THE barbarians who speak Austro-Asiatic tongues of Munda type form the lowest substratum of the population of India. Not only were they the original inhabitants, but vast reservoirs of "uncivilized" mankind still survive in the hill districts, away from the great roads and centres of culture. The Santals of Chota Nagpur and the Kols are at this stage to-day. Students of India were too long blind to the permanent influence of this factor, both racial and linguistic. It needed the curiosity of Sylvain Lévi, directed to place-names, to raise the question in all its bearings, and it needed the competence in Austro-Asiatic philology of his disciple J. Przyluski to produce definite results which are already very remarkable. It is established that the most ancient element in India is of the same race as the peoples of the Nicobar Islands and Malacca.
The religion of these tribes may be described as being based on totemism. The members of a group feel themselves to be one in that their life is the actual life of a vegetable or animal species. So the principle which animates them is at once immanent and transcendent; and here we must recognize the first experience of an attitude from which many later metaphysical systems were to proceed.
Vegetable energy is very widely venerated. Eating offers a way of capturing this force by absorption of the spiritual principle. India was to keep the conviction that existence is a question of food, and of that axiom, based on elementary observation, it would discover many subtle applications. Moreover, to primitive man the spectacle of the vegetable world bears witness to the polymorphism of nature more clearly than that of the animal kingdom. That a tiny seed should produce masses of verdure, and that all that luxuriance should end in new seeds, which men can use or destroy, a fact to be seen on every side in the jungle which was the normal environment of man in India, no doubt planted in men's minds the notion of a mechanical, spontaneous evolution, which could, however, be arrested by human intervention. Even in its most abstract interpretations, this evolution would continue to be described in "vegetable" terms: the manifestations of existence result from seeds which ripen and fructify with a view to subsequent sowings without end, unless there is destruction by torrefaction.
The lower peoples of India were many of them cannibal. For them the best sacrifice of all was the consumption of the raw flesh and hot blood of human beings. This practicc is in accordance with the principle of totemism, which finds the essence of universal life in the life of the species. These bloody rites are quite unlike Dravidian or Vedic sacrifices, and are doubtless the prototype of those exceptional rites, of which classical orthodoxy preserved the memory, in which the victim was a man.
Most of the features which Lévy-Bruhl has noted as characteristic of "primitive" thought are to be found not only among the less civilized peoples of India, but in a great many aspects of Indian thought in the syncretic form assumed by it in historical times. Such, for example, is the prelogical mentality, which never dies out, in spite of attempts almost as ancient as itself to set up a logical discipline. Such is the conviction of the efficacity of desire or thought. Such are the belief in continuance after death, and that in the power of an individual, living, dead, or divine, to be in more than one place at a time.
So it is one or the great differences between the West and India that the latter has always hoarded in her bosom, in the midst of highly refined cultures, elements which have remained, as it were, in their crude state. The "barbarian” invasions inflicted on her from outside were never as barbaric as some permanent factors in India herself.
The Dravidian cults left on the religions of the classical period traces which, if not deeper, were at least more definite than those of the pre-Dravidian cults. One religious act in particular still bears their mark—pūjā, the veneration of an idol. To do worship to an image, to water it or scent it or hang it with garlands, is quite a different act from a blood-sacrifice. Flowers, scents, and the rest may, no doubt, be regarded as offerings, but they are attentions rather than gifts. Stone or wood, a statue is the symbolic object of a cult in that it is "cultivated", rendered flourishing and prosperous, and therefore capable of radiating beneficent influences. The negritos who practise such rites are gentle creatures, very unlike the Kols with their bloody sacrifices.
In this religion grossness and ferocity are to be found in the gods rather than in man. Most of the deities are females, and this accords with the preponderant place held by women in matriarchal societies, from the Asianic peoples of Cappadocia to the dwellers on the seaboard of the Bay of Bengal. The hideous ogresses who still rule in the temples of the south-eastern coast of the Deccan perpetuate this form of divinity. There is no doubt that Kali the Black and Durga the Unapproachable would never have been brought into the Brahman pantheon if Dravidian goddesses with names ending in -amma had not stood as prototypes. It is even to be supposed that Krishna, the very much Brahmanized "black" demigod, keeps up some connexion with the primitive rites of the Dravidians especially when that mystic bridegroom of pious consciences assumes, and that too in the Bhagavadgītā, the aspect of a devouring monster.
So, then, the dark-skinned Indians were particularly alive to the maleficent and fearful attributes of the absolute, the numinosum which Rudolf Otto, following Durkheim, describes as one of the two faces of the sacred. They not only feared the gods, as did some of the contemporaries of Epicurus and Lucretius; they were terrified of them. No doubt the caresses lavished on their images tended to make them inoffensive. When Hinduism coaxes the destroying god by calling him the Propitious or the Benevolent, Siva, it is acting in the Dravidian fashion.
The gluttonous, grinning goddesses rule the fruitfulness of nature. The masculine aspect of that fecundity is represented in an impersonal manner by many phallic emblems. This is the origin of the fetishism of the linga, so widespread in classical India, where, from the beginning, the name borne by universal spirit is the Man, Purusha. Here again the religion of Siva forms the bridge between Dravidian superstitions and orthodoxy, for the swallower assumes the character of the great generator; indeed, the chief contribution of the Dravidians to the abstract thought of Hinduism might well be the idea which had such a great future before it, though foreign to Vedism—-that production and destruction come from the same principle.
We know nothing about the amalgam of Sumerian and Dravidian ideas which must have reigned in the Indus basin before the coming of the Aryans. But through it the religions of India are in part derived from the ancient religions of Mesopotamia. Animism, the belief in creation by the efficiency of the Word, and the worship of a Mother Goddess are very old Asianic and Sumerian convictions, and Iranian Ahura and Vedie Varuna are continuations of Anu, god of the sky, but also of rain and therefore of water, who was worshipped at Sippar and Nippur from the beginning of the third millennium B.C. The similarity of name between Ahura (Asura in Vedic) and Ashur, the eponymous god of Assyria, is striking, and we may add that the apsu of the Babylonians is probably equivalent to apas (waters) in the Vedas. Many Semitic legends, especially Assyro-Babylonian, were brought by the same road into the inheritance of India, such as the story of the Deluge, which must, in its most eastern form, have united with the myth of Manu. Lastly, Chaldæan astrology was also handed down to the Hindus.
In comparison with the religions of the other races of India, the primitive religion of the Aryans is one of which we can have some positive knowledge, for we have abundant evidence for it in its various forms. We must remember that the Árya, the Indo-Iranian branch of all the peoples which speak Indo-European languages, can be studied in their Iranian aspect in Avestic literature and in their Indian aspect in the Vedas, and that the two series of works show very close affinities.
Here the difficulty lies in the interpretation of the documents. The two literatures developed independently, and it is rash to draw conclusions from their likenesses about the original phase from which both have come. The most ancient parts of the Avesta, the Gathas, preach the reformation of Zoroaster, which was not earlier than the second half of the seventh century B.C., and was a reaction against the previous cults. Many expressions survive from those cults, but only appear in the "late" Avesta, which was compiled in the second and third centuries of our era, and is therefore contaminated by elements a thousand years more recent. The Ṛigveda and Atharvaveda are on the whole more ancient than the Gathas of Zoroaster, but they already have an Indian, or at least a Punjabi character. And while we have much information about the use made by the Brahmans of the Vedas, we have very little about the rites practised at the time when the hymns were evolved, about 1,000 or 1,500 years before Christ.
The document which is earliest in inspiration, although it was compiled last of all the Vedas, is the Atharvaveda, a collection of magical formulas. In all probability, the original form of the religious act, karman, was a direct action by which a man, not yet a priest but the possessor of certain forms of words, could by the efficacity of those words bend nature to the achievement of his purposes. This fundamental conviction was to leave a permanent mark, but speculative inquiry would aim at an acquisition, prāpti, not the mere satisfaction of curiosity. What the Aryans of early India sought to obtain in this way was the goods of this world—subsistence, a minimum of well-being, even wealth, a full life, not cut off by premature death, and male descendants, who alone were qualified to continue after the father's death the offerings which supported the the life of their ancestors. Now, protection against evil forces—devils, the hungry dead, plague, human enemies, wild beasts—and the provision of descendants are both to be secured by direct influence exercised on things.
What the formula needs, and all that it needs, to be successful is correctness. The Brahman word satyam, which is often loosely translated "truth”, simply means ritual correctness. This too was to become a permanent element in the thought of later times; truth would be a matter of normality and propriety, not of conformity to an object. To know "in the right way"—that was the way to succeed in all circumstances.
The axiom implied in this magic, that the right formula is infallible, holds good both for what is in order according to nature and for what alters the course of events for the benefit of man, as an individual or as a group. In other words, formulas (mantra) ensure the normal order of natural events, such as the succession of the seasons, the fertility of the land, or the fruitfulness of animals, no less than the twist given to natural laws by a particular will. Here was another factor for all later Indian thought to retain. Nothing is effected save by a law in the sense of a norm, the average, regular order or the accidental, exceptional order. Afterwards the name of dharma was given both to natural existence and to legislation, to the order of objects and to the moral order. The Indo-Iranians had not yet constructed that supreme "category"; they contented themselves with a word to designate the order of constitutive principles and of universal stability, the asha of the Iranians, the rita of the Vedas; but they brought about that very order, like the order created by a particular will, by sacrificial formulas.
Ṛita as the basis and sacrifice (yajña) as the means, and the means to that same basis—such was the foundation of Aryan beliefs. From the idea that the world subsists by sacrifices the greater part of later speculative development sprang. When it was no longer possible to suppose that the mere action of an individual man could ensure the permanence of fundamental natural conditions, it was imagined that ṛita was the result of sacrifices made by certain gods, or by all the gods. It was even conceived that the world is a sacrifice. The Vedas, in the strict sense, stop short of this point. The authors of the hymns, less exclusively and crudely magical than those of the Atharvaveda, thought it advantageous to utter the praises of the gods in order to obtain from their favour the realization of human purposes; instead of themselves producing the desired phenomena, they left it to a "wished god" (ishta devatā) to satisfy their desires. Ritual correctness influences the gods, and influences things through the intervention of the gods.
This was not such a great change as one may suppose, for the gods are natural forces, most of them barely anthropomorphized. Although the theory that the Vedic pantheon had its origin in philological circumstances has been pressed too far, it is certain that a numen implies for a certain group of facts a nomen. A type exists when a word is found to give the data of experience a sign and an expression, which soon become their essence.
Ṛita never becomes a mere arbitrary institution, the work of a god. The gods themselves, entrusted with maintaining it in the universe, are only its keepers (gopa), as a herdsman has charge of his herd. This task falls on Mitra and Varuna. These two brothers, the first sanctioning contracts and ruling the friendships of men and the second governing the revolutions of the heavens and bearing witness to all actions, predominate over the earliest Indian reflection on the cosmic order, the social order, and their relations. The mainly "formal" character of their function makes them the most typical gods of India, which in every age, as we shall see, is inclined to define objectivity by correctness of operation, by canons of right practice. The other gods are merely personifications, often pale and abstract, of natural phenomena or instruments of worship.
Perhaps we should say, not "or”, but "and”. The naturalistic explanation of former times and the ritualistic explanation, more recent in Europe but classical in native Brahmanism, might be interpreted more profoundly as systematized aspects of a single reality. Agni is the sacrificial fire as well as the element fire. Parjanya is the art of rainmaking as well as the rain itself. Ushas is the delicate charm which drives away darkness and restores the light as well as the delicate dawn. Soma is the moon and also the sacrificial drink, for the moon, emptying and refilling without end, shows itself to be a cup of immortality. With regard to this we should note that in the Indian view an object is a force, and being is thought of only in reference to action; it is important to observe this principle in operation at the very beginning of Indian thought.
In a domain where the spirit of system has given proof of its vanity we shall beware of believing in the value of one single explanation. Not all the gods in the Vedic pantheon are of this part natural, part ritual character. Underneath them we find the more ancient pair, common to the whole Indo-European stock of beliefs, of the Sky Father and the Earth Mother. Characteristic of the narrower range of Indo-Iranian beliefs is the honour paid to light in its various aspects. The gods (deva) are properly beings of light, shining or heavenly, and it was by an accident of history that the dev of the Avesta became anti-gods or demons. In sectarian and even Buddhist forms, a cult of the Sun continues all through religious development. In part it is derived from Vishnu in the Vedas; his Three Steps symbolize his conquest of the three worlds. If we look for other sources, we may remember Savitri, the vivifying power of the sun's heat, and Pushan, the stimulant which causes plants and cattle to grow. Mitra himself is an aspect of Surya, the sun.
By the side of the metaphysical and moral god, Varuna, and the ritual god, Agni, the Vedas give a very great place to another deity, Indra. The books make these three figures, in turn or simultaneously, the very prototype of godhead. The storm which fertilizes the earth by pouring water on it, according to the myth of the liberation of the clouds in the form of cows, and blasts the demons which oppose that fertilization, is simply the aspect of Indra as a natural force. He, and he alone, has a very concrete human character; he stands for the Aryan in his victorious war with Dasyu, the dark-skinned first occupant of the country, who is symbolized by the demon Vritra. Force is his essence, and if, like all the gods, he owes his continued existence to ambrosia, he consumes it like a drunkard and glutton. Having won his celestial rank by his exploits, he has the air of a deified hero. Moreover, his connexion with the company of the gods is always uncertain; sometimes he is the son of Dyaus, and sometimes of Tvashtar. From the fact that he is invoked with Varuna as a witness of the treaty struck by the people of Mitanni in northern Mesopotamia (1400 B.C.), we may wonder if he is not more ancient than the arrival of the Aryans in India. Indeed, he is known to the Iranians, but, doubtless under the influence of the moral side of Zoroastrianism, he is placed among the demons, either because he is a deva or on account of certain unedifying passages in his story. None the less, he is the most Indian of all the gods. He is advised by Vishnu, the god of sacrifice, and combines with him in a dyarchy which is an anticipation of that of the Brahman and the Kshatriya.
Gods of secondary or still lower importance are numerous, Mitra and Varuna, coupled under the name of Adityas, are sons of Aditi, the Unlimited, the Indian form of the Great Goddess of Asia Minor. The Asvins, or Horsemen, also called the Nasatyas, are the Indian Dioscuri. Rudra, the Roarer, personifies all that cattle-owners fear—storm and plague. His sons are the Maruts, the hurricanes which descend from the mountains where he dwells. Since the protection of the living depends on him, he is supposed to have remedies for the pests which he lets loose. So Rudra is a first sketch of the destroyer whom men afterwards, to conciliate him, called the Propitious, Siva.
In the composition of the Indian pantheon abstract reflection can be seen at work in two opposite directions. We have seen how the pre-Indian gods Mitra and Varuna are concerned with contracts and the maintenance of order respectively; the latter watches in particular over the oath. Aryaman presides over marriage. In contrast to these old deities, others, already more Brahmanic than Vedic, are mere personified abstractions—Sraddha, Faith; Manyu, Anger; Prajapati, the Lord of Creatures.
The confusion which we see, the different strata of religious life at which we guess, give the present-day reader of the Ṛigveda the impression of many serious discrepancies. But it has taken more than a century of Indian studies for these evidences to become manifest. The master-keys with which it was attempted to force the secret of the text were systems of etymology, language-study, mythology, or ritualism, and from Colebrooke to Bergaigne hardly anyone questioned that the content of the text was a systematic whole. L. Renou's objective examination in Les Maîtres de la philologie védique (Paris, 1928) gives an excellent account of the way in which exegesis, for a long time dogmatic, has advanced towards conclusions which are not sceptical, but relativistic. The mythology of the Ṛigveda, according to Winternitz,1 is in a state of growth, and so is the text itself. Nothing could be more artificial than the arrangement of the most ancient part (books ii to vii), where the hymns are grouped according to the god praised and the number of lines. The later parts, the first and tenth books, border on the literature of the Brāhmaṇas. No doubt when a collection (saṃhitā) was put together by arranging already existing stanzas there were many interpolations. The stanzas themselves were already a mass of inconsistencies, some aiming at flattering a god or toadying to princes, others being used to accompany a sacrifice or a funeral, and yet others giving a dramatic dialogue or a philosophic reflection. Between the expositions of priestly ritualism, like those of the earliest Brāhmanas (of the sixth century B.C., for example) and the not merely Aryan but Indo-European inspiration (the Germans unjustifiably call it Indo-Germanic) of other hymns, we may reasonably place an interval of a thousand years. We must not, therefore, be surprised to find elements which are not Indian because they are pre-Indian side by side with others which are not only Indian but Hindu, that is, which already show a mixture of Aryan and aboriginal ideas.
Then we shall see why those who have specialized in the Veda from the beginning of historical times onwards have failed to understand it. The explanations given by the earliest commentator, Yaska, about 500 B.C., are far from being concordant, and his collection of etymologies, the Nirukta, is hardly less disappointing, valuable as it is. But it gives more information than the paraphrase of Sayana, who wrote in the fourteenth century. What was already lacking in the sixth century B.C. was the prose commentary with which the singers, like the singers of the Avestic Gathas, used to accompany their recitation of the hymns. Moreover, the Brahman pundits looked in these books, which for the greater part they did not understand, for what they wanted to find there, namely hymns which could be used for their own sacrificial methods. In exactly the same way in China, Confucian tradition has been only too successful in transforming the love-songs of the Book of Verses into a treatise on morals and politics. In any case, it is obvious that the priestly scholasticism which already permeated the hymns encouraged the Brahmans in the illusion that their own scholasticism was a continuation of the same work.
Besides, mystery added to the prestige of the ancient poetry. Although lists of family traditions indicated, if not the actual authors of the hymns, at least their human origin, a metaphysical interpretation of the Veda became established as the immovable basis of Brahmanism. The hymns, according to the orthodox view, have an absolute reality, like the ideas in Plato. They are truer than empirical reality, they exist for ever and contain the secret not only of all wisdom but of all existence. They constitute a revelation, and a revelation by sound, which can be heard (śruti); to indicate exactly how far they have objectivity, they are said to have been "seen" by the first inspired bards, who, accordingly, are called ṛishis, the "seers" of the hymns.
In the magic of the Atharvaveda and the verses of the Ṛigveda we have the whole of the origins of the Aryan religion of India. All that Brahmanism afterwards claimed to be was the inheritance of the Vedic tradition; certainly it was the exploitation of it, and exploitation by the priestly caste which came into being after the age of the Ṛigveda. It is an anachronism to regard the Veda from the Brahman point of view only, but that anachronism is the corner-stone of orthodoxy from the beginnings of history to our own time. The arrangement of the hymns in saṃhitās represents the first manifestation of Brahmanism. Other manifestations, akin to the first, were the development of the two supplementary Vedas, the Chants (sāman) and the Sacrificial Formulas (yajus). The Chants are almost all stanzas of the Ṛigveda, sung to music. The Sacrificial Formulas are shown to be much later than the Chants by the fact that they are written in prose, with or without a commentary (the White or Black Yajurveda). It was considered necessary to have as many collections of texts as there were priests in the sacrificial practice of the Brāhmaṇas; the rij is recited by the hotar, the sāman chanted by the udgātar, the prayer of the yajus muttered by the adhvaryu.
The Brahrnanic conception of sacrifice, which thus governs the way in which the Vedas are presented, is at once systematic and realistic.
It revels in abstraction. To speak in concrete terms, we should say that the magical operation of sacrifice creates or preserves the world. The authors of the Brāhmaṇas prefer a different language: the universe is a sacrifice. There is a cosmic giant, the Lord of Creatures, Prajapati, or the Man, Purusha, and the parts of the world are his members. Ever afterwards India would call the parts of a whole its members, anga, for everything that is consists of the members of the macrocosm. Prajapati is at once the sacrificer and the thing sacrificed. So this demiurge, so different from the Vedic gods, is equivalent to the personification of sacrifice.
Existences are classified in several strata of reality, which correspond to the senses and the intellect, so far as the microcosm is concerned, to the elements, to the Vedas, to the Vedic metres, to the principal gods, and so on. In each of the lists thus formed, any two principles stand in the same relation to each other as two principles in another list. This kind of intelligibility, based on analogies and orders of dignity, is both an ancient and a permanent feature of Indian thought. Any fact symbolizes a number of other facts, and all facts are at the same time governed by the act of sacrifice.
Brahmanic realism is like that which has been noted among many peoples described as "primitive”. It implies that abstractions, like Prajapati, have an objective existence, reigning over the whole universe or part of it. There are sounds which give birth to things—those of the ritual formula, Brahman, those of the voice, Vāch, those of the breath, Prāṇa, three terms whose meanings partly overlap. There are forms which are creative—gestures and drawings which have magical efficacy. By imposing names and forms on a chaos, Prajapati, like the Demiurge of the Timœos, introduces order and establishes the specific nature of beings. Thought, an operation of manas (the empirical mind and foundation of the five senses), is not a subjective double of the macrocosm, an accidentally individual disposition of representations; it is a reality, not spiritual but vital, capable of going past the limits of the body by the sensorial orifices and wandering outside. Dreams, death, and the efficacity of wishes or purposes prove it.
The cosmic forces are catalogued by a kind of religious physical science, which is at once crude and subtle. They are impersonal like those just mentioned—ritual formula, voice, breath, thought. Men have a share in them, but they go from nature to man as well as from man to nature. In the latter case they are set in motion by the magical operation, but in the former they have a cosmogonic value. The chief of them are the following. Mahas and aṇu are largeness and smallness, growth and restriction. Guṇa is the characteristic quality of a being; in the strict sense, it is kind, variety. Then there are various aspects, consequences, or conditions of life—asu, the vital breath; āyus, both age and the normal length of life, the vital quantity imparted to each being; anna, food as a means of subsistence. Tanū, a sort of humbler duplicate of the concept of ātman, which was to become so very important, is the self of each individual, and even of each thing, in both its material aspect (body) and its intelligible aspect (mind, essence). Satya is the truth of the real, the conformity of an individual to its type (an abstract extension of the primitive meaning of "ritual correctness"). The list includes principles endowed with partial expansivity, which, emanating from a centre, which may be the sun or a human energy, propagate as it were their vibrations through the whole extent of the ākāśa, the universal environment and condition of all movement. Such are tejas, the cutting, the piercing of a ray of light; varchas, luminous energy; bhrājas, brilliance; śrī, splendour, beauty, power and glory; ojas, energy. The neuter gender of many of these nouns shows that the principles which they designate are impersonal. We must add enas and āgas, crime and transgression, evil as a wrongful action, but also as the calamity which results from it; pāpa, misfortune ensuing from sin; puṇya, its converse, advantage ensuing from merit. The idea of activity is implied in these concepts, which are summed up in the notion of karman. This word, properly meaning "rite", covers every activity from sorcery (yātu, māyā) to the moral action, magical power and virtue being the two poles of religious activity. So the idea of karman is very wide, being at once very primitive and very capable of refinements, which in time to come would give expression to a reflection and a moral life, at first less gross and later more and more highly developed. But the notion of activity would keep some of its primitive characteristics, even if they were not properly understood. For example, asceticism (tapas), the superior acting of the man who has risen beyond action, properly means and originally meant "heating", the warmth which makes the demiurge sweat as he fashions beings or that which hatches the cosmic egg in the old myths.
Dharman, karman, brahman, and presently ātman stand in the front rank among these original classes, almost categories, distinguished by Indian thought. They all have neuter names, which will not produce masculine terms till later. They are all anonymous forces, of which existences are only the result. Schopenhauer shows a very keen understanding of the Indian mind when he says that in this domain, contrary to what happens in the classical antiquity of the West, esse sequitur operari. The rite being the origin of all speculation, action always stands at the beginning. These are the effective bases of philosophic reflection; they lie in a metaphysical theory of the rite much more than in a somewhat arbitrary mythology.
Why arbitrary? Because most of the myths were made up in order to justify fanciful etymologies by which the theorists of sacrifice fathered their own lucubrations on the Vedic hymns. While abstraction, scholastic classification, and realism in respect of the forces connected with worship appear as the things which make up Brahmanism, there are signs of a variety of traditions and great vagueness in the legends. Take, for instance, the cosmogonic stories, the first specimens of which appear in the late tenth book of the Ṛigveda. The first being was a golden egg (hiraṇyagarbha), floating on the waters. From it came the vital breath of the gods (x, 121). But where would the waters and the egg come from? Then we are told that the universal demiurge, Visvakarman, has his foot and his foothold everywhere, it being understood thereby that the world is in him, not he in the world (x, 81). Generation began neither with being nor with not-being, but with the One, which was at once single and unity. This principle breathed without breath (x, 129), for its life required nothing outside itself. Like a germ in the void, it came to birth by the heat of its own tapas. Then it desired, and that desire was the germ of thought. Since the gods came after that first creation, nobody will ever describe the creative power, and even if there is some transcendent being who presided over this creation, does he know it himself?
The famous passage which I have thus summarized is alike contradicted and corroborated by similar passages in the Brāhmaṇas. The Śatapatha first says that thought, which was neither being nor not-being, existed, and, wishing to become an ātman, heated itself with that object, and so on (x, 5, 3, 1); and then that in the beginning there were the waters, and that they warmed themselves in order to produce the golden egg from which after a year, the beginning of time, Prajapati was born (x, 16, 1). Further on the same work brings in a Brahman, by name Svayambhu (which we may translate into Latin as in se and causa sui), who creates by sacrificing himself (xiii, 7, 1). The Taittirīya Brāhmaṇa sets forth that the world, not yet being, thought "I wish to be", and, heating itself, produced all things (ii, 2, 9, 1). The corresponding Upanishad likewise holds that from not-being being proceeded, and made itself ātman (ii, 7). The very ancient Bṛihadāraṇyaka Upanishad already places at the beginning of everything an ātman, which says to itself "I am", and is at first frightened and then bored by its solitude and therefore splits into a first pair (i, 4). It would be idle to seek for consistency in these gropings of philosophic explanation, when systematization reaches its height in the dogmatism of sacrifice.
It is in the course of the Upamshads, not in the Brāhmaṇas, that philosophic explanation becomes systematic, gradually shedding the character of myth and flowing into the moulds the sacrificial dogmatism. Two conditions led to this result. Thought became freer, less dependent on the interpretation of ritual formalism; and the entirely new idea arose that meditation is a substitute for worship, and even an act more efficacious than the rite itself. This was a decisive turning-point in religious development. On the practice of rites there was superimposed a desire to understand the conditions of existence. The true and most operative sacrifice is to know, for it is by understanding more than by formulas that evil as well as error is eliminated. Knowledge (jñāna) tends to take the place of sacrifice (yajña). In this overturning of values karman, action, lost none of its dignity. It merely took on a different meaning; it came less and less to mean a rite and more and more to mean acting in general, of which knowing is only one kind. Everything was ready for the advent of Buddhism.
This intellectual revolution was effected, and justified, by the equation of brahman to ātman, magnificently expounded in the Upanishads. To have set up as their watchword, among many others—Vāch, Prāṇa, Ākāśa—brahman, the Vedic Word, and to have made it the absolute, was the final achievement of the Brāhmaṇas and the triumph of the Brahmans. We must bear in mind that that entity, brahman, represented the essence of the Brahman caste, as kshatram did that of the Kshatriya caste. To raise that brahman to an absolute was to give a metaphysical justification to the necessary, eternal supremacy of the priesthood, the sole performer of the religious operations which preserve the cosmic order and the sole heir of Vedic knowledge (which was the same thing). I do not mean to suggest that to identify ātman with brahman was to depose that brahman and substitute a different absolute. If that had been so, the Upanishads, instead of being the great sacred texts, would have been heterodox. But the truth of the Upanishads was certainly regarded as being of a different order from that of the Brāhmaṇas, since it was considered to be secret, esoteric. According to that truth, every individual in principle partakes of brahman, and therefore has the absolute in himself. It reveals to every mind that the great cosmic god, by whom the gods of the Veda were as it were eclipsed, is consubstantial and identical with itself. "Tat tvam asi—Thou art that absolute!" The ātman of each man is his self and at the same time his vital breath, but it exists only by participation in a universal life which also has its self. Except in size, if one may say so (although there are no degrees in the absolute), ātman is identical in the macrocosm and in the microcosm. A sort of pantheism, modified, indeed, by many limitations, became established in men's beliefs—a pantheism which to a remarkable extent lost sight of Vedic tradition and even of sacrificial dogma, a pantheism emancipated from the obsession of caste and apparently ascribing the same essence to all souls alike.
This meant a complete revolution in eschatological beliefs. All that the Ṛigveda wanted was that the pious man should live out his full life without premature death. After life the dead were received into abodes which were too indeterminate to be called either heavens or hells. The Brāhmaṇas, and, so far as they are continuations of them, the Upanishads, attempt a number of solutions of the problems of the next world. Sometimes the components of a human individuality go, after death, each to its natural place. Sometimes souls, with a subtle part of their body, travel across various regions of the world along routes strictly laid down according to the manner in which life has been lived on earth. But by the side of these theories of destiny, which indicate definite "ways" (gati), we find other solutions, which are simpler and more straightforward. A funeral rite feeds the ghost and prevents it from dying again or dissolving. Just as the offering of soma maintains the life of the gods, certain offerings secure for the dead "non-re-death" (a-mṛita), an expression which should not be translated as "immortality", for the cult must be kept up for ever if the deceased are to continue to exist. The precariousness of this ritualistic solution contributed to the prestige of the purely metaphysical solution contemplated by the Upanishads. The sacrifice which is knowledge not only delivers the soul from error and sin; it reveals to it its true nature, its permanent, indissoluble consubstantiality with the absolute. It gives infinitely more than a means of prolonging existence in the next world; it causes each one of us to discover and experience, as Spinoza says, that we are eternal. This solution overshoots its mark, it is true, for it ignores the problem. Instead of securing that a man shall have a fate appropriate to the moral value of his actions, it transfigures his existence in proclaiming his identity with the absolute. All his human error is to think that he is a man, when he is a god. It is an immoral solution, for the rogue is ātman as much as the saint; or rather one still has to explain why, if at bottom we are the absolute, the relative, evil and illusion, exists. One might say that Brahmanic thought became too expert in a certain kind of dialectics in its efforts to adapt the hymns to the requirements of sacrificial methods, and therefore leapt too fast, in its advance from the Brāhmaṇas to the Upanishads, towards the most abstract ontology. A revision of the postulates on which these views were based, a loss of interest in being and a curiosity about becoming, and above all a turn of the helm which directed thought towards moral realities, were the correctives brought to bear on Brahmanic theory by the heterodox sects of the sixth century before Christ.
1 CCLXXXVIII, vol. i, p. 66.