1

Social and Literary History of Dharmaśāstra

The Foundational Texts

Patrick Olivelle

How and why did the genre of literature called Dharmaśāstra come into being? Who invented it and for what reasons? These are questions hardly ever raised by historians of this literature. Most take its existence for granted as obvious and self-evident, just like other similar genres such as the Śrautasūtras and Gṛhyasūtras. In a previous study, I raised this issue and attempted to provide a hypothetical answer (Olivelle 2010b). Here I want to emend and expand that answer, taking into account the new discoveries regarding the concept of the gṛhastha (householder) presented in Chapters 5 and 8 of this volume.

Briefly, in the previous study, I concluded that the emerging centrality of the term and concept dharma both in Buddhism and in the imperial ethics propagated by Aśoka in the middle of the third century bce prompted Brahmanical theologians to define their own religious way of life in terms of dharma . As I have shown elsewhere (Olivelle 2004a , 2005c ), dharma , a neologism coined by the poets of the Ṛgveda , 1 was not a central term in the theological vocabulary of the middle and late Vedic periods when the Brāhmaṇas and the Upaniṣads were composed. It also occupied a marginal position in the ritual sutras, the Śrautasūtras, and the Gṛhyasūtras. I argued that it is the theological development within Brahmanical thought making dharma its central concept that resulted in the creation of texts devoted to the definition and explication of the Brahmanical dharma in opposition to the various dharmas underlying the doctrines and lifestyles of non-Brahmanical ascetic communities such as Buddhism and Jainism. Although I think this hypothesis is still valid, it is, however, incomplete. There were other more specific factors influencing the emergence of Dharmaśāstra.

It is obvious that many of the provisions in Dharmaśāstras are addressed to the married householder, especially the Brāhmaṇa householder; he is the lynchpin of the Dharmaśāstric system. This is true also of the ritual sutras. In much of this literature, however, that householder is what Jamison calls “the unmarked subject” of the provisions; most frequently he is not designated by a specific term but is the implied subject of the verbs in the third person singular. The term most frequently used in Vedic texts, gṛhapati , however, is totally absent in the Dharmaśāstras, including the earliest sūtra texts. Its place is silently replaced by gṛhastha , a term that is absent in the literature prior to the Dharmasūtras (see Jamison’s remarks in Chapter 9 ). I think it is this unremarked and seemingly unremarkable terminological shift that holds the key to the beginning of the Dharmaśāstric genre of literature. As Jamison states, “This terminological demarcation hints at a conceptual discontinuity as well, and the linguistic history of the term gṛhastha illuminates the conceptual renewal.” It also, I argue, underpins the very necessity for the creation of this new genre of literature.

As I note in Chapter 5 , the term gṛhastha , although new to the Sanskrit vocabulary, is found in several Prakritic forms such as gahatta and is already used by Emperor Aśoka in the middle of the third century bce . In his usage, gṛhastha is always coupled with and contrasted to pravrajita , the ascetic who has gone forth from home into the homeless life. As Jamison points out, gṛhastha should be properly viewed as the “stay-at-home” in contrast to the “gone forth,” rather than simply as any married householder. Both of these kinds of religious people are presented as members of a religious community or organization termed pāṣaṇḍa by Aśoka (Rock Edict 12). In Pillar Edict 7, moreover, one of the pāṣaṇḍas is identified as Brāhmaṇa. We can conclude, therefore, that in the eyes of Aśoka the Brāhmaṇa community also constituted one among the many pāṣaṇḍa groups within his empire, and it, like the others, contained two kinds of members: gṛhasthas and pravrajitas.

At some point during this period, probably in the third century bce or a bit earlier, new theological developments 2 appear to have taken place within the Brahmanical intellectual classes. One such development, as I point out in Chapter 5 , was the āśrama system, which expanded the twofold Aśokan classification into four. This theological development alone, however, could not have been the catalyst for the creation of the new genre of literature. The reason is twofold. First, some early writers on dharma , such as Gautama and Baudhāyana, reject the āśrama system as propounded by its advocates proposing instead the “single-āśrama ” (aikāśramya ) theory: there is only one āśrama, namely, that of the gṛhastha . Second, none of the early Dharmasūtras incorporates the system into the structure of their compositions; the sections on the system give the appearance of being parenthetical or appendices to the main body of the works. It is the gṛhastha , now modeled after the Vedic householder and following a ritual cycle centered on the “five great sacrifices” (pañcamahāyajña ), rules of purity and diet, and Vedic recitation, who occupies center stage. Yet, the writers never use the old Vedic term gṛhapati but the novel gṛhastha , a term that had no prehistory in the Vedic literature. So, the Dharmaśāstric writers, while rejecting or marginalizing the theology of āśramas , nevertheless operated within the newly emerging conceptual world and its vocabulary. Further, the Dharmaśāstric gṛhastha is not simply a ritualist; the rules given in these texts promote virtuous and holy living. These texts present the dharma , in the sense of proper behavior called ācāra , to be followed by a gṛhastha . I want to explore further these competing theological innovations to lay the groundwork for assessing the possible motives for creating this genre of literature.

The hypothesis I propose is that, rivaling the theology of the āśramas , which presented a variety of lifestyles, especially the gṛhastha and the pravrajita , as alternative religious paths, a new theology appears to have been constructed asserting the centrality of the gṛhastha . This theology probably represented the mainstream of Brahmanical tradition. Yet it was markedly different from the Vedic theology centered on Vedic rituals and represented, as Jamison puts it, “a conceptual discontinuity.” It had a lot in common with the āśramic theology, and the two coexisted in some fashion—sometimes in conflict and sometimes in harmony, but always in tension—throughout Dharmaśāstric history. Its debt to the āśramic theology and śramaṇic vocabulary is evident not just in the adoption of the term gṛhastha for its central homo religiosus , but also in presenting the household life as an āśrama , indeed, as the only legitimate āśrama in the view of Gautama and Baudhāyana (see Chapter 5 ). I think it is the dialogue and disputes between these two Brahmanical theologies (and perhaps others that we cannot readily identify) that are captured in the texts of the Dharmaśāstric tradition, disputes that continue well into the medieval period. We see them articulated in the strong defense of the householder as the highest form of religious life. As Vasiṣṭha (8.14–8.16) says:

A householder alone offers sacrifices; a householder performs austerities. Of all the four āśramas , the householder is the best.

As all rivers and rivulets ultimately end up in the ocean, so people of all the āśramas ultimately end up in the householder.

As all living beings live dependent on their mothers, so all mendicants live dependent on the householder.

On the other hand, with the emergence of the ideal of liberation (mokṣa ) shared by Brahmanism and ascetical theologies such as Buddhism, the wandering mendicant (pravrajita or bhikṣu ) came to be seen as the figure most closely associated with that ideal. This is clearly revealed in the vocabulary of Manu, who calls it mokṣāśrama , the āśrama leading to liberation.

It was this gṛhastha theology that provided the impetus to the creation of the Dharmaśāstric genre of literature. This explains both the centrality given to the gṛhastha and the appearance, often somewhat marginally, of the āśrama system. Yet, I think that the creation of this genre was not simply the result of the interactions between these two Brahmanical theologies. A major factor was what we may call interreligious debates on the concept of dharma between Brahmanical theologians and those representing the ascetical or śramaṇa ideologies, especially Buddhism. The concept of dharma was very much a site of contention and debate. What is dharma ? And how and where do you find it? In other words, the epistemology of dharma (dharmaparmāṇa ) was a central theological issue (see Chapter 3 ). Buddhist theologians had a clear position: buddhavacana (the words of the Buddha) is the sole epistemic source of dharma . Either proximately or ultimately all valid pronouncement on dharma must go back to the ipsissima verba , the very words, of the Buddha. This position is encapsulated in the opening words of every Buddhist scriptural statement: evaṃ mayā śrutam (Pāli: evam me sutam ), “Thus have I heard.” It thus comes as no surprise that all the early Dharmasūtras begin with the epistemology of dharma . 3 This feature of the texts on dharma stands in sharp contrast to other similar Brahmanical texts such as the Śrauta- and Gṛhya-sūtras, which saw no need to state where they get knowledge from, taking the epistemological issues as self-evident and noncontroversial.

It is, then, from within this theological ferment that the genre of Dharmaśāstra was born. Yet, I think there is another significant element that, even if it was not a causal factor, shaped the structure and tenor of these texts. That is the system of varṇas . It is clear that the varṇa system was not an objective and disinterested classification of ancient Indian society. It was from the start an ideologically driven enterprise designed to place the Brāhmaṇa at the top of a pyramidal social hierarchy, supporting the claim to power of the Kṣatriya class, and in a special way, reducing the Śūdras and other lower classes to a marginal and oppressed status. This is clearly indicated in the foundational document on the varṇas, the Puruṣasūkta (ṚV 10.90), in which the vertical structure of the human body provides the basis for the hierarchical structure of the varṇa system. The Śūdra, born from the feet, is placed at the bottom. So, to uphold and to promote the varṇa system is at the same time to uphold the supremacy of the Brāhmaṇa class and its exceptional status.

This, I think, was a crucial element of the Dharmaśāstric project. One may question the need for Brahmanical theologians to assert aggressively the varṇa system; many scholars, after all, take the system to be not just old but also reflecting the social reality of the ancient Indian society. I think this is a mistake, because all the evidence presented for this claim comes from Brahmanical sources, which generally sought to shape rather than simply to reflect social reality. Aśokan inscriptions constitute one of the few independent sources, and they are completely silent on the varṇas ; the very term is absent in them and so are terms for three of the four varṇas : Kṣatriya, Vaiśya, and Śūdra. They do mention the Brāhmaṇa but not as a varṇa but as a religious group, first as a counterpart to the śramaṇa and second as one among the many pāṣaṇḍas . Further, the Aśokan reforms greatly undermined Brahmanical exceptionalism. The special relation between king and Brāhmaṇa advocated in the Vedic texts was eliminated. The need to assert and reassert the centrality of the varṇa system with the Brāhmaṇa at its apex was never more urgent.

The dharma articulated in the Dharmaśāstras was not simply a narrowly religious one centered on the holy life of a gṛhastha or of those belonging to the four āśramas ; it was also a sociopolitical blueprint for the proper management of society by the king. Even the early Dharmasūtras contain sections on family, civil, and criminal law, and on governance by the king, however rudimentary these appear in comparison to the detailed treatments of these topics by later authors such as Manu and Yājñavalkya. In the sociopolitical ethic of the Dharmaśāstra, the varṇa system and Brahmanical exceptionalism are fundamental elements.

Such, I think, was the religious, social, and ideological background for the creation of the genre of literature known as Dharmaśāstra. But what was the actual institutional framework of inquiry and education that produced the early texts on dharma ? We can look at the parallel literature, the ritual sutras, for a model. These were produced within specific “schools” or caraṇas belonging to the various Vedic branches (śākhā ). It is reasonable to assume that Dharmasūtras also were produced by the same kinds of individuals who produced the ritual sutras and within the same kinds of educational settings. Looking at the extant works, we have two that are attached to precisely such caraṇas and are ascribed to their respective founders: the Dharmasūtras of Baudhāyana and Āpastamba. 4 The other two, those of Gautama and Vasiṣṭha, however, are independent of any caraṇa . It is fair to assume that there were educational and intellectual homes other than the caraṇas to engage in scholarly activities, as demonstrated by the composition of the Upaniṣads, the grammatical treatises, and the literary activities that gave rise to the Sanskrit epics.

In this context, I think we should extend Jamison’s conclusion of “a conceptual discontinuity” from the notion of gṛhastha as such to the broader literary genre of Dharmaśāstra in which the gṛhastha occupies center stage. In other words, I think the theologians who composed these texts were operating with a different Weltanschauung than their ritual counterparts, even as they shared the Vedic ritual and mythological world. This Weltanschauung was very much molded by ascetic ideologies, values, and vocabularies, as we see most prominently in the term gṛhastha itself and in the value that these texts place on sexual abstinence, fasting, and other ascetic practices. We need not assume that these new intellectuals shared a uniform theology; but they did share a broad vision of what it was to be a good and religious Brāhmaṇa.

The Early Textual Production

When the earliest texts on dharma were composed is difficult to determine accurately, and all the dates proposed by scholars are at best educated guesses and conjectures. There are two anchors, however, that permit us to make an objective, though imprecise, assessment. The first consists of the various factors underlying the creation of this genre that I have discussed above. Although Aśoka provides us a definite date for their articulation, that is the third century bce , some of Aśoka’s vocabulary and classificatory systems, such as pāṣaṇḍa and gṛhastha , may have preexisted their use by him.

The second consists of the use of the terms Dharmaśāstra and Dharmasūtra by authors external to this literary tradition. The earliest such reference comes from the grammarian Kātyāyana in his Vārttika 39 (on Pāṇini 1.2.64): dharmaśāstraṃ ca tathā (“Likewise also the Dharmaśāstra”). Patañjali commenting on this gives the examples of such Dharmaśāstric injunctions: “A Brāhmaṇa should not be killed. Liquor should not be drunk.” 5 Elsewhere in his commentary, Patañjali (on Pāṇini 1.1.47: I: 115) himself uses the term Dharmasūtra in discussing the interpretive rule that special rules or exceptions set aside the provisions of general rules: naiveśvara ājñāpayati nāpi dharmasūtrakārāḥ paṭhanty apavādair utsargā bādhyantām iti / “Neither does the Lord command nor do the authors of dharmasūtras declare: ‘Let general rules be set aside by exceptions/specific rules.’” Clearly, here Patañjali asserts the authority of Dharmasūtras in matters of hermeneutics. Now, the scholarly consensus today is that Kātyāyana should be assigned to a period after the Maurya reforms (Deshpande 2006 ), that is, to the end of the third or the beginning of the second century bce , and Patañjali to the middle of the second century bce .

We cannot be far wrong, then, in dating the beginning of this genre of literature to the first half of the third century bce . As the reference in Patañjali indicates, the earliest extant texts of the genre were composed in aphoristic prose and were therefore called dharmasūtra . The four extant ones are ascribed to Āpastamba, Gautama, Baudhāyana, and Vasiṣṭha, listed according to their probable chronological order. The relative and to a degree the absolute dating of these texts is aided by the term and category dvija/dvijāti , the twice-born or man with two births. As I have dealt with this issue extensively elsewhere (Olivelle 2012a ), I will state its conclusions here briefly. The term and concept of dvija are absent in the entire Vedic corpus, including the ritual sutras. The term is also absent in the Āpastamba Dharmasūtra and in Patañjali; the latter is instructive, because his Great Commentary is a mine of cultural information, and he is exceptionally well informed about Dharmaśāstric notions. The earliest extant text to use the term is the Gautama Dharmasūtra . From then on it becomes a cornerstone of the Dharmaśāstric project and its use is common and frequent.

The conclusion, then, is that the category of dvija was absent in the earliest period of Dharmaśāstric textual production, and that the category was invented after about the middle of the second century bce . The purpose, as I have noted in Chapter 5 , was to bring under the hegemony of Brahmanical ideology articulated in the Dharmaśāstras all the “upper” levels of the social hierarchy and to thereby exclude other segments of society, including non-Brahmanical religious traditions, that are often termed Śūdra in these texts. That dvija was a technical term restricted to the Dharmaśāstric theologians is indicated by its complete absence in the Buddhist Pāli vocabulary. The term was probably invented toward the end of the second century or at the beginning of the first century bce .

Using this and other criteria, 6 I have assigned the following probable dates to the four Dharmasūtras: Āpastamba = third to early second century bce ; Gautama = late second to early first century bce ; Baudhāyana 7 = mid first century bce to early first century ce ; and Vasiṣṭha early to late first century ce .

These texts provide us a glimpse into what subjects the early writers on dharma wanted to include in their works and, thus, into what they considered dharma. First, we have as a preamble two topics: the sources of dharma and the four varṇas including the mixed castes and proper occupations of the respective varṇas. Then, there is a section on Vedic initiation and a student’s duties, including discussions on the teacher and the conclusion of the period of studentship. In somewhat overlapping sections, texts deal with the student who has returned home and the so-called snātaka (bath-graduate; Chapter 7 ), types of marriages and of sons (Chapter 8 ), including adoption, dharma with respect to women (Chapter 17 ), Vedic recitation, the annual course of study, and times when recitation is forbidden, salutation and rules of precedence, rules on answering calls of nature and on impurity, purification of body and articles, sexual intercourse, dietary rules (Chapter 12 ), hospitality and reception of guests, inheritance (Chapter 16 ), funerary rites, periods of impurity upon a death or birth in the family (āśauca ), ancestral offerings (Chapter 13 ), and sins and penances (Chapter 24 ). Further topics outside the main ones dealing with the householder include rules during times of adversity (āpaddharma ; Chapter 19 ), the four āśramas (Chapter 5 ), family, civil, and criminal law (Chapter 23 ), and the duties of a king (Chapter 20 ). Some texts, such as that of Gautama, have more extended discussions of lawsuits and rules of legal procedure (Chapter 22 ).

Besides the opening discussion of the epistemology of dharma (Chapter 3 ), a unique feature of the Dharmaśāstras, as opposed to the ritual sutras is that they begin with the Vedic initiation of an adolescent boy. All ritual sutras begin with marriage, given that it is the gateway to the ritual life of a Brāhmaṇa. The reason why authors of the Dharmaśāstras departed from this tradition is instructive. The theology here is that a person comes under the regimen of dharma only after he has undergone Vedic initiation. This point is accentuated when this same rite is viewed as the second birth of the initiate, thus constituting him as a dvija , twice-born. Only twice-born individuals are capable of fulfiling the requirements of dharma .

Another significant literary and doctrinal feature of the early Dharmasūtras, as opposed to the texts of Manu and his successors, is that their authors do not pretend that the doctrines and rules they enunciate are anything more than scholarly statements. There is no preamble or story that presents the text as the pronoucement of a divine authoritative figure. Thus, the authors proffer divergent points of view indicative of scholarly give and take. We have already seen, for example, that Gautama and Baudhāyana give the āśrama system as a theory of some people which should be rejected. Likewise, Āpastamba departs from the common acceptance of polygamy and hypergamous marriage and supports monogamous marriages between partners belonging to the same social class. It is within this context of citing the opinions of others, especially of opponents, that we get a glimpse into the hidden history of the early Dharmaśāstric textual production.

The four authors whose works are extant cite or refer to seventeen other experts in the tradition. 8 It is instructive to read closely two passages of Āpastamba in which the opinions of various authors are given with regard to two topics: a person from whom one may accept food (1.19.2–1.19.15), and whether and when someone can take what belongs to another without becoming guilty of theft (1.28.1–1.28.5). On the first issue, six opinions are cited: (i) Kaṇva: anyone who gives because he wants to; (ii) Kautsa: a virtuous person; (iii) Vārṣyāyaṇi: anyone who gives; (iv) Eka, Kuṇika, Kāṇva, Kutsa, and Puṣkarasādi: almsfood is always pure and may be eaten; (v) Vārṣyāyaṇi: anyone who gives food without being requested; (vi) Hārīta: provides a rider to the last opinion that it should not be given subsequent to an invitation. On the second issue, we have three opinions: (i) Kautsa, Hārīta, Kaṇva, and Puṣkarasādi: someone who takes what belongs to another, no matter the circumstance, is a thief; (ii) Vārṣyāyaṇi: some articles, such as legume pods and fodder for an ox, are exempt from the above rule; (iii) Hārīta, apparently in reply to Vārṣyāyaṇi, says that one must always obtain the permission of the owner.

These opinions provide insight, slight though it may be, to the views and personalities of these authors. Vārṣyāyaṇi appears to hold somewhat liberal views on both issues, while Hārīta here and elsewhere expresses conservative opinions. With reference to the opinion (ĀpDh 1.18.2) that certain kinds of food may be accepted from a person belonging to the Ugra caste, Hārīta objects, saying that it is permissible only if the giver is a pupil. Elsewhere, with reference to a penance for having sex with an elder’s wife that ends in the penitent’s death (ĀpDh 1.18.15–1.18.16), Hārīta objects, saying that killing oneself or another person is always a heinous sin. And he rejects the opinion (ĀpDh 1.29.15–1.29.16) that sorcery and cursing do not cause a person to lose his caste, saying that such acts do cause the loss of caste.

What we see, then, is that the first three or four centuries of Dharmaśāstric history were characterized by vigorous debates among scholars regarding the rules of dharma . Matters were not settled, and writers during this period cite the opinions of other scholars, some of which go against their own views.

Innovations of Manu

The century or so before and after the turn of the millennium was a period of profound social and political turmoil and transformation in Northern India. The period witnessed repeated invasions from the northwest: first the Śakas (Scythians) in the first century bce and then the Kushanas (Yuezhi) in late first and early second century ce . These foreign polities established kingdoms within the Indian heartland. Even though they became Indianized to some degree, the insertion of foreign values and strategies must have created a cultural shock within India, especially within the Brahmanical intelligentsia, who considered them mleccha , barbaric foreigners. They could not be legitimate kings. Further, the Kushanas converted to Buddhism and became great patrons of Buddhist projects, especially monumental buildings. The Kushana period represents also the beginning of what Pollock (2006 ) has termed the “Sanskrit Cosmopolis,” a millennium or so when Sanskrit became the medium both of expressing aesthetic beauty in literature and of projecting political power.

We, of course, do not have direct evidence about the Brahmanical attitudes toward these newcomers and their rule; Brahmanical authors hardly ever comment on contemporary social or political realities. But I want to propose that the ideologies of kingship—Who is an ideal king?—present in both Manu and the Sanskrit epics are in some way related to contemporary political realities. When Manu says: na śūdrarājye nivaset (“He should not live in a kingdom ruled by a Śūdra.” MDh 4.61), I think, the subtext is the foreign rule of the Kushanas that extended to much of Northern India during the second century bce . Śūdra for Manu appears to be an epithet that could be hurled at anyone opposed to Brahmanical privilege. The term is used with regard to both rulers and Buddhist and other non-Brahmanical ascetic traditions. He states explicitly the reason why the ruling elite of foreign countries—Greeks, Śakas, Persians, and Chinese—have sunk to the level of Śūdras: “By neglecting rites and by failing to visit Brāhmaṇas, however, these men of Kṣatriya birth have gradually reached in the world the level of Śūdras” (MDh 10.43). The royal ideal is the very opposite: “Refusal to turn back in battle, protecting the subjects, and obedient service to Brāhmaṇas—for kings, these are the best means of securing happiness” (MDh 7.88).

This contemporary political and religious situation was the backdrop for the composition of Manu’s treatise on dharma . Composed probably in the middle of the second century ce , Manu represents a watershed in the history of Dharmaśāstra when the scholarly tradition of debate and disagreement of the previous centuries was abandoned, at least at the level of literary composition, and the authoritative voice of the author rises to drown out all dissent. 9 Stylistically, Manu composed his work entirely in verse, just like the Sanskrit epics. He also presents it within a narrative framework: the story of sages approaching Manu and asking him to teach them dharma . Manu tells them the genealogy of his treatise on dharma (1.102). It was originally composed by Manu’s father, the self-existent creator of the universe, and handed over to his son (1.58).

Manu places the Brāhmaṇa and Brahmanical privilege at the very heart of his treatise. In an almost “over-the-top” eulogy of the Brāhmaṇa in the very opening chapter, Manu harks back to the Puruṣasūkta of the Ṛg Veda (10.90) to assert the supremacy of the Brāhmaṇa:

Because he arose from the loftiest part of the body, because he is the eldest, and because he retains the Veda, the Brāhmaṇa is, according to dharma, the lord of this whole creation (1.93). A Brāhmaṇa’s birth alone represents the everlasting physical frame of dharma; for, born on account of dharma, he is fit for becoming Brahman. For when a Brāhmaṇa is born, a preeminent birth takes place on earth—a ruler of all creatures to guard the storehouse of dharmas. This whole world—whatever there is on earth—is the property of the Brāhmaṇa. Because of his eminence and high birth, the Brāhmaṇa has a clear right to this whole world. The Brāhmaṇa eats only what belongs to him, wears what belongs to him, and gives what belongs to him; it is by the kindness of the Brāhmaṇa that other people eat.

(1.98–1.101)

One can feel the intensity and urgency with which Manu defends Brahmanical exclusivism and, pari passu , the relegation of the Śūdra to the level of a servile and oppressed class. For Manu, I think, Śūdra was not simply a particular social group; it was a catch-all category for all groups that would present a threat to Brahmanical hegemony. These included Buddhist and other “heretical” religious orders, whom Manu dubs śūdrapravrajita (Śūdra ascetics), as well as kings and polities that did not toe the Brahmanical party line, whom Manu calls śūdrarājya (Śūdra kings and kingdoms).

A significant advance of Manu pertains to law and legal procedure. It is clear that Manu had before him a copy of Kauṭilya’s Arthaśāstra and incorporated much of the legal and procedural material from it. 10 These sections comprise Chapters 7–9 , amounting to one third of Manu’s text. His classification of the vyavahārapadas or subjects of litigation into eighteen remained paradigmatic in later Dharmaśāstric literature (Chapter 23 ).

Another significant subject Manu introduced into Dharmaśāstric discourse is mokṣa or liberation, a central idea of Indian religions that did not play a major role in early Dharmaśāstric history. He devotes the last chapter to this topic, even though it occurs frequently in the rest of the book as well. Significantly, he calls the life of a wandering mendicant mokṣāśrama , the āśrama devoted to liberation. This will remain an integral topic in later Dharmaśāstras and even in the legal digest of the medieval period.

Dharmaśāstras After Manu

A major political transformation took place in Northern India early in the fourth century ce , about two centuries after Manu, when Candragupta and his successors asserted dominion over much of Northern India from their hereditary base in eastern India. For the first time since the Mauryas another Indian empire emerged, but, unlike the Mauryas, the Guptas were deeply Hindu and they satisfied Manu’s ideal of a king devoted to Brāhmaṇas.

The Gupta empire is generally considered to be the golden age of Indian art, architecture, and literature. The great Sanskrit poet and playwrite Kālidāsa flourished during this period. It appears that literacy also made gains, given the prominent place given to legal documents and written contracts in the legal literature of the period. The same literature shows that jurisprudential scholarship developed exponentially; it is reflected in the detailed discussions of court procedures and the nuanced technical vocabulary. In the area of religion, we have the development of strongly devotional (bhakti ) movements reflected in both literature and art. It is within the context of these sociopolitical changes that we must locate the composition of Dharmaśāstras after Manu.

All the post-Manu writers of Dharmaśāstras were indebted to him in numerous ways; they followed his lead with regard to both literary style and content. Yet we see developments in a variety of areas of dharma , most especially in religious orientation and jurisprudence. Although, in all likelihood, there were dozens of such writers, there are only four whose works have survived: 11 Yājñavalkya, Viṣṇu, Nārada, and Parāśara.

Of these, Yājñavalkya was the most influential writer after Manu in terms of his effect on the later tradition. The text was composed probably in the fourth or fifth century ce in Eastern India during the rule of the Guptas. The ascription of the work to Yājñavalkya, the celebrated theologian of the Bṛhadāraṇyaka Upaniṣad, the founder of the school of White Yajurveda, and close associate of Janaka, the renowned king of Videha in Eastern India, makes it likely that the work was comissioned at least in part to support the legitimacy of the Gupta emperors. My critical edition of the text (Olivelle Forthcoming) shows that the recension commented on by the ninth-century scholar Viśvarūpa is far closer to the original than that of the better known twenth-century commentator Vijñāneśvara. It is clear that sometime in the tenth or eleventh century a drastically emended edition of the text was made that, through Vijñāneśvara’s popular commentary, became the Vulgate version in medieval and modern times.

To a degree even more than Manu, Yājñavalkya incorporates much of the legal and procedural rules of Kauṭilya’s Arthaśāstra. Indeed, some of Yājñavalkya’s verses look much like versifications of the Arthaśāstra prose. His presentation of legal procedure, however, shows clear advances over that of Manu and Kauṭilya, pointing to a vibrant jurisprudential scholarship. For example, Yājñavalkya places emphasis on documents in his discussion of evidence; he is the first to use the technical terms lekhya and likhita with reference to legal documents. He is also the first to use the technical term divya for an ordeal, making it a significant part of legal evidence. For the first time we see in his work the enunciaton of a hierarchy of courts with the possibility of appeals from the lower to the upper ones and ultimately to the king himself.

Another significant feature of Yājñavalkya is his focus on asceticism, yoga, and the search for liberation. Indeed, the very first word of the text is the epithet of Yājñavalkya: yogīśvara , the Lord of yoga, and throughout the text he is presented as a teacher of yoga and asceticism. He moves the discussion of the two ascetic āśramas , the forest hermit and the wandering mendicant, from the section of proper conduct (ācāra ) to that on penance called prāyaścitta , which, I think, is viewed by him not simply as practices to expiate sins but more generally as extraordianry acts of penance and self-mortification. The section (prakaraṇa ) on the ascetic (2.56–2.206) comprises 151 verses, by far the longest section comprising 15 percent of the entire text.

The text of Viṣṇu is one of the latest Dharmaśāstras, composed in between the sixth and eighth century ce in Kashmir by a person belonging to the Kāṭhaka branch of the Black Yajurveda who was a devotee of Viṣṇu (see Olivelle 2009a ). Viṣṇu is presented as the person teaching the Dharmaśāstra to the goddess Earth. His authority alone guarantees the validity of the document, and this is the only Dharmaśāstra that does not present the traditional sources of dharma .

The contents of the text are unremarkable, except for the strongly devotional bent that stands in sharp contrast to all other Dharmaśāstras. The new institution that it introduces is the wife immolating herself besides her deceased husband, referring to this practice as anvārohaṇa at 25.14 and suggesting the term anugamana at 20.39.

During and after the fifth century ce , some Dharmaśāstric scholars appear to have engaged in writing texts focused on specific topics rather than the entire range of Dharmaśāstra. The two extants texts of this genre are those of Nārada and Parāśara. Nārada focuses on legal procedure (vyavahāra ) and Parāśara on proper conduct and expiation.

Tradition explicitly presents Nārada’s text as a recension of Manu, indicating its close connection to the latter. Nārada, however, far surpasses Manu, as also Yājñavalkya, in his jurisprudence; his is perhaps the most refined legal text from ancient India assording to A. Barth’s assessment: “If we except the monuments of Roman legislation, antiquity has not perhaps left us anything which is so strictly juridical” (Lingat 1973 : 102).

Nārada is probably later than Yājñavalkya and can be dated to between the fifth and sixth century ce .

Parāśara’s is a very brief and somewhat mediocre text whose existence in the mansucript tradition is probably due to the great commentary on it, the Pārāśara-Mādhavīya , written by the fourteenth century scholar Mādhava. In 592 verses, Parāśara deals with issues relating to proper conduct (ācāra ) and penance (prāyaścitta ). It was probably composed between the seventh and eighth century ce .

Extinction of Dharmaśāsstric Texts

The age of Dharmaśāstric composition came to an end probably by the middle of the second half of the first millennium, even though texts calling themselves smṛti continued to be composed well into the medieval period. A statement by the great commentator on Manu, Medhātithi, shows that scholars accepted the possibility that even their contemporaries had the authority to write Dharmaśāstras. After stating that Manu was not some exceptional and divine being but simply an entrepreneurial scholar who gathered specialists and composed his treatise, he goes on to say, “Even in contemporary times, when a person endowed with the aforementioned qualities and with those very reasons composes a treatise, he becomes authoritative for future generations just as Manu and the like” (on MDh 2.6). And it appears that many scholars did undertake such projects; many of their texts are cited in medieval legal digests.

Manu generally does not cite or refer to his predecessors, but at MDh 3.16 he refers to Atri, Utathya, Śaunaka, and Bhṛgu. Yājñavalkya (1.4–1.5) is the earliest writer to give a list of authors of Dharmaśāstras: “The promulgators of legal treatises are: Manu, Viṣṇu, Yama, Aṅgiras, Vasiṣṭha, Dakṣa, Saṃvarta, Śātātapa, Parāśara, Āpastamba, Uśanas, Vyāsa, Kātyāyana, Bṛhaspati, Gautama, Śaṅkha, Likhita, Hārīta, Atri, as well as myself.” Of these, only the compositions of Manu, Viṣṇu, Vasiṣṭha, Parāśara, Āpastamba, Gautama, and Yājñavalkya are extant. Kane (1962–75, I: 304) estimates that approximately 100 Dharmaśāstras are cited in medieval legal digests. Thus, the conclusion we have to draw is that the vast majority, perhaps as much as 90 percent of the Dharmaśāstras have been lost in the manuscript tradition. Even if we question whether all the citations in the legal digests are from actual treatises and whether some may have been floating verses in the memory of experts, yet it is clear that a large number of these texts have simply disappeared.

The reasons for this large-scale extinction are unclear. The ultimate cause, of course, is that fresh copies of the manuscripts of these works were not made because no one thought it important enough to spend time and money to do so. Given that manuscripts in the Indian climate last but a few centuries, if fresh copies are not made they will gradually become prey to decay and bugs. But why scholars thought it not important enough to copy them remains obscure. After all there were some Dharmaśāstras, namely the extant ones, that were copied and recopied. The voluminous commentaries and digests that presented topically arranged citations from the ancient texts may themselves have made experts and students alike less dependent on the original texts. There are indications in the digests themselves that later authors are citing not from the originals but from citations in earlier digests. If we knew more about the education system in medieval India, about how young students were taught the Dharmaśāstras, we would probably have a better idea about the reasons of this extinction. Did students, for example, simply study one or several legal digests rather than original Dharmaśāstras? Or, if they studied the original texts, did the curriculum include only a few major ones, such as that of Manu and Yājñavalkya?

A few lost Dharmaśāstras have been reconstituted by collecting medieval citations. The two most prominent ones are the texts of Bṛhaspati and Kātyāyana, 12 both great jurists dealing with legal procedure. Others still await close scrutiny by scholars.

The history of Dharmaśāstra from the middle of the first millennium takes a new turn. In place of original compositions, scholars began to write first commentaries on the major ancient Dharmaśāstras and then, from at least about the twelfth century ce , legal digests (nibandha ). This period of the history is taken up in the second chapter.

1 For detailed studies of the early history of the term, see Brereton 2009 and Horsch 2009 .
2 For an assessment of these developments, see Olivelle 1993 : 58–70.
3 For a detailed study of this topic, see Olivelle 2016a .
4 The Dharmasūtra of Hiraṇyakeśin also comes from a caraṇa , but this text is simply a slightly altered version of the text of Āpastamba.
5 For an extended discussion of both Kātyāyana and Patañjali, see Olivelle 2010b, 34–6.
6 One such criterion is the concept of āryāvarta , land of Āryas, first articulated by Patañjali and unknown to Āpastamba and Gautama but incorporated by Baudhāyana and Vasiṣṭha.
7 The first two Praśnas. The second two Praśnas were probably a late addition. See Olivelle 2000 : 191; Bühler 1879 –82, II: pp. xxxiii–v).
8 Aupajaṅghani (BDh 2.3.33); Bhāllavins (BDh 1.2.11; VaDh 1.14); Eka (ĀpDh 1.19.7); Hārīta (ĀpDh 1.13.11; 1.18.2; 1.19.12; 1.28.1, 5, 16; 1.29.12, 16; BDh 2.2.21; VaDh 2.6); Kaṇva (ĀpDh 1.19.3; 1.28.1); Kāṇva (ĀpDh 1.19.2, 7); Kapila (BDh 2.11.28); Kaśyapa (BDh 1.21.2); Kātya (BDh 1.3.46); Kautsa (ĀpDh 1.19.4; 1.28.1); Kuṇika (ĀpDh 1.19.7); Kutsa (ĀpDh 1.19.7); Mahājajñu (BDh 3.9.21); Manu (ĀpDh 2.16.1; GDh 21.7; 23.28; BDh 2.3.2; 4.1.13; 4.2.15; VaDh 1.17; 3.2; 11.23; 12.16; 13.16; 19.37; 23.43); Maudgalya (BDh 2.4.8); Puṣkarasādi (ĀpDh 1.19.7; 1.28.1); Vārṣyāyaṇi (ĀpDh 1.19.5, 8; 1.28.2). It is not clear whether all these actually wrote Dharmasūtras or were merely viewed as authoritative teachers.
9 For a detailed assessment of Manu’s treatise (its author, date, composition, sources, and the like), see the introduction to my critical edition: Olivelle 2005a .
10 For more details and fuller argument, see McClish 2014 , Olivelle 2004b .
11 I ignore here the Vaikhānasa Dharmasūtra , which was a sectarian text and had no impact on the later Dharmaśāstric tradition.
12 Bṛhaspati has been reconstructed by Aiyangar (1941a) and Kātyāyana by Kane (1933 ).