt-karaṇasya kaścid virodhah.

etad abhipretyaitad ucyate sāksāt-krta-dharmāņa rsayo babhūvuḥ iti.

te ‘varebhyo ‘vara-kālīnebhyaḥ śakti-hīnebhyo ’tīta-systi-krta-punya višesābhāvāt29 asāksāt-krta-dharmabhya upadeśena śisyopadhyāyikayā vrttyā mantrān granthato’rthataś ca samprāduḥ. teșu hi pūrva-systāv adhīteșu teşām upadeśa-mātreņaiva smộtir babhūva, yathedānīm chando-nastam gaņayataḥ 6 °naste guṇavatah/punyavatah?) kasyacit. 130

128If this phrase is accepted as it is printed, S-M would come across as admitting the presence of first-degree rşitva and śrutarșitva (82.18, Wezler 2001: 221) in the same person. If the phrase is split differently and printed as “dese naivā”, the meaning conveyed would be: “Anything else which has not been directly seen even by him has certainly not/has never been received in the line of Vedic) instruction.” It is difficult to see how S-M could make a bold assertion like this or why they would find it necessary to make it at all. Secondly, from the use of ata eva in the explanation of samāmnāsiṣuḥ below (note 135), we can gather that the first degree seers were capable of providing some Vedānga content too in S-M’s understanding. One would expect that they would have to receive some instruction in the meaning of (a part of) the Veda (in an earlier life) to be capable in this way. Therefore, Sarup’s way of reading the phrase here should be allowed to stand. S-M, evidently, look upon the s-k-ds as well as the avaras as persons who had exposure to the Veda in the past (cf. teșu hi … kasyacit below) and upon the Veda as an ever-existing entity.

129 Following ms A, Sarup reads saktibhāraiḥ after this word. His ms C has saktihārera and ms. B none. As nothing is lost in terms of meaning by not reading the word and as it could have come about from a redundant writing of a part of sakti hinebhyo occurring a little before, I have decided to drop it.

130S-M do not think of the avaras as radically different from the s-k-d rșis. Both are involved in shaping the Veda and composing the Vedāngas. Both have previous exposure to mantras. Both have religio-spiritual merit (punya) to their credit. However, the merit of the s-k-ds is something special. That is why, whereas the avaras need instruction to remove the blockage of their memory and receive the mantras into their being, the-s-k-ds do not. On this background and assuming that S-M use chandas in the sense of ‘Vedic recitation tradition’, I would take the intended analogy to be this: The avaras come to possess the mantras as soon as they receive instruction, very much in the manner in which a virtuous/meritorious person recalls a text lost in the Veda recitation tradition. The implication of the analogy would be twofold; (a) Recov ery of lost Veda parts can take place at the hands of persons strong in religio-spiritual merit. (b) The Veda is not revealed only once or all at once at the beginning of creation. It is eternally present, but it reveals itself only to those who have exceptional religio-spiritual quality. Such an interpretation would need the emendation proposed. While it is contextually and transcrip

mantra-grahanam cātra brāhmaṇānām pradarśanārtham. mantran brāhmaṇānīti samprāduh.131

upadeśāya glāyanto ‘vare. te ‘py avaratamebhya [/“tarebhya] upadeśāya upadeśārtham132 glāyanto, “glai mlai harşa-kşaye” [Dhātu-pātha 928, 929], upadeśa-mātrena grahītum → grāhayitum?) aśaknuvantas tad-anukampayā ksiyamāna-harşās tān anukampamānā ity arthah.

bilma-grahaņāya. bilma upāyaḥ. tena granthasya cārthasya ca grahaņārtham. granthasya grahaņopāyo ‘sta-samdhakena daśa-samdhakenal33 vādhyayanam.131 arthasyopādhyāyāt punaḥ punaḥ śravaņam. vedānga-jñānam ca.

imam grantham samāmnāsişur, vedam ca vedāngāni ca. yad yasya sākṣāt krta-dharmaņa ārşam tat tasmād upaśrutya krtsnar vedam grantha-grahaņārtham samāmnātavantah, artha-grahaņārtham [ca?) vedangāny ata135 evopaśrutya

tionally probable, it cannot be accepted as certain until the mss are restudied or new mss furnish readings in its favor.

131 Here, S-M seem to have used brāhmana in the sense ‘brahman-associated, (a text) following prayers in the Samhitās, one that comments on the mantras or sūktas (especially from the point of view of their hidden meaning and application in ritual worship).’ cf. appen dix 1, point 2, note b; also Durga’s amuşmāt … pasyanti above. S-M frequently express Durga’s (and their other predecessors’) thoughts differenly.

132 Sarup’s mss B and C do not contain upadeśāya. If haplography or dittography is consid ered possible, the reading of ms A, which is simply upadeśārtham, would be better. Since upadeśāya is included in the pratīka part, there is no need to repeat it after only three words.

133The intended meaning of asta-samdhakena daśa-samdhakena vā is probably through eight repetitions or ten repetitions.’ Dr. Parameswara Aithal kindly informs me that in the process of Veda memorization, in which the teacher typically recites a text and the student repeats after him, the repetition takes place 8, 10 or 12 times (the twelvefold repetition seems not to have existed in the time or region of S-M) and that the Kannada term for the Veda memorization process of the specified type is santa/santhi/sante. This information corrobo rates my guess that the term samdhā present in S-M’s remark is related to Marathi samthā.

134 The reading adopted by Sarup is và grahaņam adhyayanam artha”. The word grahanam occurring after vā would be strange, for one cannot say grahanopayo … grahanam (unless the intention is to say something like ’the trick to learning is learning itself,’ which is clearly not the intention of S-M, as they use asta- … vā). Secondly, Sarup informs us that mss B and C do not read daśa-samdhakena … grahaņam. Of this, the omission of daśa-samdhakena could have been due to haplography. At asta-samdhakena the scribe’s eye could have moved to samdhakena beyond daśa-. The presence of grahaņam in the remaining ms A could be a case of redundant writing. In the following part, a phrase like arthasya ca grahanopāyah is needed. The text I have given above is based on these considerations. Contextually, one expects S-M to write phrases having the following structure: granthasya grahanopāyo X, arthasya [/arthasya ca) grahanopāyo Y.’ This expectation is also satisfied by the adopted text.

The absence of ca does not pose a serious problem, and S-M could have expected their readers to import grahanopāyaḥ from the phrase granthasya grahanopāyo.

135 Sarup accepts orgānīty-ata of ms A against “ngāny anyata of mss B and C. I have used the latter reading to the extent of considering iti of the former reading dispensable. What would ity ata convey after a syntactically complete clause artha-grahaņārtham vedangāni? On the other hand, eva would become vacuous if anyataḥ of mss B and C is followed. Apparently, inVeda Revelation according to Bhartr-Hari, appendix 3

svayam ca kalpayitvā. niruktam ca vedāngam tad [… tad] 136 evam āgamikasyedam āgama-kathanam.

ye tu srsti-pralayau necchanti ta etam grantham evam vyācakșate. sākṣād dharma-vacanopadeśa-nirapeksam krtaḥ 37 pratipanno dharmo yais te sākṣāt-krta dharmāna rşayo babhūvuh te ‘varebhyo ‘sāksāt-krta-dharmabhyo veda-vacanād evopadeśa-nirapeksam dharmam pratipattum asamarthebhya ity arthah.upadesena carthasya mantrān brāhmaṇāni ca samprāduh asya mantrasyāyam, asya cāyam ity-evam [/ ayam asya mantrasyāyam aspety-evam] upadeśena manträrtham brāhmaṇārtham ca kathitavanta ity arthah. upadeśāya glāyanto ‘vare ye tu tato ‘py avare veda-vacanāt svayam138 upadeśa-mătrena vā tad-artham pratipattum asamarthās ta ātmana upadeśārtham samāmnātasya sata ‘upādhyāyā artham upadekşyanti, upadiştārthāt [► sțārthās?) tato vedārtham pratipatsyāmaha’ity evam-artham ity arthaḥ. imam grantham gav-ādi-deva-patny-antam samāmnātavantah. vedam dharma-pratipatty-artham. vedāngāni ca anyany api vedārtha-pratipatty-artham.

bilma-grahaņāya. bilma upāyaḥ, tasya nirvacanam bhilmam bhāsanam iti vā iti. bilmam (→ bhilmam?) iti bibharte rūpam, upāyo bibharty upeyam, bhāsate ca prakāsi-bhavati tat tena.139 bilma-sabdasya cāprasiddhasya loke yad upādānam nirvacana-karanam ca, kvacit tasya mantresu prayogo ‘sti tad-artham. tad 41 darśayisyāmaḥ [Rg-veda 2.35.12:] “asmai bahūnām avamāya sakhye yajñair vidhema namasā havirbhih / sam sānu mārjmi, didhisāmi bilmair, dadhamy annaih pari vanda rgbhih//“142 … didhişāmi … dhārayitum cecchāmi gārhapatyādim

S-M’s view, although the avaras are the authors of the Vedāngas, some content of the Vedāngas comes from the s-k-ds. This would be consistent with S-M’s understanding that the s-k-d rșis impart both the form and meaning of the object of revelation to the avaras. See note 128 above.

136The reading of ms A given by Sarup, dīdīpyante, is too different to be reconciled with that of B and C. It is more likely to be a remnant of some other sentence. The introduction of āgamika is rather sudden. S-M, up to this point in their commentary, have not glossed imam grantham of their commentandum. For these reasons, the problem felt here cannot be solved simply by punctuating the text as niruktam ca vedāngam. tad evam āga’. A relatively long textual loss seems to have taken place.

137The paraphrase of kyta with pratipanna seems odd. I wonder if the original reading was kytaḥ svataḥ, which through haplography was reduced to kytah.

138The reading suśryam of A could have been a corruption of something like suśrūşayoo leading to the phrase svayam susrūsayopadeśa-.

139The original readings here could also have been upāyo bibharty supeyam), upeyam bhāsate ca. prakāsībhavati tat tena, with tat standing for upeyam and tena for upāyena.

140 The reading of Sarup’s mss B and C, modified as he suggests. 14. That is, the fact that it is found in the mantras, albeit rarely.

142I have dropped most of the explanation of this rk in the following reproduction of S-M’s comments. The explanation establishes a fire-kindling context for the rk.

agnim bilmair upāyair jvalanasyādho bhasmani gomaya-nikhananādibhir… 143 evam asmin mantre bilma-sabdasya prayogād ihopādānam nirvacanam caivam-artham drastavyam.

S-M’s understanding of Nirukta 1.20 is not as different from that of Durga as may seem at first blush. Like Durga, they note that dharma cannot be an object of the senses and take imam grantham as standing for the Nighantu. Durga takes witnessing of dharma to mean witnessing the relation between ritual actions and their consequences. The latter take it to mean witnessing of the Mantra-Brāhmaṇa. However, as $2.15 points out, this may only be a difference of phrasing. As note 12la states, Durga too thinks of three groups while explaining Yāska’s passage. The main difference is that Durga interprets the Nirukta remark without referring to the possibility of cyclical creation, and S-M first adopt cyclical creation as the context. Their version of cyclical creation is interesting in that it invokes cyclicity in order to explain how there is knowledge of Mantra-Brāhmaṇa on the part of seers at the dawn of creation. But then the quandary of there being either eternal seers or there being a different set of seers and hence a different Veda collection in each creation presents itself. S-M get over it by adding: Although the seers are born anew in each creation, they perform the same actions and have the same names because of the excellent good karman they acquire in the preceding creation.’

As in some other parts of their exposition of the Nirukta, S-M could have interacted with BH’s thought, directly or indirectly, in their handling of Nirukta 1.20 (compare $4.2 and appendix 1, points 4 and 6 with the observations above). However, as point 7 in appendix 4 states, the evidence of interaction is not strong in the case of Nirukta 1.20.

Durga does not display even feeble signs of interaction.144

Translations and/or interpretations

by modern scholars Muir 1874: 118: “The rishis, who had an intuitive perception of duty, handed down the hymns by (oral) instruction to men of later ages, who had not that intuitive perception. These, declining in their power of giving instruction, compiled this work (the Nirukta), the Veda, and the Vedāngas, in order to facilitate the comprehension of details.”

Belvalkar 1915: 6: “It [= Nirukta 1.20) mentions three distinct periods of intellectual development corresponding roughly to sections 2-5 above

143 Rajavade (1940: 291) remarks that in Rg-veda 2.35.12 the word bilma “probably means Řk [= rc] or food.” He does not state why he thinks so.

144 See appendix 4, point 4 for other observations helpful in understanding the preceding reproductions of the relevant parts of the commentaries by Durga and S-M.

Veda Revelation according to Bharty-Hari, appendix 3

[ = to the sections in which Belvalkar discusses the appearance of the Vedas, the appearance of the Brāhmaṇas and the appearance of the texts meant to help in the preservation and study of the Vedas in their parişads and caraņas, i.e., the appearance of the manuals on phonetics, Pada-patha, pre-Yaska Niruktas, other Vedāngas etc. Fn 1 elaborates on this remark:] sākṣāt-krta-dharmāņa ysayo babhūvuḥ. These are the original “Seers of Mantras.” te ‘varebhyo ‘sākṣāt-krta-dharmabhya upadeśena mantrān samprāduh. These correspond to the authors of the Brāhmaṇic speculations, possibly also to the compilers of the family-books. upadeśāya glāyanto ‘vare bilma grahaņāyemam grantham samāmnāsişur vedam ca vedangāni ca. These are the authors of the Pada-pāțha, the Nighanțu, and other allied works, including possibly the prototypes of our modern Prātiśākhyas.”

Sarup 1921: 20: “Seers had direct intuitive insight into duty. They by oral instruction handed down the hymns to later generations who were destitute of the direct intuitive insight. The later generations, declining in (power of) oral communication, compiled this work, the Veda, and the auxiliary Vedic treatises, in order to comprehend their meaning. Bilma = bhilma (division or illustration).” [Fn 1 at this point: Cf. Muir, op. cit. [= Original texts …] vol. ii, p. 165; vol. iii, p. 118.]

Sköld (1926: 7), who, as far as I could check, does not offer a translation or interpretation of his own of the Nirukta passage, remarks: “Belvalkar is doubtless right, when he says, Systems of Sanskrit grammar, p. 6, that this passage “mentions three periods of intellectual development.” I am not quite sure, however, that the authors of the pada-pāțhas can be put in one line with the authors of the Nighaņțu, for I believe … that the padakāras were no authorities even in the time of the niruktakāra. [Fn l at this point: Weber remarks that even in the time of Patañjali the pada pātha had no higher authority, but was subject to criticism.]”

Falk 1990: 108: “persons who had direct insight into dharma turned into poets (“seers”). They handed down their verses by way of teaching to those who were inferior, (i.e.) who had no direct insight into dharma. These inferior persons were tired of this teaching and arranged this opus, i.e. the Veda and its ancillary literature in order to grasp (or: it with) a bilma.”

Carpenter 1995: 44: “The seers saw the ritual ordinances directly. To others who had not seen the ritual ordinances directly they proclaimed the mantras by way of instruction.“145

Wezler 2001 does not offer a continuous translation but comments on practically all significant expressions in the Nirukta passage, frequently

145 Carpenter does not translate the rest probably because he did not need it for his purpose.

taking into account the views of the scholars to whom I have referred above. I have mentioned several of his distinctive observations in the main essay itself. One additional reaction I should offer is this: I do not agree with Wezler’s (p. 232) proposal that upadesena should be taken as a comitative instrumental, functionally equivalent to constructions like putreņa sahāgataḥ pitā.

APPENDIX 4

TKV 1.5 vis-à-vis Nirukta 1.20 Some observations on the relationship between TKV 1.5 and Nirukta 1.20 have already been offered in $2.2-3, 8 above. A few more observations that have an indirect or limited significance for our main concern will be made below.

In the edns of the Nirukta, avara is the stem in the two padas or words (aparebhyah and apare) found in the TKV’s Nirukta citation. No variant readings are recorded. The glosses of the commentators Durga and S-M confirm that they knew no reading other than avara. On the other hand, the BH tradition is consistent within itself. It has apara as the stem in the citation as well as the non-citation part of the Vrtti. The reading found in the mss of the oldest accessible commentary of the Vrtti, namely Vrsabha’s, is para according to the edn available at present, which is unlikely to be an error for avara/apara, since in that case the last syllable of the preceding anyebhyaḥ would have been changed to bhyo and the phrase would have become anyebhyo ‘varebhyaḥ/“parebhyaḥ. However, para is obviously closer to apara.

Further, when Helā-rāja cites the Nirukta passage under TK 3.1.46, he cites it thus: sākṣāt-kyta-dharmāņa ļşayo babhūvuḥ. te ‘parebhyo ‘sāksāt-krta dharmabhya upadešena mantrān samprāduḥ. apare upadeśāya glāyanto bilma grahaņāyemam grantham samāmnāsişur vedam ca vedangāni ca. 46

Such consistency in textual variation - one side has only the avara forms and the other only the apara forms — is unusual. Normally, ms traditions show random variation. Here, we have absence of variation despite the fact that the confusion of pand vis very common in mss (p and vhave similiar appearance in most Indian scripts), that both avara and apara would not conflict with the immediate context, that BH and his commentators were almost certainly acquainted with the Nirukta and that the s-k-d passage

146 The readings my readers will find in Subramania Iyer’s edn conform to the printed edns of the Nirukta. In this instance, lyer has acted unlike a critical editor. He should have allowed the Prakīrņaka-prakāśa mss to speak for themselves. The apara readings have the support of the largest number of geographically spread-out mss. On this objective criterion of textual criticism, Iyer should have given preference to them.

Veda Revelation according to Bhartr-Hari, appendix 4

must have been one of the most commonly mentioned passages in circles of Brahmanic thinkers.

The Yukti-dipikā citation of the passage under Sāmkhya-kārikā 51 (Wezler-Motegi edn pp. 251-252) is made after explaining the ūha/tāraka and sabda/ sutāra siddhis. It is made to explain the adhyayana/tārayantam siddhi. It, too, displays apara:147 yadā tv anyopadeśād apy asamarthaḥ pratipattum adhyayanena sādhayati, sā tytīyā siddhis tārayantam ity [/ity apy] apadisyate, tad etat tāraņa-kriyāyā adyatve ‘py avyāvyttatvān mahā-vişayatvāt (/°şaye) tārayantam ity apadistam, ta ete trayaḥ sādhanopā<yā> yairl4s ā brahmanaḥ prāņino ‘bhipretam artham prāpnuvanti. āha ca “sāksāt-kyta dharmāņa rsayo babhūvuḥ. te ‘parebhyol19 ’ sākṣāt-krta-dharmabhya upadesena mantrān samprāduḥ. upadeśāya glāyanto ‘pare bilma-grahaņāyemam grantham samāmnāsişur vedam ca vedangāni ca” [Nir. 1.20] iti, bilmam bhāsanam samyak pratibhāsasya višistaḥ samketa uktah.

Nirukta 1.20 is also quoted in the Nārāyaṇīya commentary on Kaiyata’s Mahābhāşya-pradīpa 6.3.109 (prşodarādīni yathopadistam). The mss of this commentary used by the editor M.S. Narasimhacharya, p. IX. 260) read aparebhyoand apare, as in the TK tradition and the Yukti-dīpikā. The editor has, however, changed the readings to avarebhyo and avare, probably to make them agree with the Nirukta edns. After citing the Nirukta passage, the commentator Nārāyaṇa indicates that he took the passage as implying three groups of Veda recipients: anena veda-vedangadişu tri-vidhā adhikarina uktāh. tathā vyākhyātam [source not specified] “prathamāḥ pratibhānena, dvitīyās tūpadeśatah/ abhyāsena tytīyās tu vedārthan pratipedire //”iti.

The support these outside citations give to the feeling the TK tradition generates, namely that apara must be the genuine reading for that tradition, is significant. avara has a strong association with ’later’ and ‘inferior’ (Wezler 2001: 218-223 ), whereas apara does not.150 One would thus be justified in

147 I have not recorded the obviously insignificant variations noted by Wezler-Motegi. Given the precarious survival of the YD, they needed to record such ms variations. We can overlook them for our present purpose.

148 Wezler-Motegi fn 13: “All the Mss read sādhanopāyair.” The emendation effected by the editors through the addition of yā is transcriptionally probable and hence very good. They should have only separated yair by a space from sādhanopā[yā).

149 Ms D: parebhyo [AA: i.e, as S. Iyer’s edn of Vrşabha has it].

150 (a) Recall the glosses of Durga and S-M: avara-kālīnebhyaḥ sakti-hīnebhyaḥ “to those belonging to a later time, lacking the necessary) capacity.” A qualitative gradation is not generally suggested by apara. When in certain traditions it implicitly incorporates a qualitative judgement, it in fact functions as an antonym of avara, connoting acceptability and respect (cf. Helā-rāja 3.14.615). As for connoting anteriority or posteriority, it draws a blank. Spatial distinction is also not a natural part of its meaning as it is in the case of para in some occurrences.

suspecting that the ‘avara: apara’ variation reflects a consciously and/or carefully maintained tradition.151

Why would such a tradition be maintained? Perhaps because there was historical memory preserved that, in the view of the Yāskīya Nairuktas, the recipients of the mantras from the s-k-ds were to be ranked lower and/ or that BH and certain other thinkers like the Yukti-dipikā author did not favor the suggestion that the recipients were inferior. Their resistance to such a suggestion becomes all the more probable because the recipients in this case were the arrangers of the Veda and composers of the Vedāngas. They had been venerated with such epithets as śrutarși. Also, BH’s temperament as a thinker is that of a perspectivist. He tries, as far as possible, to preserve different ways of solving grammatical and philosophical problems by assigning those ways to the appropriate levels and contexts. Using adjectives of negative import is not his style.

The Nirukta passage concerned does not speak of s-n-a-vāc (it can at the most be said to have presupposed such a vāc) 152 or the mantras as what the s-k-ds see. The seeing of the mantras, however, can be read in it by implication. Since the s-k-ds later transmit mantras, they must have somehow come to possess them in the first place. However, theoretically, it is possible that what was directly seen by the seers in Yaska’s view was only the dharmas/ dharmans. As a result of that or independently of that, they came to possess the mantras (in a way which was familiar to Yaska’s readers and hence he did not feel the need to clarify).

Relatable to the preceding possibility is the possibility that Yaska intended sākṣāt-krta-dharmāṇaḥ to be the predicate, a possibility discussed

(b) Rajavade (1940: 290) comments: “Durga calls these inferior men śrutarsi because they heard first and then saw what the Veda was…. But I think Yāska did not mean that these inferior men were Rsis.” There is no suggestion in Durga’s use of śrutarsi that the avaras later saw the Veda or that they saw the way the s-k-ds saw it. However, I think that Rajavade is correct in suspecting that the avaras may not have been rșis in Yaska’s view.

151 The other possibilities are (a) that there were different versions of the Nirukta text at an early time and (b) that the Nirukta is not the direct source for the TKV author. But these possibilities lack external corroboration. Unless the evidence extends beyond Nirukta 1.20, we cannot entertain them seriously. The farthest we can justifiably go in conjecturing in the current state of our information sources is that a ms tradition of the Nirukta or of a text quoting the Nirukta that was accessible to the TKV and YD authors read apara instead of avara or wrote vin such a way as to be mistaken for p.

152 Yaska cites some vāc verses of the RV that speak of the difference between knowing vāc superficially and knowing it really as it is. If the evidence in the Nirukta-parisista is admitted, as I think it should be, it almost becomes a certainty that Yaska knew a vāc philosophy developed from the RV times.

Veda Revelation according to Bharty-Hari, appendix 4

but not accepted by Wezler (2001: 229-230).153 The meaning of the first two sentences in the Nirukta passage then would be: “Seers (at a distant past) became/were (persons) who had directly seen dharma(s)/ dharman(s). (As a result, they could compose mantras, utterances that can affect physical reality). Through teaching, they have entrusted (note 43) the mantras to some inferior (or later) persons who had not directly seen the dharma(s)/dharman(s).”

However, tisunlikey thatsākṣāt-kyta-dharmāṇaḥ was meant to be a predicate. Rşis or seers are so called because they have seen something that ordinary people do not see. If sākṣāt-krta-dharmatva is predicated of them, it would be something over and above that which, after being seen, made them seers. Therefore, “Seers became s-k-d” is unlikely to be said unless there is a specification in the context of what made them seers in the first place. But there are no syntactically or conceptually connected sentences before the sentence sākṣāt-kyta-dharmāņa rşayo babhūvuḥ. That very sentence introduces a new topic. Yāska’s stance must, therefore, be one of assuming that his readers know what makes a person a rși or what the different types of rșihood are.

Under the first alternative, sākṣāt-krta-dharmāņa rsayo babhūvuḥwould, in effect, mean “Those whom you readers know as seers became s-k-d.” Yāska’s readers would then expect him to tell what the difference is, that is, to provide some description of what sākṣāt-krta-dharmatva consists in. Such a description is missing.

If, on the other hand, Yāska expected his readers to know the different possibilities through which one could become a rsi, his statement “seers became s-k-d” would be odd in the subject part. Contrary to our expectation that he would refer to a seer type distinct from the s-k-d type, he would come across as referring to seers in general. The discontinuity of thought between the first sentence and the second sentence would also continue to afflict the interpretation.

Thus, the best way in the given hermeneutic situation is to assume that Yāska expected his readers to know the connection between sākṣāt kệta-dharmatva and the ability to impart mantras and that his first sentence is to be understood as “s-k-d rșis became/were” → “There came to be/ there were s-k-d rsis” > “Seers who had seen dharma came about (were born, appeared on the scene),” that is, with sākṣāt-kyta-dharmāṇaḥ as an attributive adjective in the subject part.

If the connection between dharma and mantra is presumed in the passage as known, Yāska must be assumed to know it too — in other words,

153 Wezler rightly rejects Falk’s attempt to take the first sentence in the sense “Persons who had direct insight into dharma became poets.”

to have a ’theory of language that connected dharma and mantra (82.3). On the mantra side the connection is obvious, since any mantra is a realization in language. On the dharma side, the connection is anything but obvious. At least a few elements of the thought of BH and his grammarian predecessors on the relation between vāc and dharma must have been known to Yaska and his readers.

Despite the absence of s-n-a vācin Nirukta 1.20, therefore, there is unlikely to be a great distance or essential difference between how BH saw the religio-spiritual universe and how Yaska saw it.

The object of transmitting, as well as that of seeing, is vācin BH’s statement. In Yaska’s, it is mantras The employment of mantradrśaḥ shows BH’s awareness of the implication of mantrān samprāduh (“the rșis must have come to possess the mantras if they conveyed them to others’) in the Nirukta. This implication could have allowed BH to choose mantras as the object of pravedayisyamāṇāḥ as those who will reveal/ convey’. Yet it is tām, referring to vāc, that figures as the object of transmitting in his statement. He must have understood mantras as essentially identical with the s-n-a vāc as having a diference of extent but not of kind.

Just as the relation between dharma (or the sāksātkāra of dharma) and mantra is not specified in the Nirukta, the relationship between the mantras and bilma is left unspecified. Depending on how we interpret bilma and, partly, on how we dissolve the compound bilma-grahaņāya, it could be the s-k-ds or the second group members characterized as a-s-k-ds who fashion the bilma. With bilma-grahaņāya as a genitive tat-purusa, the implication would be that the bilma has already been created by the s-k-ds and the a-s-k-ds are supposed to grasp it. On the other hand, if bilma-grahaņaya is taken to be an instrumental tat-purusa, the implication would be that the a-s-k-ds have access to the bilma as an instrument — (a) as something created by others for them or (b) as something they created — for their own use or (c) for enabling others to grasp the mantras.

Under (a), the instrumentality between the mantras and bilma, on the one hand, and the relationship between bilma and the referents of imam … vedam ca vedāngāni ca would remain unspecified. Questions about how the first instrumentality is different from the instrumentality of upadesa mentioned in the preceding sentence, about why and how the s-k-ds give something in addition to the mantras and about the connection between instruction fatigue and the bilma remedy for it would arise. 154 If what bilma refers to is co-extensive with the referents of imam … ca, then the question

154 Consequently, context would offer less help to us in the direction of determining the meaning of bilma.

Veda Revelation according to Bharty-Hari, appendix 4

about whether the s-k-ds can be thought of as the authors of the Vedāngas would also present itself. Our certainty would not extend beyond the understanding that bilma is something situated between the mantras and the referents of imam … ca.

There will be no essential difference between (b) and (c). The avaras, who have received mantras, can always (or relatively easily) use for the benefit of others what they have created for their own benefit. What would make a difference would be the following considerations: (i) How much is included in upadeśena mantrān samprāduh? What is it precisely that creates the need for grasping with/through the bilma? (ii) How is the dative in upadeśāya to be understood? (iii) Whose fatigue is it? Who are the locus of it? The ones who received the mantras from the s-k-ds or those a-s-k-ds who are continuing the process of instruction directing it toward the later generations? (iv) Should we proceed on the assumption that the referent of bilma is co-extensive with the referents of imam … ca or should we assign some such general meaning as ’efficient method’ to bilma? The answers to these questions, in turn, would depend on how we answer the questions about why the Nirukta passage has been composed and why it appears where it does (see appendix 3, point 1).

The traditional commentators like Durga and S-M must have asked the specified questions of themselves. Their answers show careful thinking and excellent judgment. They recognize that the situation calls for inclusion of ‘meaning explanation’ in the sense of upadeśa and that, if the sentence te … samprāduh conveys success in transmitting to the a-s-k-ds, the extra means or methods must have been felt necessary in the case of those who came after the successful a-s-k-ds — that the means and methods must have been developed by the a-s-k-ds for use in the case of their students, that we have to think of three groups, (although the Nirukta speaks of only the s-k-ds and the a-s-k-ds), that bilma is best understood as referring to a generally settled (if not written) text of the mantras (Veda) and the Vedāngas (see appendix 3).

BH’s words leave no doubt that in his view the bilma creators are the s-k-ds. They are the grammatical subject of his verb samamananti, of which the object is bilmam. He views bilma as something that is involved in the process of transmission of the s-n-a vāc and as anukāra of sabda-tattva brahman itself. It must, therefore, stand for what the a-s-k-ds receive. As, at the other end, he does not indicate any disapproval of Yaska’s words … avare/apare bilma-grahaņāya imaṁ … ca, his bilma must be related to the Veda but different from it, an object coming into existence at a stage preceding the composition of the Veda text by the a-s-k-ds. He does not have to get into the question of whether this composition is for the use of

the a-s-k-ds or for their students. Nor would the persons suffering from instruction fatigue be persons other than the a-s-k-ds in his understanding. Consequently, if he needs to understand bilma-grahaņāya as an instrumental tat-purusa, it can only be done by attaching a sense of manner or adverb (‘for grasping easily/ efficiently,’ ‘in order to grasp in a manageable/ shorter form’). But he has clearly taken bilma in the objective sense of ‘anukāra’. Therefore, the dissolution of bilma-grahaņāya presumed by him must have been a genitive tat-puruşa dissolution (bilmasya grahaņāya). 155

A theory, a distinctive theory at that, of how extraordinary perception takes place was known to BH (cf. $3.6). Such a theory might not have been known to the author of the Nirukta.

The use of the present tense forms in BH’s statement in the place of the perfect and the aorist forms of the Nirukta passage indicates that BH thought of creation (and of dissolution) as recurrent. Yāska does not provide evidence of being under the influence of such thinking at least when he wrote the sentence with which we are concerned (the Nirukta-parisista may turn out to be an exception). 156 Consequently, Yaska’s three sentences have the tone of reporting a one-time historical or mythic event, whereas BH’s words have the tone of making the general statement, but this does not mean that Yāska must be opposed to repeated creation of the universe.

In the case of Nirukta 1.20, S-M have not explained the details of the Nirukta statement in the light of what is found in BH’s work or in the commentaries thereto. Their explanation does not introduce the notion of the subtle form of language. The way they paraphrase bilma is different. The reference they make to sleeping in supta-pratibuddha is not at all like BH’s reference to svapna. An extension of BH’s mantra to mantra-brāhmana is seen in their comments. Evidently, the Nirukta commentators’ tradition of understanding the part cited by BH was different from that of BH’s

155 Vrsabha glosses the compound bilma-grahaņāya with another compound expression, praticchandaka-grahaņāya. As a result, we do not know whether he understood Yaska (and, indirectly, BH) as employing a genitive tat-puruṣa or an instrumental tat-puruṣa.

156 Durga does not impute the sarga-sthiti-laya way of thinking to Yāska. S-M recognize the possibility of being able to do so (see appendix 3). They could have been aware that a recurring creation does not rule out the possibility that an author living in one creation may speak of an earlier age in that creation as a historical fact and with a perfect tense form. Such an author has no need to watch his tenses unless the context consisted of a question put to him: “Do you, Philosopher P, think that this world is made again and again?” or “Philosopher P, is the world like things of Play-Doh that children make or is it like some structures that are meant to last for ever (that the structures do not, as the dictum yat kytakam tad anityam says, is another matter.)91

Veda Revelation according to Bhartı-Hari, appendix 4 commentators, and the two have probably been kept apart deliberately (cf. point l above). It is also possible that BH has preserved for us an older and a philologically and philosophically sounder understanding of the Nirukta statement. This understanding does not necessarily conflict with the understanding reflected in the comments of Durga and S-M ($2.15), but it behooves us as historians to note that there is a different understanding.

BIBLIOGRAPHY AND ABBREVIATIONS Aklujkar, Ashok (1982): “Interpreting Vākyapadīya 2.486 historically

(part 2).” Indological and Buddhist Sudies — Volume in Honour of Porfessor J.W. de Jong on His Sixtieth Birthday, pp. 1-10. Ed. Hercus L.A. et al. Canberra: Faculty of Asian Studies [Australian National University]. -, (1991a): “Bhartr-hari’s concept of the Veda.” In Panels of the VIIth World Sanskrit Conference, Volumes IV and V, pp. 1-18. (eds) Bronkhorst, Johannes; Deshpande, Madhav M. Leiden: E.J. Brill.

, (1991b.): “Syntactic Gleanings from Bhartr-hari’s Trikāņdī.” In Studies in Sanskrit Syntax, pp. 1-11. Ed. Hock, Hans Henrich. Delhi: Motilal Banarsidass. - , (1994): “The Tīkākāra mentioned by Skanda- Maheśvara.” In Vācaspatyam. Pt. Vāmanshastri Bhāgwat Felicitation Volume, pp. 9-25. Eds. Bhate, Saroja; Deshpande, Madhav M. Pune: Vaidika Samsodhana Mandala. -, (1999): The Theory of Nipātas (Particles) in Yaska’s Nirukta. Pune: Bhandarkar Oriental Research Institute. Post-graduate and Research Department Series No. 42. Pandit Shripad Shastri Deodhar Memorial Lectures, sixth series. -, (2001): “The word is the world: nondualism in Indian philosopy of language.” Philosophy East & West 51.4: 452-473.

, (2004): “Can the grammarians’ dharma be a dharma for all?”

Journal of Indian Philosophy 32: 687-732. Alper, Harvey P. (1989): (ed) Understanding Mantras. Albany: State University

of New York Press. This book, consisting of 540 pages, has a shorter edn consisting of 343 pages, titled Mantra and published by the same publisher in the same year. Its bibliography pages are renumbered. Indian edn of the larger edn, 1991: Delhi: Motilal Banarsidass. Indian edn of the shorter edn, 1997: Delhi: Satguru Publications.

a-s-k-d = asākṣāt-krta-dharman. Belvalkar, Shripad Krishna. (1915): An Account of the Different Existing Systems

of Sanskrit Grammar … Poona: Author. Second revised edn 1976. Delhi: Bharatiya Vidya Prakashan. I do not think that this is really a

second ’edition’ or is revised. BH = Bhartr-hari. Bhadkamkar: see under Yaska. Bhagavad-datta (1974-78): Vaidika Vanmaya kā Itihāsa. 3 vols. (in

unexpected reverse chronological order: 1978, 1976 and 1974) out of 5 projected vols. Expander and editor: Satyaśravā. Nai Dillī: Praņava Prakasana. Hindi. Originally published: Vol. 1, Pt. 1: 1927.

Vol. I, Pt. 2: 1931. Lahore: Dayananda Anglo-Vedic Studies. Bhandarkar, Ramakrishna Gopal (1868): Second Book of Sanskrit. My access

to the Preface of the first edn of this title is through the twenty second edn published in 1952 by Karnatak Publishing House, Bombay. The twenty-second edn is probably a reprint of the sixth

revised edn. Bharati, Agehananda (1965): The Tantrik Tradition. London: Rider and

Co. Reprint 1970: Garden City, N.Y.: Anchor Books. Reprint 1975:

New York: Samuel Weiser. Bhartr-hari. Mahābhāsya-tīkā. Published under the title Mahābhāsya-dīpikā

of Bhartr-hari. Ed. Abhyankar, K.V. Other ed. V.P. Limaye. Poona: Bhandarkar Oriental Research Institute. 1967-1970. Post-graduate

and Research Department Series No. 8. Bhartr-hari. Trikāndī or Vākyapadīya. I have followed the enumeration of

kārikās in Rau’s edn and reproduced the text of the Vrtti from my critical edn under preparation. Those wishing to verify my references to the Vștti prior to the publication of my edn should consult the edns by Subramania Iyer. The numbers in my edn and those in the edns by Subramania Iyer do not always match. However, they are not far removed from each other. The specifics of the edns I have in mind are: (a) Vākyapadīya of Bhartr-hari with the Vrtti and the Paddhati of Vrsabha-deva. Ed. Subramania Iyer, K.A. Poona: Deccan College Postgraduate and Research Institute. 1966. Deccan College Monograph Series 32. (b) The Vākyapadiya of Bhartr-hari, Kanda II with the Commentary of Punya-raja and the Ancient Vitti. Ed. Subramania lyer, K.A. Delhi: Motilal Banarsidass, 1983. (cl) Vākyapadīya of Bhartr hari with the Commentary of Helā-raja. Kānda III, Part 1 [Samudeśas 1 7]. Ed. Subramania Iyer, K.A. Poona: Deccan College Postgraduate and Research Institute. 1963. Deccan College Monograph Series

Veda Revelation according to Bharty-Hari

(c2) Vākyapadīya of Bhartr-hari with the Prakīrņaka-prakāśa of Helā rāja. Kānda III, Part II [Samudeśas 8-14]. Ed. Subramania Iyer, K.A. Poona: Deccan College, 1973. [Continuation of Deccan College Monograph Series no. 21?]. (d) Bhartr-hari’s Vākyapadīya, Die Mula kārikās nach den Handschriften herausgegeben und mit einem Pāda-Index versehen. Ed. Rau, Wilhelm. Wiesbaden: Kommissionsverlag Franz Steiner GMBH. 1977. Abhandlungen für die Kunde des Morgenlandes (Monograph Series of the Deutsche Morgenländische

Gesellschaft) XLII, 4. Biardeau, Madeleine (1964): (ed, tr.). Vākyapadīya Brahma-kānda avec la

Vrtti de Hari-vrsabha. Paris: Editions E. de Boccard. Publications de

l’Institut de Civilisation Indienne. Série IN-8°. Fascicule 24. Brereton, Joel P. (2004): “dhárman in the Rg-veda.” Journal of Indian

Philosophy 32: 449-498. Bronkhorst, Johannes (2004): From Pāṇini to Patañjali: the Search for Linearity.

Pune. Bhandarkar Oriental Research Institute. Post-graduate and

Research Department Series No. 46. Carpenter, David Wesley. (1985): “Revelation and experience in Bhartr

hari’s Vākyapadīya.” Wiener Zeitschrift der Kunde Südasiens 29: 185-206. - , (1995): Revelation, History, and the Dialogue of Religions: a Study of Bharty-hari and Bonaventure. Maryknoll, New York: Orbis Books. Originally a Ph.D. dissertation submitted to the Divinity School, University of Chicago, in 1987 under the title The Light of the Word: a

Comparative Study of the Phenomenon of Revelation. Despande, Madhav M. (1992): “Justification for verbal root suppletion in

Sanskrit.” Historische Sprachforschung 105.1:18-49. Dimmit, Cornelia (1978): Classical Hindu Mythology: A Reader in the Sanskrit

Purānas. Other author: J.A.B. van Buitenen. Philadelphia: Temple

University Press. Durga: see under Yāska. ed = editor, edited by. edn = edition. Eliot, Charles (1921): Hinduism and Buddhism: An Historical Sketch.

London. 7 books/vols. Reprint 1977 by Curzon has the 7 books in 3

vols. Falk, Harry (1990): “Goodies for India. Literacy, Orality, and Vedic Culture.”

In Erscheinungsformen kultureller Prozesse (Script Oralia, 13), pp. 103 120. Tübingen: Gunter Narr Verlag.

Falk, Harry (1993): Schrift in alten Indien. Ein Forschungsbericht mit Anmerkungen.

Tübingen: Gunter Narr Verlag. Fitzgerald, James L. (2004): “Dharma and its translation in the

Mahābhārata.” Journal of Indian Philosophy 32:671-685. fn = footnote. Hacker, Paul. (1972): “Notes on the Māndūkyopanişad and Sankara’s

Agama-śāstra-vivarana.” In India Maior, Congratulatory volume Presented to J. Gonda, pp. 115-132. Ed. Ensink, J. Other ed: P. Gaeffke. Leiden:

Brill. Reprint: Kleine Schriften, pp. 252-269. Halbfass, Wilhelm (1988): India and Europe: an Essay in Understanding. Albany:

State University of New York Press. ., (1991): Tradition and Reflection: Explorations in Indian Thought.

Albany: State University of New York Press. Helā-rāja: see under Bhartr-hari. Holdrege, Barbara (1994): “Veda in the Brāhmaṇas: Cosmogonic Paradigms

and the Delimitation of Canon.” In Patton 1994, pp. 35-66. Horsch, Paul (2004): “From creation myth to world law: the early history of

dharma.” Journal of Indian Philosophy 32: 423-448. Translated by Jarrod L. Whitaker. Originally published in 1967 as “Vom Schöpfungsmythos zum Weltgesetz” in Asiatische Studien: Zeitschrift der Schweizerischen

Gesellschaft für Asiankunde 21: 31-61. imam … ca = imam grantham samāmnāsişur vedam ca vedāngāni ca. Kahrs (1998): Indian Semantic Analysis: the Nirvacana Tradition. Cambridge,

U.K.: University of Cambridge. University of Cambridge Oriental

Publications 55. Kane, Pandurang Vaman (1973): History of Dharma-śāstra (Ancient and

Medieval Religious and Civil Law).5 vols. Poona: Bhandarkar Oriental Research Institute. My reference is to vol. 3, second edn, published

by the original publisher. Larson, Gerald James (1987): Ed. Samkhya: a Dualist Tradition in

Indian Philosophy. Other ed: Ram Shankar Bhattacharya. Princeton: Princeton University Press. Also, Delhi: Motilal Banarsidass, Encyclopedia of Indian Philosophies, vol. IV. General Ed.: Karl H.

Potter. Mayrhofer, Manfred. 1956, 1963, 1976, 1980, Kurzgefasstes Etymologisches

Wörterbuch des Altindischen. 4 vols. Heidelberg: Carl Winter. Universitätsverlag. Indogermanische Bibliothek. II. Reihe. Wörterbücher.

Veda Revelation according to Bharty-Hari

Medhātithi. Manu-smrti with Commentaries of Medhātithi, Sarvajña-nārāyana,

Kullūka, Rāghavānanda, Nandana, Rāma-candra, and Govinda-rāja.

Ed. Mandlik, V.N. 3 vols. Bombay: Ganapat Krishnaji’s Press. 1886. Mitchiner, John E. 1982. Traditions of the Seven Rșis. Delhi: Motilal Banarsidass. ms = manuscript. Muir, John. 1868-1872. Original Sanskrit Texts on the Origin and History of the

People of India, Their Religion and Institutions. 5 vols. London: Trübner. Revised edn of vol. 3 in 1874. Reprint of all vols. 1967: Amsterdam:

Oriental Press. Nārāyaṇa Commentary on Kaiyața’s Mahābhāsya-pradīpa. Mahābhāsya

pradīpa-vyākhyānāni, (ed) M.S. Narasimhachārya. Pondichéry:

Institute Français d’Indologie. 1982. O’Flaherty, Wendy Doniger. 1975. Hindu Myths: A Sourcebook Translated From

The Sanskrit with an Introduction. Penguin Classics. Oliver, Curtis F. 1979. “Some aspects of literacy in ancient India.” The

Quarterly Newsletter of the Laboratory of Comparative Human Cognition

1: 57-62. Reference from Falk 1990. Padoux, André. 1990. Vāc: the Concept of the Word in Selected Hindu Tantras.

(tr.) Gontier, Jacques. Albany: State University of New York Press. Patañjali. Vyākarana-mahābhāsya. Ed. Kielhorn, Franz Lorenz. 1880-1885.

Revised third edn. K.V. Abhyankar. Pune: Bhandarkar Oriental

Research Institute. Patton, Laurie L. 1994. (ed) Authority, Anxiety and Canon: Essays in Vedic

Interpretation. Albany: State University of New York Press. Rajavade: see under Yāska. Ruegg, D. Seyfort. 1994. “Pramāņa-bhūta, *pramāņa-(bhūta)-puruşa, pratyakşa

dharman and sāksāt-kyta-dharman as epithets of the rsi, ācārya and Tathāgata in grammatical, epistemological and Madhyamaka texts.” Bulletin of the School of Oriental and African Studies 57: 303-320. Most of the content is available in French in Ruegg 1994, Wiener Zeitschrift für die Kunde Südasiens 38: 403-420, “La notion du voyant et du

“connaisseur suprême” et la question de l’authorité épistemique.” RV = Rg-veda, Rig-veda. Sarup: see under Yaska. Šāstrī, Siva-nārāyana. 1972. Vaidika vānmaya mem bhāṣā-cintana. Varanasi,

Delhi: Indological Book House. Hindi. Skanda-Maheśvara: see under Yāska. s-k-d: sāksāt-krta-dharman.

Sköld, Hannes. 1926. The Nirukta, Its Place in Old Indian Literature,

Its Etymologies. Lund: C.W.K. Gleerup. s-n-a: sūkṣmā nityā atīndriyā. Srinivas Sastri. 1976. “ sākṣāt-kyta-dharmāņa rşayo babhūvuḥ. “Samskyta-vimarśaḥ

4 (1-2): 7-10. Not seen. Summary from R.N. Dandekar’s Vedic Bibliography: “seeks to determine the meanings of dharma and rși

with the help of Nyāya-śāstra.” Subramania lyer, K.A. 1965 (tr) The Vakyapadīya of Bhartr-hari with the Vrtti.

Chapter 1, English Translation, Poona: Deccan College Building Centenary & Silver Jubilee Series 26. See under Bhartr-hari for his

edns. TK = Trikāņdī, frequently referred to as Vākyapadīya. See under Bhartr

hari. TKV = Trikāņdi-vrtti. See under Bhartr-hari. tr = translator, translated by. Tripāthī, Rāma-deva. 1976. “Vaidika vānmaya mem bhāṣā-darsana.” In

Bhāratīya bhāṣā-śāstrīya cintana, pp. 167-175. Ed. Miśra, Vidyā-nivāsa

et al. Jaya-pura [= Jaipur]: Rājasthāna Hindi Grantha Akādamī. Vịtti: see under Bhartr-hari. Vrsabha: see under Bhartr-hari. Vākyapadīya: see Trikāņdi. VP = Vākyapadīya. Wezler, Albrecht. 2001. “Some remarks on Nirukta 1.20 sākṣāt-kyta-dharmāna

rşayo, etc.” In The Pandit. Traditional Sanskrit Scholarship in India. Festschrift Parameshvara Aithal, pp. 215-248. Ed. Michaels. Axel. New

Delhi: Manohar. South Asian Studies series. Yāska. Nirukta: (a) (ed) Sarup, Lakshman. 1921. The Nighantu and the

Nirukta … English Translation and Notes. Oxford: Oxford University Press. Also considered a part of the Panjab/Punjab University Oriental Publications Series. Reprint, along with Sarup’s The Nighantu and the Nirukta … Introduction (1920) and The Nighantu and the Nirukta … Critically Edited …. (1927) in one vol.: 1967, 1984 etc.: Delhi: Motilal Banarsidass. (b) (ed) Bhadkamkar, H.M. 1918. The Nirukta of Yaska with Durga’s Commentary. 2 Vols. The second vol. ed. R.G. Bhadkamkar, 1942. Poona: Bhandarkar Oriental Research Institute. Bombay Sanskrit and Prakrit Series, nos. LXXIII, LXXXV.

157

157 Recently, I have come to the conclusion that the form of this name should be “Sri-vrsabha.

Veda Revelation according to Bhartr-Hari Vol. 1 reprint in 1985 by the original publisher. (c) (ed) Rajavāde, Vaijanātha Kāšīnātha. 1921. … Yaska … pranītam Durga … kyta-vrtti sametam Niruktam. Pune: Anandāśrama. Anandāśrama-samskrta granthāvali, no. 88. 2 vols. Offset reprint 1990 by the original publisher.(d) (ed) Sarup, Lakshman. (i) 1928. The Fragments of the Commentaries of Skanda-svamin and Maheśvara. Chapter I. (ii) 1931. Commentary of Skanda-svāmin & Maheśvara on the Nirukta. Chapters II-VI. (iii) 1934. Commentary of Skanda-svāmin & Maheśvara on the Nirukta. Chapters VII-XIII. All vols.: Lahore: Punjab University. Punjab University Oriental Publication Series. Reprint 1982 of (i), (ii) and (iii) in 2 vols. under the title Commentary of Skanda-svāmin & Maheśvara on the Nirukta with Additions and Corrections by Acharya V.P. Limaye. New Delhi: Panini. Pāṇini Vaidika Grantha-mālā 11. (e) Ed. Rajavade, Vaijanātha Kashinath]. 1940. Yaska’s Nirukta. Poona: Bhandarkar Oriental Research Institute. Government

Oriental Series, class A, no. 7. Yudhisthira Mīmāmsaka. 1961 (samvat 2019] and onwards. Samskrta

Vyākarana-śāstra kā Itihāsa. 3 vols. Some in revised edns. Ajmer: Bhāratīya Prācya-vidyā Pratisthāna. Hindi. 1961: Dvitīya Bhāga (= vol. II]. Ajmer: Bhāratīya Prācya-vidyā Pratisthāna. 1962 (samvat 2020]: revised edn. vol. I. Ajmer: Bhāratīya Prācya-vidyā Pratisthāna. 1972 (samvat 2030]. 1984 [samvat 2041] Bahālagadha, Harayāņā: Author. Distributor: Rāmalāla Kapūra Trust.