The Inner Conflict of Tradition: Essays in Indian Ritual, Kingship, and Society c17: 13 Caste and Karma: Max Weber's Analysis of Caste

The Inner Conflict of Tradition: Essays in Indian Ritual, Kingship, and Society c17

Language: English

13 Caste and Karma: Max Weber's Analysis of Caste

c17.1

Apart from its obvious place in Max Weber's study of Hinduism, caste, though fairly often mentioned, does not appear to be a central concern in his thought. Weber, it would seem, is mainly interested in caste as a particular, even extreme case of the closed estate (*Stand*) as different from *Klasse*. The distinctive element that turns *Stand* into its extreme form, caste, is that it is underwritten by the ritual sanction of pollution.¹ With Louis Dumont, we may note in passing how Weber on this basis achieves a remarkable junction of hierarchy, ethnicity, and division of labor.² In an imaginative tour de force, he pulls together disparate notions on the nature and origin of caste in a succinct, logically coherent formula. In Weber's view, the caste phenomenon comes down to "eine umgreifende Vergesellschaftung die ethnisch geschiedenen Gemeinschaften zu einem spezifischen politischen Gemeinschaftshandeln zusammenschliesst".³ This formula may count as a school example of Weber's particular use of the concepts of *Gemeinschaft* and *Gesellschaft*.⁴

c17.2

For Weber, however, the point is the difference between the situation where ethnic groups, such as Jews and Gypsies in Europe, are tolerated, sometimes even privileged, because of their particular economic function and so are juxtaposed to a political society and the caste order that joins together the ethnic communities in a hierarchy of honor to create a political society in the common interest. The formula has been criticized for attempting to do too much, namely, connecting the hierarchical view of society with the exceptional situation of minority communities through the intermediary of the supposedly racial origins of caste.⁵ Its advantage, however, is that it keeps caste within the bounds of a comparative discourse.⁶ But

c17.3

This essay will appear in slightly different form in *Max Weber's Studien über Hinduismus und Buddhismus: Interpretation und Kritik*, edited by Wolfgang Schluchter (forthcoming).

c17.4

what makes caste in Weber's view a uniquely Indian or Hindu phenomenon is not hierarchy underwritten by the sanction of pollution, but its tie-in with the brahmanic karma doctrine of retribution, the most consistent theodicy ever produced, as he puts it.⁷

c17.5

Analyses of the caste system, however much they may differ from each other, agree in at least one point. They all stress its immutability and total command over society. So also Weber's analysis. One may enter caste society as a group—by way of *Vergesellschaftung*—or one may leave it as an individual by renouncing the world, but the system remains unchangeable. Its grid can be endlessly refined, but it cannot be breached or transformed. In fact, Weber steps up and finalizes this line of thinking by bringing in the karma theodicy. In his view, this doctrine is the coping stone of the caste system, for it entails strict adherence to one's caste duties and rewards such adherence with the prospect of ever-improving rebirths, which will eventually lead to the ultimate goal of salvation. In this connection, Weber even comes to qualify the junction of caste and karma as a *Kastensoteriologie*.⁸ And although he is perfectly aware that his intellectual doctrine cannot have been comprehended by all and sundry, Weber does argue that the promise of a progression to higher and better rebirths held out to the lower castes in a significant degree contributes to the latter's often observed conservatism.⁹

c17.6

This linkage of caste and karma, Weber insists, is a pure product of rational ethical thought.¹⁰ The caste order is in the last resort explained as the junction of two tendencies whose opposition is fundamental in Weber's thought. On the one hand, he analyzes caste society in magico-religious terms and qualifies it as thoroughly traditionalist and antirational.¹¹ On the other hand, it is inextricably linked with the ethical rationalism of the brahmanic karma theodicy. The remarkable point is that it is exactly this otherwise volatile mixture that is made responsible for the caste system's exceptional coherence and stability. If there is a paradox here, Weber does not seem to have been aware of it.

c17.7

This does not mean, of course, that Weber was wrong in juxtaposing the two opposite principles of magico-ritualism and ethical rationalism. He shows himself, moreover, most sensitive to the tensions resulting from their competition and interpenetration.¹² However, in his view, this tension is blunted by an attitude of relative indifference or passivity toward the mundane order.¹³ Deprived of its absolutist claim to transform the world, ethical rationalism gives way to the relativistic accommodation of an "organicist" social ethic.¹⁴ In the Indian case, however, Weber argues that the perfect fit, the *sehr besondersartige Verknüpfung*,¹⁵ of caste and karma did not produce just another accommodation but an absolute and unassailable social order. Far from prising it open, the ethical rationalism of the karma theodicy closed the mundane order of caste and sealed it with the prospect of ultimate salvation. The question is, Can we accept this monolithically coherent picture?

c17.8

Now, Weber is certainly right in recognizing that in the Hindu view the mundane order is not as denuded of soteriological value as the prominence given to world renunciation may suggest. It is, however, more than doubtful whether this value can be located in the caste system. For one thing, the connection of caste with karma retribution is not as generally and unequivocally stated as the importance ascribed to it by Weber would require. We do find strong, authoritative statements to the effect that salvation is open, not only to the renouncer, but equally to the householder (*gṛhastha*).¹⁶ But it should be noted that it is the order of the gṛhastha—irrespective of caste—that is endowed with soteriological value, not caste as such. This may seem a quibble. The point, however, is that the nexus of caste and karma is far from unequivocal.

c17.9

Let us look into it more closely. One should, of course, conform to the ways and habits of one's caste, for it is better to perform the duties of one's own caste, one's *svadharma*, indifferently than those of others with outstanding distinction. But what if these customs and duties, such as meat eating and impure occupations, are contrary to the precepts of dharma? For instance, to refuse to eat meat or, say, to participate in pig hunting if one happens to be a royal rājput would cause a scandal in one's community. Such refusal is therefore stated to be reprehensible. Conversely, one can not be blamed for following the code of conduct that is particular to one's community, even if it would constitute *adharma*.¹⁷ In a similar vein, the authoritative *Mānava Dharmaśāstra* concludes its exposition of nonviolence (*ahiṃsā*) and the consequent prohibition of meat by saying that, after all, there is no sin in eating meat, this being simply the way of the world (*pravṛttir eṣā bhūtānām*), but that abstention (*nivṛtti*) is rich in religious reward.¹⁸

c17.10

On the face of it, this statement may seem to be a rather lame attempt at harmonization. It does show, however, precisely where the problem lies. Ahiṃsā, abstention from destroying life, is an absolute, transcendent value, and for that reason, it is beyond the grasp of the world and totally at variance with normal practice, with *pravṛtti*. Pravṛtti, then, means going on with the business of life, following the practices of the community and thereby occasionally departing from the ultimate requirements of the dharma. As pointed out by Robert Lingat, there are no temporal sanctions against such ultimately adharmic behavior, nor can there be.¹⁹ But one does not escape the religious or metaphysical sanction of one’s karma. That is, as long as one follows the ways of the world, one is obliged to adhere to the customs of the community and so remains enclosed in it, for the inexorable karma accumulated by the faithful member of his caste will necessarily keep him in line with his community from rebirth to rebirth. And this situation will endure as long as the community—in contradistinction to the individual renouncer—does not conform its practices to the precepts of the dharma, a step that is nowadays known to sociologists by the confusing name “Sanskritization.” However, such conformity can at best be only partial, for pravṛtti, going on with the business of life, as any community is bound to do, is incompatible with its opposite, nivṛtti, turning away from the ways of the world. On the other hand, the individual who would renounce his community’s way of life would thereby break its rules and risk considerable scandal. Whichever way we look, the dilemma remains wide open.

c17.11

Viewed in this light, the nexus of karma and caste does not look very promising as a way to ultimate salvation, and it is certainly wide off the mark to speak of a *Kastensoteriologie*. Incidentally, we can now well understand that insofar as the mundane sphere offers a way to salvation, it is not primarily attached to caste and hierarchy but, as we saw, to the dutiful performance of the householder’s rites and duties enjoined by the Veda. This does not solve the essential problem either, but only moves it on to another field, namely, that of the relationship of the Veda to the mundane sphere.²⁰ This problem need not concern us here. At this point, it may be sufficient to note that insofar as a nexus between caste and karma is acknowledged, it can hardly lead to salvation. When caste is brought into play, it is not so much the loyal adherence to one’s caste duties but the renunciation of the “fruit” of one’s dutiful activity that neutralizes the inexorable law of karma and holds out the promise of salvation—an inner-worldly renunciation through totally disinterested action, which then threatens the world with complete meaninglessness and so again leads to an unsoluble dilemma.

c17.12

More important, however, is that, as we saw, the dharma shows itself to be not so much a closed and fixed order as a wide-open problem. Now one may try to circumvent the problem by reasoning that dharma is not a law that is imposed, but one that is proposed, as Robert Lingat aptly puts it.²¹ Or as Louis Dumont, borrowing the well-known phrase about constitutional kingship, says, “Le dharma règne de haut sans avoir, ce qui lui serait fatal, à gouverner”.²² However, whether dharma is only proposed without being imposed, or reigns from on high without actually governing the world, it cannot resign its function of ordering the affairs of the world. Here, as before, we come up against the unforgiving rift between the mundane sphere and otherworldly renunciation, between pravṛtti and nivṛtti. The dharma is and must be transcendent and eternal, but by the same token, it is divorced from the world. Coming down to the mundane sphere to govern it is indeed fatal for the dharma. Yet it cannot avoid doing so on pain of losing its universal validity.

c17.13

Allegedly there once was a golden age, the *kṛtayuga*, when the dharma not only reigned from on high but fully governed man’s affairs on earth. But even in that golden age, the matter was not unambiguous, for how is it that in spite of the dharma’s governance, the world could progressively fall away from it till it reached our own wretched age, the irredeemable *kaliyuga*, fourth and last of the world periods? The answer, as given in a passage from the *Mahābhārata*, is in the very perfection of the dharma’s governance.²³ We are told that in those blessed times, people protected each other with the help of the dharma alone (*dharmenaiva*)—as, of course, they should. But it is precisely thereby that, as our text dryly explains, the people became overwhelmed and confusion entered the world. This, then, is meant to explain the necessity of kingship. Now the king is indeed the mediator between dharma and world. But this does not solve the problem, for kingship could do little if anything to stop the progressive weakening and eventual dissolution of dharma in the successive world ages. The institution of kingship could only offer a focus for the unresolved and essentially insoluble tension. Far from patching over the rift, mythic imagination shows it to have been there from the very beginning, waiting to be focused on the king. We can now see how fatal its descent into the world is for the dharma, for in order to be universal, it has to accommodate within itself its own opposite, the principle of confusion and disorder. That, after all, is what is meant when the texts emphasize the utter subtlety of the dharma. This is not the subtlety of literate doctrine but the tension that only the sage can handle, even though he cannot resolve it.

c17.14

When the dharma shows an essential “fault line,” we should expect the caste order to be similarly broken. In fact, this has already become apparent from our discussion of the conjunction of caste and karma. Far from forming a coherent whole, caste and karma are at opposite ends and mutually disruptive. The point is that karma refers to the individual’s fate; caste and its svadharma, however, refer to the collectivity. As we saw, it is the caste-bound collectivity that determines the individual’s karma and thereby prevents him from working out his own salvation. Only when man realizes his individuality to the full extent and steps out of collectivistic society, will he be able to stop the inexorable karma mechanism. This, rather than an organistic theodicy, is the actual message of the karma doctrine. It breaks away from the collectivistic world toward individual salvation. Here the organicist restraints of caste are in conflict with the rationalistic karma doctrine. The juxtaposition of caste and karma, then, is indeed a volatile mixture. For the consequence of the karma doctrine is no less than a call to deconstruct the world. It is not fortuitous that its first formulation in the Upanishads takes the form of a secret doctrine not to be publicly divulged but to be discussed in private, outside the assembly.²⁴ Its ultimate consequence, world renunciation, threatens to break up society. Understandably, renunciation has always encountered muted but obstinate hostility.²⁵

c17.15

The conflict arose from the infusion of the renouncer's individualistic thought and rationalistic values into the otherwise devalued magico-religious world of caste. This created an irreparable rift, not only between the renouncer's otherworldly sphere and the mundane order of caste, but also within the caste world itself. Although usually supposed to be completely closed and internally coherent, caste does indeed show a rift. This is not just a matter of actual practices where the rift is obvious but all too easily dismissed as another instance of India's proverbial diversity. It is equally and clearly expressed in theoretical statements in the dharma texts. There we find at least two terms, *varṇa* and *jāti*, where we make do with only one, caste, suggesting an unbroken unity. Admittedly, *varṇa* and *jāti* are often used interchangeably, but the question is how the set of four *varṇas* are related to the practically innumerable *jātis*. One might, of course, take the *jātis* simply as subdivisions of the four canonical *varṇas*. The explanation of the dharma texts is, however, subtly different. The *jātis*, we learn, arose from marriages—that is, mixed marriages—between members of the "original" four *varṇas*.²⁶ This explanation has, of course, much to recommend itself. But the point is that here the preordained and immutable caste order breaks open. With a clarity that leaves nothing to be desired, we are told that the caste order with its innumerable *jātis* arose from the evil of *varṇasaṃkara*, the mixing of the *varṇas*. The caste order, then, is undermined from the beginning, built as it is on the "fault line" of *varṇasaṃkara*.

c17.16

Now it can be shown, as G. J. Held did almost half a century ago, that originally the *varṇa* concept, instead of prescribing strict separation, implied a system of connubial and other exchanges.²⁷ The decisive turning point was the rejection of such inter-*varṇa* exchanges and their replacement by the opposite rule, which set the *varṇas* rigorously apart from each other. It will not be possible to date this turning point in absolute chronological terms. It can, however, be related to the rationalized Vedic ritual outlined in the brāhmaṇa texts. Here, as also in the rise of Buddhism and Jainism, we find the reflection of India's version of the "Axial Age," or in Weber's terms, the onset of "disenchantment" and transcendentalism that split the world between two principles.²⁸ The unified world of sacrifice, held together by agonistic, potlatch exchanges is irreparably broken.²⁹ The new rationally systematized ritual excludes the community, and the mythical presence of the gods evaporates. Alone the sacrificer stands on his place of sacrifice. There, through the artifice of the ritual, he creates his own rigorously ordered universe separate from and transcending the mundane world.

c17.17

The motive behind this radical change was the rejection of the ultimately self-defeating conflict and violence that were the essence of the original sacrificial exchanges. To eliminate conflict and violence, the community had to be broken up and the parties separated. This, not ethnic "apartheid," is the reason why the *ārya* and *śūdra* varṇas had to be separated from each other. Originally both had taken part in the agonistic sacrificial festival as opposite parties—a situation preserved in an innocuous form in the Vedic *mahāvrata* ritual, where the two parties are still described as contending with each other for the "goods of life."³⁰ In contrast, the rationalized Vedic ritual excludes the opposite party and sets the sacrificer apart in a transcendent world of his own. The end result is the individual sacrificer performing unaided and unopposed the sacrificial ritual in himself and by himself; in short, he is the ideal brahmin, in this respect, no different from the renouncer.

c17.18

In the same way as the Vedic ritual, the varṇa theory breaks up society. It does not stop at dividing society into four strictly separate blocks but goes a long way in breaking down the internal cohesion of each varṇa too. This becomes clear when we look at the rules pertaining to marriage. One should, of course, marry within one's varṇa, but the rules about forbidden degrees of relationship (*sapiṇḍa* relatives) are such that they exclude the otherwise normal—even normative—practice of marrying within the circle of known relatives (such as one's mother's brother's daughter). If strictly adhered to—as, of course, it hardly ever is or can be—the prohibition would prevent the formation and maintenance over the generations of a web of marriage exchanges. The exclusion of the sapiṇḍa relatives would cause the affinal network to crumble with each new generation, leaving the single patriline isolated. The intention clearly is to atomize society.

c17.19

Of course, no society can rest on the single principle of separation alone. And so varṇa, like dharma, has to encompass its own opposite, namely, varṇasaṃkara, the confusion of the varṇas, resulting in the proliferation of the innumerable jātis. The seemingly monolithic caste order, then, is torn apart by two diametrically opposed principles, strict separation as required by the varṇa theory and conflictual interlinking in the actual jāti order. To handle this unresolved tension is the function of the king. As upholder and mediator of the dharma, the king is called upon to set the world of jāti right when confusion and conflict are rife. However, belonging himself to the jāti world, he lacks the authority to do so. Such authority has to be derived from the otherworldly sphere and can only be imparted by the brahmin (or the renouncer).³¹ For the brahmin is ideally above and beyond the jāti order. He is complete in himself and independent from the world. Only the brahmin can qualify for all six dharma activities—performing the ritual for himself as well as for others, learning and teaching, giving as well as receiving gifts. The two other "twice-born" varṇas, the kṣatriya and the vaiśya, are qualified for only one of each pair: sacrificing, learning, and giving, while the śūdra is only allowed to serve. That is, they are all dependent on others. Only the brahmin combines in himself both ends of each of the three pairs of activity and so he is, ideally, fully independent from others, transcending thereby the mundane order of conflict and interdependence. By the same token, however, he is—like the renouncer—cut off from the king's world. Indeed, if he is to preserve his transcendent authority, the brahmin must shun the king and all his works, even his gifts and benefices.³² Yet he cannot live on his purity and transcendence. For his sustenance he is obviously, though contradictorily, dependent on the king whom he should avoid. So king and brahmin need but cannot reach each other, because the brahmin must opt out of the necessary compact.

c17.20

So the unresolved tension within the worldly order of caste bears down in full force on the dual agency of king and brahmin. On the one hand, we find the collectivistic world of the jātis tenuously held together, or rather constantly rearranged, by the king; on the other hand, there is the brahmin whose individualistic varṇa concept is aimed at decomposing this world. In the open space circumscribed by these two opposite and incompatible principles, the arrangements for handling their tension have to be worked out time and again. The tension is insoluble and the arrangements for handling it therefore remain open to recall and replacement. In other words, the very insolubility of the tension provides a dynamic that opens up the world, invites new ways of handling it, and favors pluralism.

c17.21

One question remains. How is it that Weber, who otherwise was so sensitive to the tensions created by transcendental thought and world rejection and who shows the way to analyze such tensions, has failed to recognize and exploit them in his analysis of Hindu society? The answer, at least in part, is simply in the sources he had to use, notably, the Indian census publications and studies deriving from these materials. He had to base his theory on the view of Indian society, and in particular of caste, produced by the modern state for its own purposes. Now the modern state—in contradistinction to the ancien régime—is hived off from society and pretends to govern it by remote control as it were. To that end, it first of all needs an all-inclusive and immutable grid of rigidly bounded and inflexible categories. The grid can be refined, but it cannot be shifted or altered, on pain of losing its universal character and usefulness. This need for an immutable grid of categories was filled with deplorable obviousness by caste, seemingly custom-made for the purpose, especially in its brahmanic form of varṇa separation. Conversely, the modern state and its census grid of caste could not but project the image of an unchangeably fixed order of society. One may wonder whether and how far the notion of a never-changing, utterly tradition-bound, and stagnating India has been formed by the modern state's view of society. But this is another matter. For our present purpose, it may perhaps be relevant that the census definition of caste reached its height of authoritative elaboration with the censuses of 1901 and 1911, that is, at the time of Weber's study of Hinduism.³³

c17.22

Given this material and its background, we can hardly expect Weber to have set himself against the current of authoritative opinion in an area where he was a stranger. What he did achieve, though, was a highly original attempt at a consistent explanation of caste as it was presented to him, fitting it into his overall scheme of thought.

c17.23

Incidentally, it may be that Weber's thoroughgoing "rationalization" of the still fairly chaotic and uneven information on caste may have contributed to his failure or, at least, to his ambivalence in recognizing the full impact on Hinduism of the "Zerspaltung der Welt in zwei Prinzipien" and the quality of the resultant tensions. He speaks of an ontological dualism of worldly transience and unmutable eternal order, but repeatedly refers to the transcendental sphere as a *Hinterwelt*, thereby suggesting the unbroken monism of the magico-religious world view.³⁴

c17.24

Returning to our query, the census view of caste created by the modern state will only provide part of the explanation for Weber's monolithic view of caste and karma—and the easy part at that. Perhaps, at a deeper level, Weber emotionally felt the need for the wholeness of a perfectly ordered world under the authority of a rational social ethic, while at the same time fearing its constricting consequences. Hinduism, then, may have been for him the screen on which to project this double and contradictory image. He seems to refer to this when, at the end of his chapter on caste, he points out that the very consistence, as he saw it, of caste and karma is a fearful perspective for the individual who asks for the meaning of life caught in the inexorable mechanism of repeated death.³⁵ Here Weber touches upon the inner rift of the caste order, for the tension created by the perspective of meaninglessness not only sets the mundane order apart from the renunciatory sphere, but breaks open the caste order.

Linked texts (3)

Mānava Dharmaśāstra — Manu & Bhṛgu, Sanskrit
Mahābhārata — Vyāsa, Sanskrit
Upaniṣad — Sanskrit