Origin Of Mimamsa As A School Of Thought A Hypothesis

Author(s): Johannes Bronkhorst
Publisher: Johannes Bronkhorst
Catalog link: https://jainqq.org/explore/269462/1

Page #1

THE ORIGIN OF MĪMĀMSĂ AS A SCHOOL OF THOUGHT: A HYPOTHESIS
Johannes Bronkhorst
Lausanne
We owe to Professor Parpola two illuminating articles on the formation of Mimāmsā! (Parpola 1981; 1994), and at least one further article on this topic is expected from him. In the articles that have so far appeared, Parpola
argued for the original unity of a single Mimämsäsutra ..., which was later split into two: the Pūrvamimamsāsūtra ... ascribed to Jaimini, and the Uttaramīmāmsāsūtra ... ascribed to Bädarāyana. (He) also analysed the teacher quotations of the [Mimamsasutra) and (compared) them with the evidence found in the ritual Sütras of the Veda, (both of the Black Yajurveda (and) the White Yajurveda. (Parpola 1994: 293.)
These two articles, by their very nature and intent, concentrate on the parallels between the Mimāmsāsūtra and the ritual Sūtras, and therefore on the continuity between them. However, Mimāmsă - and from now on I will use this expression primarily to refer to the so-called Pūrvamīmāmsā - is more than merely the outcome of a continuous development of the ideas and concerns which we find in the ritual Sütras. At some period in its history Mimāmsă underwent one or more dramatic breaks with its predecessors, which allowed it to become an independent school of thought.
Two discontinuities in particular deserve attention: (1) The Srauta Sūtras belong, each of them, to their own Vedic schools, and describe the rituals as carried out in those schools; as against this, Mimāmsā claims the unity of ritual practice and the fundamental identity of the ritual acts prescribed in the different schools. (2) Mimāmsă further innovates in introducing and elaborating a number of "philosophical" notions, most important among them the belief in the beginninglessness
I
Parpola speaks of the Mimāmsā; I will simply speak of Mimämsā. Cf. Parpola 1981: 164: "There can be no doubt that the Mimämsäsutra directly continues the tradition of the Vedic ritualists ... The formation of the Mimāmsāsūtra can certainly be reconstructed to a great extent by comparing it carefully with the existing Kalpasūtras."

Page #2

JOHANNES BRONKHORST
(anádiva), authorlessness (apauruseyarva) and self-sufficient validity (svatahpramanya) of the Veda. It seems likely that the attempt at unification that expresses itself in the first discontinuity was the result of an increasingly frequent interaction between at least certain representatives of the different Vedic schools. The second discontinuity - the introduction and elaboration of a number of remarkable "philosophical” notions - may, as I will argue, be accounted for as an attempt to face Critical outsiders.
Mimāmsā never fully replaced the ritual traditions of the Vedic schools. We know, for example, that Bharthari, a philosopher from the 5th century C.E., though acquainted with Mimāmsā, refers for ritual details to the handbooks of his own Vedic school, that of the Mänava-Maitrāyaniyas (Bronkhorst 1985; 1989: 105 (375-376)). Other authors explicitly prescribe that sacrificers should adhere to the manuals of their own schools (Deshpande 1999). The Mimāmsāsūtra itself (2.4.8-9), finally, first records the position according to which there are differences between the rituals in different Vedic schools, then rejects it. All these passages reveal a certain amount of resistance against Mimāms, that was apparently felt by a number of orthodox Brahmins, presumably from the very beginning."
This is not the place to study in further detail the first discontinuity mentioned above. Instead we turn to the second one: the introduction and elaboration of the three doctrines of the beginninglessness (anāditva), authorlessness (apauruseyatva) and self-sufficient validity (svatahprāmánya) of the Veda. In combination they constitute a peculiar set of doctrines, even in the Indian context in which they arose. There is nothing in the contemporary schools of thought, whether Brahminical, Buddhist, or Jaina, corresponding to this set as worked out in Mimāmsā. The preceding Vedic tradition itself contains nothing of the kind, either. Indeed, the Vedic Brahmins held - still in the days of Megasthenes. - the opposite opinion that the world (and therefore presumably the Veda) does have a beginning in time. The schools of philosophy that arose beside Mimämsä believed in the beginninglessness of the universe, to be sure, but they all accepted, unlike Mimāmsā, the periodic destruction and recreation of the world. Why then did Mimämsä invent and accept this strange set of doctrines? What could the Mimāmsakas possibly gain by doing
Parpola is of the opinion that Kätyāyana the author of the Katyayana Srauta Sutra is later than Jaimini (Parpola 1994: 303). He further states (p. 305): "Kätyāyana's work proves that there was a close connection between the Yajurveda and the Sämaveda (i.e., the Veda to which Jaimini belonged, JB) around the time when the Mimamsasutra) came into being." Parpola (1981: 172) is nevertheless of the opinion that "mimamsă discussion involving two opposing protagonists were a regular institution of each Vedic school in the Sutra period.. And it is from these discussions that the Mimamsasutra has directly grown". Schwanbeck's fragment 41; tr. McCrindle 1877: 101. The Mahābhārata characterises the Veda (besides many other things and beings) as being sanātana 'etemal(?)'; e.g. Mhbh 1.1.52.

Page #3

The Origin of Mimämsä as a School of Thought
85
so? Predictably, none of our sources proposes any answers, for these doctrines are not presented as new inventions but as eternal truths. But we are entitled to ask what benefit these strange doctrines brought with them. What could be the advantage for the Brahmins concerned in accepting them??
These three doctrines, most specifically the first of them, have a consequence of which the Mimamsakas themselves were very much aware: since the Veda has no beginning in time, none of the events recorded in it can ever have taken place. An event must have taken place before it came to be recorded; in the case of the Veda this is impossible, for the Veda does not post-date any event. This consequence is most convenient in the case of Vedic stories and remarks that are totally implausible to begin with, but covers quite generally all Vedic statements about what presumably happened in the past. This is clear from Sabara(-svāmin)'s observations in his Mimāmsābhāşya, some of which we will now consider
Sabara is aware that Vedic myths are occasionally in contradiction with reality as we know it. He even provides examples. "The trees sat down for a sacrificial session", "The snakes sat down for a sacrificial session" and "The old bull sings mad (songs)”, all these statements are in contradiction with our experience. They are, Sabara explains, not to be taken literally. They are there in order to praise the sacrificial activities that are enjoined. Similar reasoning applies to all stories in the
Cf. Frauwallner 1968: 107: "eine philosophische Lehre (gewinnt für uns erst Leben und Bedeutung ..., wenn wir verstehen, warum sie geschaffen wurde, welche Probleme sie lösen sollte und warum gerade diese Lösung gewählt wurde..." Cf. Sabara on MiS 1.1.28 and 31: jananamaranavantaś ca vedarthah fruyante "babarah praváhanir akamayata", "kusuruvinda auddalakir akamayata" ity evamadayah | uddala. kasyāpatyam gamyata auddalakih | yady evam präg auddalakijanmano nayam grantho bhutapūrvah levam apy anityata || ... yac ca praváhanir itiran na pravahanasya puruşasyasiddharván na praváhanasyāpatyam praváhanih praśabdah prakarse siddho vahatiś ca prāpane na tv asya samudayah kvacit siddhahikaras tu yathaivāpatye siddhas tatha kriyāyam api kartari tasmad yah praváhayati sa prāváhanih babara iti śabdānukrtih tena yo nityarthas tam evaitau sabdau vadisyatan. [Objection:) Objects are recorded in the Veda that are subject to birth and death. For example: "Babara Prāvāhani (= son of Pravāhana) desired", "Kusuruvinda Auddalaki (= son of Uddalaka) desired" Auddälaki is understood to be the son of Uddalaka. In that case, this book (i.e., the Veda) (can) not have existed prior to the birth of Auddālaki. In this way, too, (the Veda must be non-eternal. ... [Reply:) What [has been said) with regard to Prāvāhani is not correct) Prāvāhani is not the son of Pravāhana, because no such man (called) Pravahana is known to have existed). The linguistic element pra is known as signifying 'excellence, and the verbal root) vah as signifying 'conveying'. But its combination is not known to signify anything. The sound i sin praváhani), on the other hand, is known to signify 'son of as well as the agent of an activity. For that reason praváhani means that which carries in an excellent manner'. Babara imitates the sound (of wind (?)). Therefore these two words (babara and práváhani) will refer to something eternal. The two quotations occur at Taits 7.1.10.2 and 7.2.2.1 respectively. Sabara on Mis 1.1.32: vanaspatayah sattram ásata, sarpah satiram ásata; jaradgavo gayati maliakáni. None of these three citations seems traceable in the Veda as we know it.

Page #4

86
JOHANNES BRONKHORST
Veda, to all Vedic myths; all the passages that contain them are either arthavada or mantra, neither of which is to be taken literally.
These and similar remarks deny the validity of all Vedic myths. None are to be taken literally, all of them have only one function, viz. to encourage, or discourage, people to carry out certain actions. But not only myths are discarded. Sabara goes further, and reduces the deities, presumably the recipients of the sacrifices that must be carried out, to mere names that possess no power and have no anthropomorphic features. His Bhāşya on Mimāmsāsūtra 9.1.9, for example, argues in detail against the notion that deities have bodies and eat. On Mimāmsāsūtra 10.4.23, having first rejected the proposal that deities are the beings living in heaven that are described in traditional stories of the type itihäsa and purāņa, he goes as far as to agree that deities may be nothing but words: "This (position, according to which deities are nothing but words) will not be refuted by us, for this position), when expressed, is not in conflict with our view."10
It will be clear that Sabara discards here, in one fell swoop, all contents of the Veda. The only exceptions are the injunctions, because these cannot be in conflict with other sources of information (Bronkhorst 1997: 367-368; cf. Devasthali 1959: 15). But what could be the point of discarding the contents of the literary corpus which the Brahmins, including the Mimāmsakas, make such a major effort to preserve?
Two possible answers come to mind. The first is as follows. The religious convictions of the Vedic Brahmins are likely to have changed profoundly since Vedic times, so much so that the contents of the Veda no longer agreed with the beliefs they actually held. Mimämsä philosophy offered an elegant way out: the Brahmins could henceforth reject the conceptual side of Vedic religion while remaining guardians of the Veda and continuing Vedic ritual, thus illustrating the observation that ritual traditions can be far more persistent than belief systems (Staal 1985). Unfortunately there is little textual evidence to support this position. It is no doubt significant and in any case highly suggestive that the Mimāmsaka Kumārila Bhatta (7th cent. C.E.) begins his Slokavārttika with a dedicatory stanza to Siva.!! It may be no less significant that his commentator Pärthasarathi Miśra makes an attempt to explain this away. 12
10
Sabara on Mis 10.4.23: nany evam sabda eva devatá prápnoti atrocyate naitad asmábhih parihartavyam | na hidam ucyamanam asmatpaksam badhate. Slokavärttika, Pratijnadhikarana 1 : wisuddhajñanadehåya trivedidivyacaksuse sreyahpräprinimittaya namah somårdhadharine. There are further indications suggesting that Kumärila may have been concerned to integrate "Hinduistic elements, such as his acceptance of the idea of liberation (see Mesquita 1994, there is no reason to think that earlier Mimamsakas had accepted this idea, cf. Bronkhorst 2000: 100). See further below. Cf. Biardeau 1964: 145: "Est-ce ... que la Mimämsä épuise la croyance religieuse des brahmanes qui l'enseignent ou qu'elle l'ait jamais épuisée? Pour l'époque contemporaine, il est
12

Page #5

The Origin of Mimamsá as a School of Thought
87
There is another possible reason why the Mimāmsakas explicitly rejected the contents of the very texts whose guardians they were. To appreciate it one should recall that early in the first millennium C.E. (or even earlier) a tradition of rational debate had established itself in India which came to exert a determining influence on the development of speculative thought. It is not at all clear why and how, and even when exactly, this tradition made its appearance, but once it had appeared, Indian philosophy was never to be the same again; it might even be argued that this tradition allowed classical Indian philosophy to come into existence. Thinkers, it appears, were henceforth obliged to defend their positions against the attacks of outsiders who felt no sympathy for them, and victory in the debates that took place was apparently considered so important that participants modified their positions where necessary so as to make them more coherent and therefore more defensible. The challenges resulting from these confrontations are responsible for much of what might be called the history of Indian philosophy: positions were polished and improved, new ideas introduced, arguments analysed and sharpened.
This development did not affect all those who held views and opinions. The mathematical sciences were not affected until late (Bronkhorst, forthcoming). In philosophy itself it appears that Jainism joined the debate rather late, and Kashmir Saivism only did so until almost a millennium after its initiation. Others may have avoided these debates. Many sacrificing Brahmins may have belonged to this category. They adhered to their traditions, which they did not need to defend, at least not in debates, and continued as much as possible as before. They had no need for verbal confrontations with outsiders, nor indeed for the systematizations of Mimāmsa.
However, sacrificing Brahmins, too, needed royal support, which may occasionally have been contingent upon their skill in defending their positions in confrontations with others, at the royal court or elsewhere. Circumstances of this kind may account for the fact that a number of sacrificing Brahmins joined the tradition of critical debate. This involved exposing themselves to often severe criticism from unsympathetic outsiders. The outsiders concerned were first of all, no doubt, Buddhists, very active participants in the debates of that early period; Buddhists may indeed have played a major role in establishing the tradition of critical debate (cf. Bronkhorst 1999). What would those Buddhists criticize above all in conservative Brahmins who spent their lives reciting the Veda and carrying out complicated rites? Primarily, one would think, the contents of the Veda. The Vedic Brahmins, whether they liked it or not, could in this way be held accountable for myths that were often highly improbable and which they themselves may have long since ceased to take seriously. And yet, the Brahmins would not be able to reject
certain que non: les rares Mimamsaka d'aujourd'hui se disent généralement smärta et se rattachent donc aux disciples de Sankara."

Page #6

JOHANNES BRONKHORST
these myths without damaging their own credibility. Once again, the Mimämsä philosophy offered a way out. The Brahmins who adopted this philosophy did not believe these myths, to be sure. The reason was not however that they were lax, or ignorant about their own tradition, nor that their attachment to the Veda was a mere facade; quite the opposite, they did not believe these myths because they knew, better than their critics, how to interpret the Veda. These myths were not meant to be believed, and those who thought otherwise displayed their own ignorance in doing so.
Seen in this way, Mimāmsā as a system of thought owed its origin, at least in part, to the need to defend the Vedic tradition against outsiders. The doctrine of the beginninglessness of the Veda, along with its corrollary of authorlessness, have as a consequence that all but the timeless" parts of the Veda no longer have to be interpreted literally. The third fundamental principle of classical Mimāmsā, the Veda's self-sufficient validity (svatahpramanya) along with "proximity" as interpretative principle (Bronkhorst 1997) was a doctrinal extension guiding the practice of interpretation. If, then, we recall that the Veda's beginninglessness (anāditva), authorlessness (apauruşeyatva) and self-sufficient validity (svatahpramanya) constitute the three pillars of classical Mimamsā as a system of thought, it can be seen that this whole theoretical construction may find its raison d'être in the need to preserve the Vedic way of life - i.e. the sacrificial tradition - without being bound by most of the contents of this body of literature.
What reason is there to think that the traditional Brahmins may have been criticized for the myths they presumably believed in? Most of the surviving philosophical discussions of classical India concern philosophical problems, and rarely do we come across attacks on the personal beliefs of the participants. This, however, may be due to the fact that most of the surviving philosophical literature of India dates from a time when the participants in the debates had developed a public image far removed from popular beliefs. Yet there are clear traces of evidence to show that the Buddhists, at any rate, had been critical of Brahmanical myths from an early date onward. We will briefly review the Buddhist criticism of one particularly important Brahmanical myth, a myth invoked by the Brahmins to justify their division of society into different castes, varnas, an idea which the Buddhists did not share.13
The myth concerned finds its classic, and probably earliest, exposition in the Purusasūkta of the Rgveda (RV 10.90), but important parts of it recur in many later
13
Some further texts critical of Brahmanical and Hindu mythology, from the side of Jainas and Buddhists respectively, are discussed in Osier 2000 and Masset 2000.

Page #7

texts. It recounts how the world and its inhabitants came about as a result of a sacrifice in which the primordial giant, Puruşa, is dismembered. The most important parts for us read, in the (slightly adjusted) translation of Wendy Doniger O'Flaherty (1983: 30-31):
The Origin of Mimämsă as a School of Thought
14
15
The Man has a thousand heads, a thousand eyes, a thousand feet. He pervaded the earth on all sides and extended beyond it as far as ten fingers. (1)
It is the Man who is all this, whatever has been and whatever is to be. He is the ruler of immortality, when he grows beyond everything through food. (2)...
16
When the gods spread the sacrifice with the Man as the offering, spring was the clarified butter, summer the fuel, autumn the oblation. (6) ....
The hymn to Puruşa is, in the words of Louis Renou (1965: 8), "the, major source of cosmogonic thought in ancient India"; elsewhere he says:
When they divided the Man, into how many parts did they apportion him? What do they call his mouth, his two arms and thighs and feet? (11)
His mouth became the Brahmin; his arms were made into the Warrior, his thighs the Common man, and from his feet the Servant was born. (12)
89
Jan Gonda (1968: 101) calls it "the foundation stone of Vişnuite philosophy".14 Especially the part concerning the creation of the four main divisions of society, the four varnas, has been taken over in numerous texts belonging both to the Vedic and to the classical period. We find it, for example, in the Taittiriya Samhita (7.1.1.4-6), the Rāmāyaṇa (3.13.29-30), but also in the first chapter of the Manu Smrti. The Lord, we read there, created, "so that the worlds and people would prosper and increase, from his mouth the Brahmin, from his arms the Ksatriya, from his thighs the Vaisya, and from his feet the Śūdra."15 Elsewhere the same text refers to this myth as common background knowledge, and as an alternative way of speaking about the four varnas.16
Il n'y a guère de poème cosmologique de l'Atharvaveda où l'on ne retrouve quelque allusion voilée au mythe du Géant sacrifié et au schéma évolutif qui en résulte ... C'est encore le thème du Géant qui sous les traits de Prajapati 'le seigneur des Créatures' ressurgit dans les Brāhmaṇa et en commande la plupart des avenues. (Renou 1956: 12.)
It is open to question to what extent the Purusasukta is representative of Rgvedic religion; Staal (1995: 30) calls it "an atypical, late and isolated composition".
Manu 1.31: lokänäm tu vivṛddhyartham mukhabähürupadataḥ | brāhmaṇam ksatriyam vaisyam sudram ca niravartayat. The translation follows, with modifications, Doniger & Smith 1991. The Bhavisya Purana has the same verse (Lásló 1971: 117).
Manu 10.45: mukhahāhurupajjänäm ya loke jätayo bahiḥ | mlecchavacas caryavācaḥ sarve te dasyavaḥ smṛtäh. Tr. Doniger & Smith 1991: 241: "All of those castes who are excluded from the world of those who were born from the mouth, arms, thighs, and feet (of the primordial Man) are traditionally regarded as aliens, whether they speak barbarian languages or Aryan languages."

Page #8

90
-
JOHANNES BRONKHORST
These and many other references!? to the myth of the Purusasūkta do not allow us to decide with certainty whether the authors concerned took this myth literally. Modern authorities have a tendency to suppose that they did not. Ninian Smart, to mention but one example, has the following to say about myths in general and the way they are understood in the present and in the past (Smart 1996: 138):
[1t) seems ... that we are moving out of the age of what may be called "fanciful" myth into that of "factual" myth. I do not mean by this that the more fanciful myths have not been believed in some sense to be factual: describing reality. But now there is a more earthbound understanding of what is factual. So Adam and Eve have to be real persons: or if they are not they have to be symbolic representations of a real human condition that can be described metaphysically or existentially.
And again (Smart 1996: 161):
As we move towards another century and into it, the divergence, considered phenomenologically, between the old myth and the new history tends to fade away. Legends of Moses and Krishna and the Buddha and Confucius tend to solidify. Since historicity is regarded as a plus, there is a trend towards thinking of the legendary as historically real. In any case, it becomes a problem to distinguish between the two.
These passages suggest that, at least according to Smart, there was a time when myths were not understood to be true in an earthbound factual manner, not historically real. Unfortunately he does not elaborate or clarify this suggestion, and nor does he give any specification as to the date or period during which the important change referred to in these passages has taken place. Moreover, no attempt is made to explain why such a change should take place. What is it exactly that pushes "us" to change our understanding of myths? Are we here presented with a new variant of the now-to-be-discarded distinction between mythical, i.e. pre-logical, and logical thought? If so, some clarifications would have been useful.
Whatever modern authorities may have to say about the question, there is evidence that Indian thinkers, or at least some of them, did take the myth of the creation of the four varnas out of the initial giant quite seriously, i.e. literally -- as being literally true. Part of the story is retold in the Padārthadharmasamgraha, also known as Praśastapädabhāsya, which is the classical surviving treatise of the Vaišeşika philosophy, written by Praśasta, alias Prasastapāda. The passage concerned reads:
When in this way the four composite elements have come into existence, a great egg (mahad andam) is formed, caused solely by God's (maheśvara) meditation/volition (abhidhyāna), out of atoms of fire with an admixture of atoms of earth (i.e., gold). In it (God) creates Brahmā, with four faces like so many lotuses, the grandfather of all worlds (sarvalokapitámaham brahmanam), and all worlds; he then enjoins him with the duty of creating living things. That Brahma, thus enjoined by God, and endowed
17
For a discussion of the importance of the Purusasükta in later literature and practice, see Shende 1965; Gonda 1977: 98-105 (390–397).

Page #9

The Origin of Mimämsä as a School of Thought
18
with abundant knowledge, complete absence of passion and absolute power, knows the effects of the deeds of living beings; he creates the Prajapatis, his mind-created (mānasa) sons, with knowledge, experience and span of life in accordance with their [past] deeds; [he also creates] the Manus, Devas, Rsis and groups of Pitrs (pitrgana), the four varnas out of his mouth, arms, thighs and feet (mukhabahurupadataḥ) [respectively], and the other living beings, high and low (uccavacāni bhütäni); he then connects them with dharma, knowledge, absence of passion and power in accordance with their residue of past deeds. 18
In order to correctly evaluate this passage, it is important to realize that the Padarthadharmasamgraha is no book of stories and myths, nor is it meant to be read as literature. On the contrary, it is a very serious treatise about the constitution of reality, of which it presents a coherent and systematic explanation. It is out of the question to read any passage of this serious work, including the one just cited, as not intending to convey reality, and convey it, not in any metaphorical, but in a most literal manner. It is true that the contents of this passage may not have been part of the Vaiseṣika philosophy during the time preceding Prasasta. There are reasons to believe that the very notion of a creator God may have been introduced into the system by this author, and that he borrowed this notion from the religious current to which he may have belonged, that of the Pasupatas. This does not, however, mean that this notion is to be taken less seriously than the remainder of the Padarthadharmasamgraha.19
19
91
The explicit mention of the creation of the four varņas out of the mouth, arms, thighs and feet of the creator in a work as serious and reality-oriented as Prasasta's Padarthadharmasamgraha shows that at least one participant in the tradition of critical reflection accepted this myth as literally true. It seems likely that many other Brahmanical intellectuals of that period did the same.
As stated above, the Buddhists rejected the fourfold division of human beings, and also rejected the myth that was meant to lend credence to it. A number of Buddhist authors criticize the very same myth which Prasasta (and probably many others with him) explicitly accepted, the myth that the four varṇas were originally
WI, p. 11: evam samutpanneşu caturşu mahabhüteşu maheśvarasyabhidhyānamātrāt taijasebhyo 'nubhyaḥ parthivaparamāņusahitebhyo (variants: parthivadiparamāṇusahitebhyo. parthiväṇusahitebhyo) mahad anḍam arabhyate (some editions read utpadyate) | tasmims caturvadanakamalam sarvalokapitämaham (variant: caturvadanakamalasakalalokapitā maham) brahmanam sakalabhuvanasahitam utpädya prajäsarge viniyunkte (variant: niyunkte) | sa ca maheśvarena viniyukto (variant: niyukto) brahma 'tiśayajñānavairagyaiśvaryasampannaḥ präninām (variant: sarvapraninām) karmavipäkam viditva karmānurupajñānabhogayuṣaḥ sutan prajapatin mänasan manudevarṣipitṛganan (variant: manun deva°) mukhabähürupadataś caturo varṇan anyani cocca vacani bhütäni (variants: bhutani ca; anyani coccävacani ca sṛṣṭvā) sṛṣṭvā, āśayanurüpair dharmajñānavairágyaiśvaryaiḥ samyojayatiti.
On the philosophical reasons underlying the introduction of the notion of a creator God into Vaiseşika, see Bronkhorst 2000: § 7, esp. pp. 37-38; further Bronkhorst 1996.

Page #10

JOHANNES BRONKHORST
created out of the mouth, arms, thighs and feet of the original being. They do so by showing that it is incoherent, in that it has implications which even the Brahmins would be loath to accept.20
We already find such criticism in the Aggañña Sutta of the Digha Nikāya. The Brahmin Vāsesha here reports the position of his fellow-Brahmins, according to whom "only the Brahmins are the real sons of Brahmā, born from his mouth, born from Brahmă, produced by Brahmä, heirs of Brahmā".21 The Buddha responds that they maintain this position, "forgetting what is old" (porāņam assaranta). This expression has been variously interpreted by the commentators: some speak of an old tradition, 22 others of ancient history.23 The context, however, favours a third interpretation: these Brahmins forget the past, that is to say the relatively recent past of their own birth. This is shown by what follows.24 According to the Buddha it is undeniable that the wives of Brahmins (brahmanānam brahmaniyo) have their periods, become pregnant, give birth and feed; in spite of being thus born from a human womb, the Brahmins maintain that they are born from Brahmä.25 In doing so, these Brahmins insult (abbhācikkhanti) Brahmä.26 This criticism is obviously based on the most literal interpretation of the Brahmanical myth. The claim of the Brahmins to have been born from Brahma is in conflict with their birth from a human mother. In other words, the Brahmins are credited with the belief that they were bom, at the beginning of their present life, from the mouth of Brahmä.
The Vajrasūci proceeds in a similar manner. Here the following argument is found:
There is another defect [in your proposition). If the Brahmin is born from the mouth, where is the Brahmin woman born from? Certainly from the mouth. Alas! Then she is your sister! So, you do not regard the convention of licit and illicit sexual intercourse! But that is extremely repugnant to the people of this world. 27
20
Vincent Eltschinger's recent book (2000) has been particularly helpful in writing the • following paragraphs. See further Renou 1960: 43.
DN 111.81: brahmanà va Brahmuno puttă orasa mukhato jätá Brahma-ja Brahma-nimmitá Brahma-dayada. Cf. Meisig 1988: 80-81 for the Chinese parallels. Walshe 1987: 408 ("ancient tradition"); Rhys Davids & Rhys Davids 1921: 78 ("ancient lore"), Sv III. p. 862: poranan ii poránakam aggannam lok'uppartint cariya-vamsam; Franke 1913: 275 (es ist nicht uralte Erinnerung an eine wirkliche Tatsache'). The following remarks also occur in the Assalāyana Sutta (MN 11. 148). DN III81-82: dissanti kho pana Vasertha brahmanānam brahmaniyo utuniyo pi gabbhiniyo pi vijayamaná pi payaman, pi, te ca brahmana yonija va samanā evam ahamsu: brahmană va ... Brahmuno puttå orasă mukhato játå Brahma ja Brahma-nimmitá Brahmadâyáda. Cf. Meisig 1988: 86–87. This last remark does not occur in the Assalāyana Sutla. Vajrasuci, ed. Weber (1860), p. 225, 11. 6-8; ed. Mukhopadhyaya (1960), p. 9 (JJ): anyac ca disanam bhavariyadi mukhato jato brahmano brāhmanyah kuta utpartih mukhad eveti
27

Page #11

The Origin of Mimāmsă as a School of Thought
The Särdülakamāvadāna states essentially the same point:
If this world has been created by Brahmä himself, the Brahmin woman is the sister of the Brahmin, the Ksatriya woman the sister of the Ksatriya, the Vaisya woman (the sister of the Vaiśya, or the Šūdra woman [the sister) of the Sūdra; if she has been created by Brahma, [a woman of the same caste), being a sister (of her husband), she will not be a suitable wife.28
This is not the place to investigate how the Vaišeşikas answered, or might have answered, the criticism of the Buddhists. It must here be sufficient to note that the three classical commentaries on Praśasta's Padārthadharmasamgraha - the Vyomavati, the Nyāyakandali, and the Kiraņāvali - devote long discussions in this connection to the question of the existence of a creator God, but fail to say a word about how this particular myth is to be interpreted so as to avoid contradictions. The discussion stays on a highly abstract, philosophical", level, where inferences and logical analyses have their place. The details of the myth, on the other hand, do not receive attention.
Perhaps the authors of the Vyomavati, the Nyāyakandali, and the Kiranavali were right in ignoring the tricky challenge posed by the Buddhists. Their task would certainly have been difficult. The position of the Mimämsakas, on the other hand, was simple and straightforward. They, the guardians of the Veda, made no effort whatsoever to justify the historical contents of this corpus, because they denied its accuracy. Not only the Purusasūkta, but any historical event seemingly described in the Veda was to be interpreted differently, so as to lose all the historical content it might have seemed to possess. The criticisms uttered by the Buddhists constituted no threat to the Mimāmsakas.
It will be clear that the above-mentioned three doctrines of the beginninglessness (anāditva), authorlessness (apauruşeyatva) and self-sufficient validity (svatahpramanya) of the Veda constituted a wonderful protection for Brahmins confronted with outsiders intent on making fun of the Veda. The introduction and elaboration of these elements - it would appear - turned a school of Vedic interpretation into a school of thought based on a coherent vision of the unique position of the Veda in the world. Do we know when these changes took place?
28
cel hanta tarhi bhavatam bhaginiprasangah syat tathả gamydgamyam na sambhavyate | fac ca loke 'ryantaviruddham. Tr. Mukhopadhyaya 1960: 20. Divy(V) no. 33 verses 76–77, p. 332: yadi tavad ayam loko brahmana janitah svayam brahmani brāhmaṇasvaså ksatriya ksatriyasvasa || atha vaisyasya vaiśyā vai sūdra śüdrasya vă punah na bharya bhagini yukta brahmana janita yadi.

Page #12

JOHANNES BRONKHORST
Some sūtras of the present Mimāmsāsūtra support the idea of apauruşe yatva and its consequences, at least in the interpretation of Sabara. Francis X. Clooney (1990: 51) agrees, and points out that "apauruşeyatva finds its roots, through (sic; this must no doubt be though) not explicit mention, in Jaimini's text". Without saying as much, he probably thinks here of sūtras 1.1.27-32, which he translates as follows (p. 166-167):
1.1.27 vedamś caike samnikarşam purusākhyah | Some people say that the Vedas are similarly (i.e. like sentences in the ordinary world, JB) composed (samnikarsa) because they are named after persons 1.1.28 aniryadarśanac ca Also, because we find ephemeral things (mentioned in the Veda). 1.1.29 ukram tu sabdapūrvatvam But we have already explained that the word is prior (to usage: sabdapúrvalvam). 1.1.30 akhya pravacanát The names (connected with various texts) are due to expounding (and not due to composing) the texts. 1.1.31 param tu śrutisāmänyamatram In regard to the latter argument (28), there is merely a similarity of sounds (fruti samanyamátram). 1.1.32 krte va viniyogah syat karmanah sambandhät (In contrast with the words of ordinary language, Vedic words) apply to what has been accomplished; for words are thus related to action.
Sabara and Clooney may be right in their interpretation of these sutras. If so, we must conclude that two of the above-mentioned three elements - anăditva and apauruşeyatva, along with their consequences - were not introduced by Sabara, but well before him. We cannot however conclude with certainty that the notion of apauruşeyatva, along with the consequences which the Mimāmsā draws from it, already existed at the time of, and found expression in, the hypothetical original Mimāmsāsūtra, the source of the more recent Pūrva- and Uttara-mimāmsāsūtras. As already observed by Parpola (1981: 151-152) and others before him, it seems certain that the Mimāmsāsūtra as we have it contains interpolated passages.
At this point we must try to refine our understanding of the idea of a Veda without beginning and its consequences. The idea that the Veda is eternal in itself appears to be old, and may have also been current in other circles than only those of the early Mimāmsakas. Early (and datable) evidence occurs in the Mahābhāsya of Patañjali, which may convey a reliable impression of the way in which at least some Brahmins thought about this issue in the 2nd century preceding the common era:
Has it not been stated that Vedic texts are not made, that Vedic texts are eternal? (True, but) even though their meaning is eternal, the sequence of their sounds is not eemal.

Page #13

The Origin of Mimämsä as a School of Thought
Here the idea of an eternal Veda is present, but interpreted in a way which renders it relatively harmless.
It is on account of that difference that we have [different recensions of the Veda, such as] the Kathaka, the Käläpaka, the Maudaka, the Paippalādaka. 29
There is another way in which the idea of an eternal Veda can be deprived of its most disturbing aspects, and it appears that many orthodox thinkers - with the exception of the Mimamsakas, of course - resorted to it. A beginningless Veda was conceived of as existing in and alongside a world which passes through cycles of creation and destruction without beginning or end.30 The eternal Veda was believed to be reintroduced after each renewed creation, exactly in the same shape as before. The advantage of this model would be that the Veda, although without beginning, might yet contain information about the world, for the simple reason that the world infinitely repeats itself from beginningless time. We find this position, for example, in the first chapter of the Manusmrti where it describes how Brahma milked the triple eternal Veda out of fire, wind and the sun.31 It seems that this is the position taken in the Uttaramīmāmsāsūtra and later Vedanta. Uttaramīmāmsāsūtra 1.3.29 and 30 (as interpreted by Sankara) maintain that the Veda is eternal. Sūtra 1.1.2 informs us that the world is periodically recreated. Sūtra 1.3.28 (as interpreted by Sankara) adds that the world arises out of the Vedic word. 32
30
A particularly clear description of this Vedanta position occurs in the much later Vedantaparibhāṣā. It reads as follows:
31
32
95
The Mimamsakas who occupy themselves with the sacrifice (i.e. the Pūrvamimämsakas) maintain that the Vedas are valid because they are eternal and therefore free from all human faults. In our opinion (i.e., that of the Vedantins), on the other hand, the Veda is not eternal, because it has an origin.
[Objection:] The fact that the Vedas have an origin and have been made by God proves that they have an author; such being the case, your position according to which the Vedas have no author is shown to be incorrect.
29 Maha-bh II, p. 315, II. 13-15 (on P. 4.3.101 vt. 3): nanu coktam na hi echandāmsi kriyante nityäni cchandāmsiti | yady apy artho nityo ya tv asau varṇānupurvi sanityä | tadbhedāc caitad bhavati kathakam käläpakam maudakam paippaladakam iti.
Note that Kumārila (Tan Vär on sūtra 1.3.7, p. 122-123) is not averse to the idea of world periods.
Manu 1.23a-c: agnivayuravibhyas tu trayam brahma sanatanam | dudoha... Kane (196877, II: 352) claims that "[a]ll dharmasastra writers proceed on this axiom of the eternity of the Veda", without however giving references in support of this.
[Reply:] Not so, for "having an author" does not, to begin with, mean "being uttered by a person". Nor does it mean "having an origin that depends on a person".
To explain: at the beginning of creation God made the Veda in such a way that its composition is identical to the composition of the Veda established during the pre
Šankara explains the words ataḥ prabhavat of sutra 1.3.28 with the words: ata eva hi vaidikāc chabdad devadikam jagat prabhavati.

Page #14

96
JOHANNES BRONKHORST
It appears, then, that the idea of a beginningless Veda (and perhaps even that of an authorless Veda) may not have been an invention of early Mīmāmsā. However, only the Mīmāmsakas (and this does not include the Vedantins) drew from it the far-reaching conclusions which turned their school into an impenetrable bastion for those defenders of the Veda who did not wish to identify with its myths. Who did so, and when, remains obscure. Why they did so may have become clearer after the preceding reflections.
vious creation, not a different Veda. The Vedas have, as a result, no author in the sense that they are not the object of an utterance that is independent of a similar utterance (made during an earlier creation). The utterance of the Mahabharata etc., on the other hand, is independent of a similar utterance (during an earlier creation), and therefore these texts do have an author. In this way tradition has been defined as being divided into parts that have and those that do not have an author.33
With regard to the introduction of the third element, the self-sufficient validity of the Veda (svataḥprāmāṇya), we are on firmer ground. Erich Frauwallner (1968: 107ff.) has adduced convincing reasons to show that this doctrine was created by the so-called Vṛttikära, the anonymous author a long passage of whose work is cited in Sabara's Bhasya on sūtras 1.1.4-5. Not only does the Vrttikara explain this doctrine in the passage concerned,34 but there are various indications to show that he introduced this doctrine as a novelty. In view of what has been said earlier in this article, it is significant to note that this Vṛttikara is very much concerned, and involved in a debate, with Buddhist positions.
33
34
Summarizing the reflections presented so far, it seems likely that Mīmāmsā - that is to say Pūrvamīmāmsā, i.e., that which finds expression in the Purvamimämsäsūtra and its commentaries - underwent an important modification, and became more than before a "school of thought", through the introduction and elaboration of three doctrinal elements: the claimed beginninglessness (anaditva), authorlessness (apau
Text and translation as in Bronkhorst 1998: 12-13: vedänam nityatvena nirastasamastapumdüṣaṇataya pramanyam ity adhvaramimämsakäḥ | asmākam tu mate vedo na nityaḥ utpattimattvät... | nanu.... utpattimattvena parameśvarakartṛkataya pauruşeyatvasiddhau apauruşeyatvam vedänām iti tavapi siddhanto bhajyeta | iti cet na | na hi tavat puruşena uccāryamāṇatvam pauruşeyatvam... näpi puruṣādhinotpattikatvam [pauruşeyatvam] | kimtu sajatiyoccaraṇanapekṣoccaraṇaviṣayatvam pauruşeyetvam | tatha ca sargadyakale parameśvaraḥ purvasargasiddhavedänupürvisamānānupurvikam vedam viracitavan | na tu tadvijātīyam vedam | iti na sajâtiyoccaraṇanapekṣoccaraṇaviṣayatvam pauruşeyatvam [vedänăm] | [mahabharatadinām tu sajatiyoccaraṇam anapeksyaivoccaranamı iti teşam pauruşeyatvam | evam pauruşeyāpauruşeyabhedena agamo dvividho nirupitahl. I translate pauruşeya with 'having an author'.
For text and translation, see Frau wallner 1968: 24ff.

Page #15

The Origin of Mimämsä as a School of Thought
ruşeyatva) and self-sufficient validity (svataḥprāmāṇya) of the Veda. It seems likely that this modification took place in two steps, presumably connected with two persons: the author of Mīmāmsāsūtra 1.1.27-32, and the Vṛttikara cited by Sabara respectively. Together these modifications provided Mīmāmsā with a global, overarching and coherent vision. This vision is unique in the sense that it is radically different from anything else produced by Indian philosophers, 35 and even from the Vedic thought which this school is supposed to represent and continue. The reasons for the creation of such an extraordinary system of thought - even by contemporary Indian standards - must be sought in the particular circumstances and challenges that accompanied its beginnings. We know little about the beginning of Mīmāmsā as a system of thought but for the fact that it must have occurred when a tradition of rational debate and criticism had established itself in India, a tradition which came to determine the shape and development of the main schools of philosophy. All schools that participated in this tradition had to make sure that their systems were coherent and defensible in debates with unfriendly critics. Mīmāmsā in its new garb was coherent and eminently defensible. Even its Achilles heel - the obligation to defend the Veda and therefore its contents, including the many improbable stories it contains - had been properly taken care of: Mimämsä after its transformation no longer had to defend anything found in the Veda except for its injunctions, for it had effectively discarded everything else.
35
97
Having discussed the origin of Mimämsä as a school of thought, I add a few provisional remarks, not about its end, but about the end of the circumstances that gave rise to it. I have suggested that the presence of unfriendly critics, along with the wish or obligation to listen to their criticisms, were responsible for the systematisations resulting in "Mīmāmsā as a school of thought". Among these critics the Buddhists played a particularly important role. Buddhism, however, was in serious decline in the 7th century of the common era. Chinese pilgrims inform us that Buddhist monasteries were largely deserted, a development which went hand in hand with an increase in the number of Hindu temples ("Deva-temples"). 36 In other words, the most redoubtable critics of Brahmanical orthodoxy were losing their position in society, and their criticism - whatever the logical value of their argu
36
The Samkhya philosopher called Madhava must here be mentioned, who, for theoretical reasons, appears to have rejected the idea of world periods followed by renewed creation; cf. Bronkhorst 2000: 61.
Eltschinger 1999, which is in this respect based on Joshi 1967, Chapter XII; the Chinese pilgrims are primarily Hsüan-tsang and I-ching, among others.

Page #16

98
JOHANNES BRONKHORST
ments - no longer constituted the threat it once had. What would be the effect on a school like Mimāmsā?
Our attention is inevitably drawn to Kumārila Bhatta, influential Mimāmsā author of the 7th century. In another study (Bronkhorst 2000, $ 13) I have pointed out that Sabara appears to have made an effort to conceptualise the mechanism of karmic retribution by reducing all the relevant elements of the sacrifice (the sacrifice itself, its result: heaven, the gods) to mental entities. Kumärila, on the other hand, did not do so, leaving karmic retribution essentially unexplained. Is it possible that Sabara, under the perceived pressure of Buddhist critics, felt obliged to offer explanations where Kumarila, no longer under threat, could do without?
Another feature deserves attention. Early "philosophical" Mimāmsā was primarily concerned with the validity of the Veda. This does not mean that it was uninterested in non-Vedic texts, texts composed by human authors. A few sutras deal with the validity of the Smrti, and Sabara's discussion shows that injunctions - presumably occurring in Kalpa Sūtras and the like - are at stake.37 Such injunctions are valid if they concern invisible things and are not in contradiction with the Veda; it must indeed be inferred that they are based on Vedic texts that may have been lost. Other injunctions in the Smộti are valid because they serve a useful purpose.38 Kumārila extends the list of valid texts so as to include the Vedāngas, in particular, all of which are, at least in part, based on Vedic texts.39 Even the sciences of reasoning (tarkaśāstra) are bom from worldly experience, arthavadas and Upanişads (lokārthavādopanisatprasūta), whatever that may precisely mean. More important are his remarks elsewhere to the effect that the epics and Purānas (?; Kumārila says bhāratādi 'the Bhārata etc.' and mentions the authors 'Vālmīki, Dvaipāyana, etc.). though of human origin, are to be interpreted like the Veda, i.e. in Mimāmsā fashion.40 We find indeed that Dharmaśāstra commentators - among them Kumārila's contemporary Bharucio! - start to use Mimāmsā methods in interpreting their Smrti
37
Agrawal (1985: 25) traces Sabara's quotation astakaḥ kartavyäh to Asvalāyana Grhyasūtra 2.4.1: gurur anugantavyan to Vasisthasmrti 8.9: tadagam khanita vyam to Manusmrti 8.264, prapa pravartayitavyå to Vasisthasmrti 2.38; śikhäkarma karta yam to Väräha Grhyasūtra 4.24; audumbaryāḥ sarvavestanam to Lātyāyana Srautasūtra 2.6.2; astācanvārimsad varsani vedabrahmacaryacaranam to Gautama Dharmasūtra 1.2.51-53; kritarājako (']hhojyánnah to Bharadvāja Srautasūtra 10.9.3.4. See however Garge 1952: 245-246; 248-249. Sabara on sutras 1.3.1-4. Cf. Kane 1968–77, III: 827-828 and V: 1260-1261. Tan Vär on sūtra 1.3.2, pp. 79-80. Cf. Ayyar 1952: 43-44; Jha 1903: 119-120 Tan Vär on sūtra 1.2.7. p. 14 I. 20 - p. 151. 13. Cf. Eltschinger 1999: Ayyar 1952: 40–41; Jhã 194) 3: 25-26. Derrell (1975. 1: 14) proposes "between A.D. 600 and 650" as "conservative" dates for this author. Derelt (1973: 15) mentions Bharuci's Vivarana on the Manusmrti. Visvarüpa's Balakrida on the Yājñavalkyasmīti and Maskarin's bhāsya on the Gautama Dharmasūtra as constituting the carliest group of commentaries in Dharmaśāstra, all of which must have been composed before the end of the 7th century.

Page #17

The Origin of Mimämsä as a School of Thought
texts.42 Treating Smrti texts like the Veda implies, among other things, accepting their prescriptions without needing to justify them,43 or to worry about the intentions of their authors.44 Bhāruci's way of interpreting the Manusmrti illustrates this. Not only does he account for every statement in the Manusmrti as being vidhi, niyama, parisamkhyā or arthavāda (Derrett 1975, I: 25), as would a Mimāmsaka when dealing with a Vedic text, but also no reasons are given to justify the contents of those statements. What is more, passages where Manu himself gives reasons embarrass the commentator. Rather than taking them as reasons, Bhāruci sees them as arthavadas, "whereupon they cease to embarrass" (Derrett 1975, I: 27).45 An example is Manu 11.12(13): "He may take three or two things at his pleasure from the dwelling of a Sūdra (for the success of the sacrifice), for the Šūdra has no business with sacrifices. 46 The second half of this verse would seem to give a reason for the first half, but Bharuci explains it otherwise: it is an arthavada. Still on the same verse, Bharuci points out that Manu elsewhere forbids asking property from a Südra for a sacrifice, and obviously anticipates surprise that one can take what one cannot ask for. His response: "There is nothing which is too heavy for a text, for our śāstra is concerned to teach us."47 Derrett explains in a note: "It seems unreasonable that a Sudra's property should be forbidden if it is asked for, but suitable if purloined. But if that is what the text requires, we must accept it."
42
43
It seems that the importance of Mimämsä in earlier Dharmaśāstra is Lingat 1973: 148 (similarly Keith 1921: 97) writes: "Vasistha (III. 20), Baudhāyana (I.1.1.8). and Manu (XII111) call a mimamsaka to sit in the parişads which are given the role of resolving controversial questions. It seems that very early the Mimāmsā was regarded as an indispensable science for the interpreter." None of these passages uses the term mimamsaka. Manu 12.111, for example, has the word tarki which some later commentators - but not Bhāruci and Medhätithi, the earliest ones - associate with Mimämsä. The fact that the Yajnavalkya Smrti (1.3) ranks the Mimamsa amongst the bases (sthana) of the knowledge of dharma, along with Nyaya and the Vedāngas, does not at all need to imply that Mimämsä is to be used in interpreting Dharmaśāstra texts (such as the Yājñavalkya Smrti itself). Cl. Lingat 1973: 107: "In the time of the commentators) the human origin of the dharma. Śástras) had ... been completely obliterated. It was an article of faith that the precepts which they contained derived from Sages of the remotest antiquity, and their authority was accord ingly beyond dispute. They appeared as if they were scripture, timeless, etemal; the whole of them, along with the epics and the puranas, brought to men the voice of a tradition which was both holy and in conformity with the order of nature. The commentators and authors of juridical treatises could not imagine their role as anything other than that of interpreters, concerned only to explain the meaning of texts whose authenticity and religious importance they did not doubt for one moment." On Medhātithi's ideas about the role of Manu, see Wezler 1998. For the way reasons are dealt with, see further Lingat 1973: 154-155. Derrett 1975, I: 234: aharet trini và dve vá kämam Sudrasya veśmanah | na hi sudrasya yajneşu kaścid asti parigrahah. Tr. Derrett 1975, II: 345-346. Derrett 1975, I: 234: na vacanasyåribhåro 'sty upadeśaparatvác chástrasya. Tr. Derrett 1975, II: 346.

Page #18

100
JOHANNES BRONKHORST
If then, as was argued above, "philosophical" Mimāmsā developed its views and methods in order to defend its "way of life" against unfriendly critics, these same views and methods came to play an altogether different role by the time the unfriendliest of critics, the Buddhists, were losing influence. They became a way of (and an excuse for) explaining all traditional texts without ever needing to look for justifications. In this way the whole of traditional literature was excluded from critical debate, and the question whether this or that aspect of it could stand up to criticism lost its importance. Mimāmsā thus came to contribute, not so much to the preservation of Vedic sacrificial activities, as to the myth that all norms emanated from a superhuman source" and to the assumption that innovation was decay, and that change must be, not merely for the worse, but an infringement of the natural order of things" (Derrett 1973: 27).48
REFERENCES
Texts and Abbreviations
Divy(V) = Divyåvadāna - Edited by P. L. Vaidya. (Buddhist Sanskrit Texts, Darbhanga, 20.)
Darbhanga 1959. DN = Dighanikaya - Edited by T. W. Rhys Davids & J. E. Carpenter. 3 vols. London: Pali Text
Society, 1890-1911. Maha-bh = Patañjali, (Vyakarana-Mahabhasya - Edited by F. Kielhorn. Bombay 1880-85. Manu = Manusmrti - Edited by J. L. Shastri. Delhi: Motilal Banarsidass 1983. Mhbh = Mahabharata - V. S. Sukthankar et al. (eds.), Poona Critical Edition. 19 vols. Poona:
Bhandarkar Oriental Research Institute, 1933-66. Mis = Mimämsäsúrra - Anandāśrama Sanskrit Series, 97. Poona 1973-84. MN = Majjhima-Nikaya - Edited by V. Trenckner & R. Chalmers. 3 vols. London: Pali Text
Society, 1888-99. Sv - Buddhaghosa, Sumarigalavilasini, Dighanikaya-atthakatha - Edited by T. W. Rhys Davids,
J. E. Carpenter & W. Stede. 3 vols. London: Pali Text Society, 1886-1932. TaiS = Taittiriyasamhità - Anandāśrama Sanskrit Series, 42. Poona 1900-08. Tan Vär = Tantravårtrika of Kumärilabhata - Anandāśrama Sanskrit Series, 97. Poona 1973-84. WI = Johannes Bronkhorst & Yves Ramscier, Word Index to the Prasastapädabhâsya. A Com
plete Word Index to the Printed Editions of the Prasastapādabhäsya. Delhi: Motilal Banarsidass, 1994.
Literature
AGRAWAL, D. J. 1985. Mimamsa Uddharana Koša. The Citations from Sabara-bhasya Traced
to their Original Sources. Pune: Vaidika Samsodhana Mandala. AYYAR, A. S. Nataraja 1952. Mimamisa Jurisprudence. The Sources of Hindu Law. (Ganganatha
Jha Research Institute Series, 2.) Allahabad: Ganganatha Jha Research Institute.
48
I thank Kiyotaka Yoshimizu for useful criticism.

Page #19

The Origin of Mimämsä as a School of Thought
101
BIARDEAU, Madeleine 1964. Théorie de la connaissance et philosophie de la parole dans le brahmanisme classique. (Le Monde d'Outre-Mer Passé et Présent, première série, études. 23.) Paris: Mouton.
BRONKHORST, Johannes 1985. The origin of an Indian dietary rule: evidence for a lost Mänava work on Dharma. Aligarh Journal of Oriental Studies 2(1-2) (Ram Suresh Tripathi Commemoration Volume): 123-132.
1989. Studies on Bhartṛhari, 2: Bhartṛhari and Mimämsä. Studien zur Indologie und Iranistik 15: 101-117. (Reprinted in: R. C. Dwivedi (ed.), Studies in Mimämsä. Dr. Mandan Mishra Felicitation Volume: 371-388. Delhi: Motilal Banarsidass, 1994.
1996. God's arrival in the Vaiśeşika system. Journal of Indian Philosophy 24: 281-294. 1997. Philosophy and Vedic exegesis in the Mimāmsā. In: Eli Franco & Karin Preisendanz (eds.), Beyond Orientalism: The Work of Wilhelm Halbfass and its Impact on Indian and Cross-Cultural Studies (Poznań Studies in the Philosophy of the Sciences and the Humanities, 59): 359-371. Amsterdam: Rodopi.
1998. Does the Veda have an author? Asiatische Studien / Études Asiatiques 52(1): 5-14. 1999. Why is there Philosophy in India? (Sixth Gonda lecture, held on 13 November 1998 on the premises of the Royal Netherlands Academy of Arts and Sciences.) Amsterdam: Royal Netherlands Academy of Arts and Sciences.
2000. Karma and Teleology. A Problem and its Solutions in Indian Philosophy. (Studia Philologica, Monograph Series.) Tokyo: International Institute for Buddhist Studies. (forthcoming). Panini and Euclid: reflections on Indian geometry. Ingalls Commemoration Volume.
CLOONEY, Francis X. 1990. Thinking Ritually. Rediscovering the Purva Mimämsä of Jaimini. (Publications of the De Nobili Research Library, 17.) Vienna: Gerold & Co. - Leiden: E.J. Brill Delhi: Motilal Banarsidass..
DERRETT, J. Duncan M. 1973. History of Indian Law (Dharmasastra). (Handbuch der Orienta
listik, 2.3.1.) Leiden: E.J. Brill.
1975. Bharuci's Commentary on the Manusmrti. The Manu-Sastra-Vivarana, Books 612. Text, translation and notes, I-II. (Schriftenreihe des Südasien-Instituts der Universität Heidelberg, 18.) Wiesbaden: Franz Steiner.
DESHPANDE, Madhav M. 1999. Vedas and their Sākhās: contested relationships. Paper read at the Second International Vedic Workshop, Kyoto University, October-November 1999. DEVASTHALI, G. V. 1959. Mimämsä, the Vakya-Sastra of Ancient India. Bombay: Booksellers' Publishing Company.
DONIGER, Wendy & Brian K. SMITH (tr.) 1991. The Laws of Manu. (Penguin Classics.) New York: Penguin Books.
ELTSCHINGER, Vincent 1999. Dharmakirti contre Kumarila et l'apauruşeyata. Lecture given 12.3.
1999 in Lausanne.
2000. "Caste" et philosophie bouddhique. Continuité de quelques arguments bouddhiques contre le traitement réaliste des dénominations sociales. (Wiener Studien zur Tibetologie und Buddhismuskunde, 47.) Vienna: Arbeitskreis für Tibetische und Buddhistische Studien Universität Wien.
FRANKE, R. Otto 1913. Digha Nikaya. Das Buch der langen Texte des buddhistischen Kanons. (Quellen der Religionsgeschichte 8:4.) Göttingen: Vandenhoeck & Ruprecht.
FRAUWALLNER, Erich 1968. Materialien zur ältesten Erkenntnislehre der Karmamīmāmsā. (Sitzungsberichte der Akademie der Wissenschaften in Wien, Phil.-hist. Kl., 259. Bd. 2. Abhandlung.) Wien: Hermann Böhlaus.
GARGE, Damodar Vishnu 1952. Citations in Sabara-Bhāṣya. A Study. (Deccan College Disserta
tion Series, 8.) Poona: Deccan College.

Page #20

102
JOHANNES BRONKHORST
GONDA, Jan 1968. The Mudgalopanisad. Wiener Zeitschrift für die Kunde Sudasiens 12-13.
(Beiträge zur Geistesgeschichte Indiens, Festschrift für Erich Frauwallner): 101-113. 1977. Vedic cosmogony and Vişnuite bhakti. Indologica Taurinensia 5: 85-111.
(Reprinted in: Selected Studies, VI.1: 377-403. Leiden: Brill, 1991.) JHA, Ganganatha 1903. Kumärila Bhatta, Tantravārtika. A Commentary on Sahara's Bhasya on
the Purvamimåmså Sätras of Jaimini. Translated into English, I. (Reprint: Delhi: Sri
Satguru, 1983. JOSHI, Lalmani 1967. Studies in the Buddhistic Culture of India (during the 7th and 8th
centuries A.D.). Delhi : Motilal Banarsidass. KANE, Pandurang Vaman 1968-77. History of Dharmśástra, I-V. (Government Oriental Series,
B 6.) Second ed. Poona: Bhandarkar Oriental Research Institute. KEITH, Arthur Berriedale 1921. The Karma-Mimamsā. (Reprint: Delhi: Kanti Publication, 1989.) LÁSLÓ, Franz 1971. Die Parallelversion der Manusmrti im Bhavisyapurāna. (Abhandlungen für
die Kunde des Morgenlandes, 40.2.) Wiesbaden: Franz Steiner. LINGAT, Robert 1973. The Classical Law of India. Translated from the French (Les sources du
droit dans le système traditionnel de l'Inde, The Hague 1967) with additions by J.
Duncan M. Derrett. (Reprint: New Delhi: Munshiram Manoharlal, 1993.) MASSET, Danièle 2000. Une lecture bouddhiste des mythes hindous: la mise en question des
modèles brahmaniques dans le Canon bouddhique tibétain. In: Marie-Luce BarazerBilloret & Jean Fezas (eds.), La norme et son application dans le monde indien: 155
173. Paris: Ecole française d'Extreme-Orient. MCCRINDLE, J. W. 1877. Ancient India as Described by Megasthenes and Arrian. Calcutta.
(Reprinted in: Ramchandra Jain (ed.), McCrindle's Ancient India. New Delhi: Today &
Tomorrow's Printers & Publishers, 1972.) MEISIG, Konrad 1988. Das Sutra von den vier Ständen. Das Agganna-Sutta im Licht seiner
chinesischen Parallelen. (Freiburger Beiträge zur Indologie, 20.) Wiesbaden: Otto
Harrassowitz. MESQUITA, Roque 1994. Die Idee der Erlösung bei Kumarilabhatsa. Wiener Zeitschrift für die
Kunde Südasiens 38: 451-484. MUKHOPADHYAYA, Sujitkumar 1960. The Vajrasuci of Asvaghosa. A Study of the Sanskrit Text
and Chinese Version. Revised 2nd edition. Santiniketan: Visvabharati. O'FLAHERTY, Wendy Doniger 1983. The Rig Veda. An anthology. One hundred and eight
hymns, selected, translated and annotated. (Penguin Classics.) Harmondsworth: Penguin
Books. OSIER, Jean-Pierre 2000. Une critique satirique des normes brahmaniques chez les jaina:
Dhūrtākhyāna de Haribhadra et Dharmapariksă de Harişeņa. In: Marie-Luce BarazerBilloret & Jean Fezas (eds.), La norme et son application dans le monde indien: 139
153. Paris: École française d'Extreme-Orient. PARPOLA, Asko 1981. On the formation of the Mimamsă and the problems concerning Jaimini,
with particular reference to the teacher quotations and the Vedic schools. Wiener Zeitschrift für die Kunde Südasiens 25: 145-177. 1994. On the formation of the Mimams and the problems concerning Jaimini, with par ticular reference to the teacher quotations and the Vedic schools (Part I). Wiener
Zeitschrift für die Kunde Südasiens 38: 293-308, RENOU, Louis 1956. Hymnes spéculatifs du Véda. (Connaissance de l'Orient: Collection UNESCO
d'auvres représentatives, Série indienne, 3.) Paris: Gallimard. 1960. Erudes védiques et påninéennes, VI: Le destin du Véda dans l'Inde. (Publications de l'Institut de Civilisation Indienne, 10.) Paris: E. de Boccard. 1965. The Destininy of the Veda in India. Delhi: Motilal Banarsidass. (English translation of Renou 1960.)

Page #21
The Origin of Mimamsa as a School of Thought 103 RHYS DAVIDS, T. W. & C. A. F. RHYS DAVIDS 1921. Dialogues of the Buddha, III. (Reprint: London: Pali Text Society, 1977.) SHENDE, N. J. 1965. The Purusa-sukta in the Vedic Literature. (Publications of the Centre of Advanced Studies in Sanskrit, A 4.) Poona: University of Poona. SMART, Ninian 1996. Dimensions of the Sacred. An Anatomy of the World's Beliefs. London: HarperCollins. STAAL, Frits 1985. Substitutions de paradigmes et religions d'Asie. Cahiers d'Extreme-Asie 1: 21-57. 1995. Mantras between Fire and Water. Reflections on a Balinese Rite. (Koninklijke Nederlandse Akademie van Wetenschappen, Verhandelingen Afdeling Letterkunde, Nieuwe Reeks, 166.) Amsterdam: North-Holland. WALSHE, Maurice 1987. The Long Discourses of the Buddha. A Translation of the Digha Nikaya. Boston: Wisdom. (Reprint: 1995.) WEBER, A. 1860. Uber die Vajrasuci (Demantnadel) des Acvaghosha. Abhandlungen der konig lichen Akademie der Wissenschaften zu Berlin, phil.-hist. KI., 3, 1859: 205-264. WEZLER, Albrecht 1998. Medhatithi on the role of Manu, the prayojana of the Manusmrti, and the incentive of the Brahmins to study it. In: Paul Harrison & Gregory Schopen (eds.). Suryacandraya: Essays in Honour of Akira Yuyama on the Occasion of his 65th Birthday (Indica et Tibetica, 35): 217-240. Swisttal: Indica et Tibetica Verlag.