Maṇimēkhalai in its Historical Setting


[xvii] In the following lectures I have attempted to consider first of all the question what the position of Maṇimēkhalai is among the Tamil classics generally and how far the general judgement of the Tamil literary public that it is one among the five great classics is justifiable on grounds of literary merit and general classical excellence. As such, it was necessary to consider whether it could be regarded a Śangam work, and if so, in what particular sense of the term, whether as a work which was presented to the Śangam and which received the Śangam imprimatur, or whether it should be taken to be merely a literary work of classic excellence, as often-times the expression is used in that sense in later Tamil literature. The investigation and enquiry into Tamil literary tradition leads to the conclusion that it is a work of classic excellence in Tamil literature and may be regarded as a Śangam work in that sense. We have no information that it was ever presented to the Śangam, although, according to Tamil tradition, the author was one of the Śangam 49, and, being so close to the age of the Śangam itself, it may be spoken of appropriately as a Śangam work, though not presented to the Śangam.

This position receives additional support in the contents of the two works, which constitute a twin Epic, namely, Śilappadhikāram-Maṇimēkhalai. The subject-matter of the two is one continuous story, and describes what befell a householder and his wife of the city of Puhār, and, as a consequence, the renunciation [xviii] of the daughter of the hero of her life as the first courtesan of the Cōḻa capital. The author of the one is described to us as the brother of the contemporary Cēra ruler, Śenguṭṭuvaṉ, a Śangam celebrity, and the author of the other is similarly introduced to us as a personal and admiring friend of the Cēra sovereign and his ascetic younger brother. Other details of a contemporary character introduced in the story, all of them, are referable to incidents which find mention in relation to various rulers of the Tamil land in the Śangam classics. Thus the mere external circumstances and the few details that we possess of the life and life-time of the authors, as well as the Tamil tradition that the author of the Maṇimēkhalai himself was one of the Śangam 49, all alike seem to tend to the conclusion that the work was a product of the age which may be generally described as the age of the Śangam, that is, the age of Śenguṭṭuvaṉ Cēra as the dominant ruler of South India.

Tamil early adopted a system of grammar, and so far as literary productions in the language go, follow the prevalent system of grammar and rhetoric. As such these works do not lend themselves exactly to that kind of investigation of a linguistic and philological character which could be more appropriately adopted in regard to works where the language is more flexible and has not attained to the classic fixity of an accepted system of grammar. But it still lends itself to a certain amount of investigation as a work of literature, and such an investigation clearly reveals the intimate connection between the Śilappadhikāram and the Maṇimēkhalai itself as literary works, products of a single age, a single tradition, and of a very similar atmosphere. If comparisons are made of these with genuine Śangam classics themselves, the similarity is no less pronounced, apart [xix] from the similarity of historical matter and of geographical surroundings. Thus from the point of view of literary criticism, we have good reason for regarding these as classics of Tamil, which may be treated as of the same literary character as Śangam works.

The historical and geographical details which can be gathered round a character like Śenguṭṭuvaṉ Cēra, and just a few others who happen to figure in these romantic poems, when carefully collected and collaborated, tell the same tale of contemporaneity between the works themselves and between the two works and other Śangam works so-called. Specific instances of historical incidents are dealt with in full detail in the lectures themselves. We need hardly do more here than merely to point out that the four capitals of Puhār, Madura, Vañji, and Kāñcī occur in the poem. Their condition and the rulers that held sway over them are described incidentally in the course of the story, and these admit of definite treatment in comparison with the condition of these capitals, as we find them described in the Śangam works. One point which clinches the matter and provides a definite test of the age is that throughout the story as narrated in these two works, Kāñcī remained a viceroyalty under the authority of the Cōḻas, who, under Karikāla, are credited uniformly by Tamil tradition with having civilized this land and brought it into the pale of Tamil civilization.

Without going into too much detail here, it may be said that the country round Kāñcī which became peculiarly the territory of the Pallavas, remained under Cōḻa rule, and a Cōḻa, a prince of the blood very often, held the viceroyalty. The one remarkable change for which we have evidence in the Śangam works is the placing of this viceroyalty in the hands of a Toṇḍamāṉ chief by name Iḷaṁ-Tiraiyaṉ. [xx] This took place in the last period of the age of the Śangam from the evidence of the Śangam literature itself. In the classics with which we are concerned, there is no evidence of our having reached the stage when Kāñcī was under the rule of Toṇḍamāṉ-Ilaṁ-Tiraiyaṉ; nor have we any vestige of evidence that would justify the assumption that the Toṇḍamāṉ chief had ruled and passed away. Other historical details can be recited in number. It is hardly necessary to take up those details here, which are discussed elsewhere in the course of this work and in other works of ours. The conclusion to which we are, therefore, irresistibly driven is that we are in an age when the Śangam activity had not yet ceased, and this view is in full accord with all the evidence available regarding the Śangam and its age in the vast mass of literature in which that evidence lies scattered.

Our main purpose in this thesis has been to consider what light the philosophical systems and the religious condition of the country as described in the Maṇimēkhalai throw upon this important question of the age of the work itself and of the Śangam literature generally. It is with a view to this that the examination was actually suggested by Professor Jacobi and was taken up by ourselves. The chapters bearing upon the questions are three, namely, books XXVII, XXIX and XXX. Book XXVII discusses the heretical systems from the point of view of orthodox Buddhism.

Maṇimēkhalai discusses, with orthodox professors of the various schools, the tenets of their particular systems on the basis of their authoritative works with a view to learn what exactly they might have to teach. She begins with a discussion of the Pramāṇas applicable generally as instruments of knowledge, and, under the general grouping [xxi] Vaidikavāda, five separate systems are described, all acknowledging the authority of the Veda. The first statement of importance contained in this particular part has relation to pramāṇas as applied to the Vaidika system. Three authorities are mentioned, Vēdavyāsa, Jaimini, and Ktakoṭi. Of these, the first is said to have formulated ten pramāṇas of which the second rejected four and accepted only six. The third one, however, seems to have accepted eight and rejected only two of the ten. After a detailed discussion of the pramāṇas and what they are, the discussion winds up with the conclusion that the pramāṇas current at the time are six, and they applied alike to the six systems commonly recognized as such.

The six pramāṇas as given are, Pratyakṣa, Anumāna, Sabda, Upamāna, Arthāpatti, and Abhāva. The six systems held as orthodox are Lōkāyatam, Bauddham, Sāṁkhyam, Naiyāyikam, Vaiśēṣikam, Mīmāṁsam, with the respective authors, Bhaspati, Jina, Kapila, Akṣapāda, Kāṇāda, and Jaimini. In this recital of six, the omission of Jainism is interesting, but may be understood as being due to its not following the Vaidika pramāṇas. While Nyāya and Vaiśēṣika are both of them mentioned, Yoga is not mentioned along with Sāṁkhya. Bauddham is mentioned as a religion to which these pramāṇas were applicable, and that is in accordance with the opinion that the Bauddhas from Buddha onwards to Vasubandhu who adopted the system of Akṣapāda, and perhaps other teachers of the Pramāṇa Vāda likewise. Mīmāṁsa is mentioned as one Śāstra ascribed to Jaimini, not as two as in later times and in orthodox parlance it had come to be recognized. This leaves out the Brahma Kaṇḍa of Vyāsa.

Apart from this general system of pramāṇas, others from whom she attempted to learn their tenets were Śaivavādī, Brahmavādī, [xxii] Vaiṣṇavavādī and Vēdavādī. At the end of the book, in summing up the totality of the systems she attempted to learn she includes these first five as one. The two following next, the teaching of the Ājīvaka and that of the Nirgrantha she apparently counts as one. The chapter winds up with the statement that thus she had learnt from their respective teachers the five systems, and Maṇimēkhalai herself repeats the statement later when she mentions it to Aṟavaṇa Aḍigaḷ that she had learnt the five systems according to their authoritative texts from those who had specially studied them, in book XXIX.

This kind of reference to the five heretical systems makes it clear that at that time there was a fixed notion that six were regarded as the prevalent systems to which, from the point of view of the Buddhists, five were heretical. Including the orthodox Buddhist system, it made up six, and therefore, we are justified in regarding the six systems as those referred to as current by Maṇimēkhalai herself in an earlier place. The points of importance for our investigation in this chapter are that the six accepted Vaidika pramāṇas applied even to Buddhism, that Buddhism regarded itself as within the fold of Vaidika pramāṇas, even though Jainism was not. The Mīmāṁsa is regarded as a single system as yet, while the Yoga system had not been known, at any rate had not become a recognized system in this part of the country. The later recognition that two pramāṇas were alone valid by the Buddhists, namely, Pratyakṣa and Anumāna, had not yet been adopted exclusively as a cardinal doctrine of Buddhism.

Passing on to chapter XXIX, we are here introduced to a system of Buddhist logic where the teaching of the Buddhists assumes more definite shape and is in the course of the full definition to which it attained under Dignāgācārya. [xxiii] Here the author puts the Buddhist teaching of logic in the mouth of Aṟavaṇa Aḍigaḷ, introduced as a Buddhist saint of the highest reputation in the Tamil country at the time. The chapter begins with saying that the highest authority for the system is Jinēndra, the Buddha, and that the pramāṇas are actually only two, Pratyakṣa and Anumāna. After defining these two, a general statement is put in that all the other pramāṇas are capable of inclusion in Anumāna. Then we are led to the five Avayavas or organs of syllogism, Pratijña (proposition), (2) Hētu (reason), (3) Dṣṭānta or Udāharaṇa (example), (4) Upanaya (application) and (5) Nigamana (conclusion). After having defined and illustrated the first three, the last two are passed over as being capable of inclusion in Dṣṭānta. Then follows a further discussion that these three could be valid and invalid, and the sub-division of each one of them is given with illustrations. Thus we are taken through a regular course of logic, the purpose of the cultivation of which is stated to be, at the end of the chapter, that by means of the validity of reasoning and its invalidity, one may understand that which is truth from that which is other than truth.

In the details that are given of the whole discussion and in the general trend of the discussion itself, the Maṇimēkhalai seems to us to follow the prevalent teaching of logic current in Kāñcī at the time. From Kāñcī however, there hailed a logician of great reputation, known by the name of Dignāga, who wrote, according to Chinese authority, a number of treatises on the subject, and thereby had become the authoritative teacher of the system. Some of his works have been continuously in use as text-books in China, and from a somewhat later period, in Tibet. They were apparently in use in India as well, but had long since [xxiv] gone out of use.

Within the last twenty years some of these have been recovered by various scholars, Chinese, Tibetan, Indian and European of various nationalities. The best known works of this Dignāga are Pramāṇasamuccaya and Nyāyapravēśa. These are sometimes criticized in commentaries by Brahmanical commentators as well as Jain, and, needless to say, quoted with approval and elaborated by Buddhist commentators. The two works quoted above constitute the final authoritative texts of this author on the subject, of which the Nyāyapravēśa seems from the information available to us at present, the fuller. For our purpose the similarity between this work of Dignāga and chapter XXIX of the Maṇimēkhalai runs through all details, and even the examples happen to be the same. This is nothing surprising as, in the treatment of technical subjects like this, examples are chosen for their peculiar aptness and all teachers accept them generally for purposes of illustration. Having regard to the great reputation that Dignāga has achieved as a logician, it may seem a natural inference that a poet like the author of the Maṇimēkhalai should have borrowed the teaching from a treatise like the Nyāyapravēśa.

Notwithstanding the closeness of similarity, there are a few points in which the Maṇimēkhalai treatment of the subject seems to mark a transition from, it may be, the Naiyāyikas to the teaching of Dignāga himself, particularly so in the two points to which attention had been drawn, namely, in the statement that the pramāṇas are only two, others being capable of inclusion in the second, Anumāna; the reference is obviously made to the other four pramāṇas out of the six already referred to as current at the time and applicable to the six systems in book XXVII. We have no right to interpret the other pramāṇas there as any other than the four of the six, to [xxv] which the work made explicit reference in book XXVII, whereas Dignāga seems to have no such qualms, and actually deals with the four pramāṇas of the Naiyāyikas, retains the first two, and rejects the other two, after examination, positively.

Similarly in the discussion of the Avayavas, the Maṇimēkhalai seems to mark a transition. It mentions the five Avayavas accepts the three, and does not consider the other two as they are capable of inclusion in the third. There is nothing like the rejection of these as invalid as in the case of Dignāga. Then there is a third point. Dignāga solemnly lays himself out to consider the Svārtha and Parārtha form of syllogism, that is, syllogistic ratiocination with a view to convincing oneself, and with a view to convincing others. After a serious discussion, he comes to the conclusion that the latter being included in the former, it is superfluous to treat of it separately. To the Maṇimēkhalai, it does not seem necessary to discuss the latter at all. In regard to the Pakṣa-ābhasas discussed, the Nyāyapravēśa is supposed to make a new classification and describes nine which are found described almost in the same terms in the Maṇimēkhalai itself.

Here comes in a discussion which may seem alien to the course of this argument, but which, as will be noticed, has an important and vital bearing on the question itself. Who is the author of the Nyāyapravēśa? The text of the Nyāyapravēśa not having been available, there were two clearly divided schools of thought, one of them regarding the Nyāyapravēśa, both in the Tibetan and Chinese version as well as the now available Sanskrit version, is the work of Dignāga; another school, basing itself chiefly on an examination of the Chinese originals, regards it as the work of Dignāga’s immediate disciple Śaṁkarasvāmin. Without going into the arguments [xxvi] which will be found elsewhere, we may state here that although Buddhist tradition had known of Śaṁkarasvāmin as the disciple of Dignāga, no authority bearing on Buddhist literature has mentioned a work, Nyāyapravēśa ascribing it to Śaṁkarasvāmin as the author. It is now clear that the Nyāyapravēśa was ascribed to Śaṁkarasvāmin very early in China, and that the work had been constantly in use there. There is no mention, however, of Nyāyapravēśa as the work of Śaṁkarasvāmin anywhere in the works of Hiuen T’sang [Xuan Zang] among treatises on logic which were being studied by students at the schools and Universities in India.

I’Tsing [Yi Jing] who gives a complete list does not mention the Nyāyapravēśa as the work of a Śaṁkarasvāmin, but seems to mention it, not perhaps exactly in the same form, as the work of Dignāga himself. So the work may well have to be regarded as the work of Dignāga, which his disciple Śaṁkarasvāmin perhaps taught and his teaching spread into China, and gave him the reputation of being the author of the work. But whether the Nyāyapravēśa is the work of Dignāga himself, as we prefer to take it, or whether it should turn out to be the work of Śaṁkarasvāmin, actually his disciple, it does not materially affect our question, as the difference of time could be hardly a generation. The real question is whether the Nyāyapravēśa is copied in the Maṇimēkhalai or, as we take it, whether the Maṇimēkhalai marks a transition between the Naiyāyikas and the Nyāyapravēśa of Dignāga. In the latter alternative, the date of the work could be much earlier than A.D. 400; and in the former alternative it must be held decisively to be a work of the fifth century at the earliest. We have good reason for regarding Maṇimēkhalai as a work anterior to Dignāga, and we shall see that it is in away supported by what the [xxvii] work has to say actually of Buddhism in the following book.

In book XXX the author of the Maṇimēkhalai lays himself out to give the actual teaching of the Buddha ‘according to the Piṭakas’, and gives a clear but succinct statement of the main Buddhistic theory of the ‘Four Truths’, ‘the twelve Nidānas’, and the means of getting to the correct knowledge, which ultimately would put an end to ‘Being’. There is here none of the features that the later schools of Buddhism indicate, so that we cannot exactly label the Buddhism contained in book XXX as of this school or that precisely.

It may be said, however, to be of the Sthaviravāda and of the Sautrāntika school of Buddhism, which seems to be the form in vogue in this part of the country, and coming in for much criticism later. This position is, to some extent, supported by the expression used in the text itself elsewhere that it is the ‘Path of the Piṭakas of the Great One’. Even in this abridged form, it is not without points that indicate a transition similar to those indicated in book XXIX. There is nothing that may be regarded as referring to any form of Mahāyāna Buddhism, particularly the Śūnyavāda as formulated by Nāgārjuna. One way of interpreting this silence would be that Nāgārjuna’s teaching as such of the Śūnyavāda had not yet travelled to the Tamil country to be mentioned in connection with the orthodox teaching of Buddhism or to be condemned as unorthodox. This is to some extent confirmed by the fact that in referring to the soul, the reference in book XXX seems clearly to be to the individual soul, not to the universal Soul, which seems to be a development of the so called Sātyasiddhi school which came a little later. These points support the view to which we were led in our study of the previous books, [xxviii] and thus make the work clearly one of a date anterior to Dignāga, and not posterior.

This general position to which we have been led by our own study of the philosophical systems, though at variance with the views to which Professor Jacobi has arrived on the same material, cannot by itself be held decisive of the age of the Tamil classic. This question has to be settled actually on other grounds, of which we have indicated the general position in some detail already.

Kāñcī is referred to as under the rule of the Cōḻas yet, and the person actually mentioned as holding rule at the time was the younger brother of the Cōḻa ruler for the time being. Against this vice royalty an invasion was undertaken by the united armies of the Cēras and the Pāṇḍyas which left the Cēra capital Vañji impelled by earth hunger and nothing else, and attacked the viceroyalty. The united armies were defeated by the princely viceroy of the Cōḻas who presented to the elder brother, the monarch, as spoils of war, the umbrellas that he captured on the field of battle.

This specific historical incident which is described with all the precision of a historical statement in the work must decide the question along with the other historical matter, to which we have already adverted. No princely viceroy of the Cōḻa was possible in Kāñcī after A.D. 300, from which period we have a continuous succession of Pallava rulers holding sway in the region. Once the Pallavas had established their position in Kāñcī, their neighbours in the west and the north had become other than the Cēras. From comparatively early times, certainly during the fifth century, the immediate neighbours to the west were the Gangas, and a little farther to the west by north were the Kadambas, over both of whom the Pallavas claimed suzerainty [xxix] readily recognized by the other parties. This position is not reflected in the Maṇimēkhalai or Śilappadhikāram. Whereas that which we find actually and definitely stated is very much more a reflection of what is derivable from purely Śangam literature so-called. This general position together with the specific datum of the contemporaneity of the authors to Śenguṭṭuvaṉ Cēra must have the decisive force. Other grounds leading to a similar conclusion will be found in our other works:– The Augustan Age of Tamil Literature (Ancient India, chapter xiv), The Beginnings of South Indian History and The Contributions of South India to Indian Culture. The age of the Śangam must be anterior to that of the Pallavas, and the age of the Maṇimēkhalai and Śilappadhikāram, if not actually referable as the works of the Śangam as such, certainly is referable to the period in the course of the activity of the Śangam.