The term impurity appears so simple and unproblematic. Something or someone has become dirty and needs cleaning. But even in modern Western languages impurity is a loaded term with a multiplicity of meanings, ranging from the purely (note the word) material to the social, ritual, moral, and mental. The bodily experience of becoming dirty, and getting clean is one of those basic sources of metaphor, without which we have no language.
Nevertheless, it must be admitted that the multiplicity, refinement, and significance of this particular metaphor is extraordinary in the Hindu legal texts, the Dharmaśāstra. The object of this chapter is therefore to draw out some general structures from this multiplicity and outline the specific practical contexts that generated this kind of language.
The Sanskrit vocabulary of the impurity–purification–purity complex is vast. Different words are applied in different contexts, but they are not always limited to one. These are just three examples of words often simply translated as “impure”: ucchiṣṭa (“with leftovers”) typically occurs in the contexts of leftovers from a meal and the need to wash hands and mouth after eating; amedhya (“unfit for sacrifice”) often occurs in connection with bodily impurities or decomposition, such as corpses, feces, and urine; and malina (“stained”) refers to physical dirt (mala ), especially to exudations from the body, in which case it overlaps with amedhya . 1 The systematization of the vocabulary was a gradual process, just as it was within other central subjects of this literature. Early texts contain a number of rules targeted at different contexts and gradually through the centuries these different sets of rules became systematized, each with its own theoretical metalanguage.
As occasions for impurity differ, so do the methods of purification. Impure substances can be washed or rubbed away depending on the material, but often ritual means are also needed for certain types of pollution that cannot be removed by physical methods alone. Impurity in the Dharmaśāstra is clearly not only a matter of physical dirt. In a very general sense, purity in this context must be understood as ritual purity, that is, as a prerequisite for performing successful ritual acts. This is spelled out in a definition of purification (śuddhi ) from a late text, that of Govindānanda’s sixteenth-century treatise on the subject, Śuddhikaumudī . It goes: “To be purified is to be worthy of performing the rituals known in the Vedas.” 2
We may use a parallel from speech act theory. Certain verbal phrases such as a marriage oath or the passing of a sentence in a court of law can have the character of binding actions, but only on the condition that the person who utters the phrase has the acknowledged authority to do so. John L. Austin formulated the rule that “the particular persons and circumstances in a given case must be appropriate for the invocation of the particular procedure invoked” as one of several felicity conditions of the successful speech act (Austin 1975 : 15). Likewise, “purity” (in its many forms) can be understood as a felicity condition of the successful ritual action.
However, if purity must be understood as ritual purity, we also need to know more precisely what is meant by ritual in the context of Dharmaśāstra, and especially how far the notion of ritual was applicable and relevant in the social life of Brahmins. Here, it is important to be aware of the historical contexts that motivated the priestly class (Brāhmaṇas) to produce this literature when it emerged in the post-Vedic period.
It has been argued that the earliest Dharmaśāstra texts, the Dharmasūtras, were composed during the Mauryan rule (322–185 bce ) as an attempt by the priestly class to reconsolidate its close economic and political ties with rulers (Olivelle 2004a ; 2005b ). In the pre-Buddhist Vedic corpus, the word dharma is not very frequent, but it does occur, mostly in relation to royal rituals, such as the royal consecration (rājasūya ) and the horse sacrifice (aśvamedha ). Early Buddhism appropriated the term, together with its associations to kingship, and in the process added new connotations to the word that became a shorthand term for the Buddhist teachings (Olivelle 2004a : 150–1). These connotations were centered on the moral and religious authority that had its roots in the monastic, celibate life of the monks. Through the charisma emanating from the extraordinary lifestyle of asceticism and celibacy, these religious leaders were seen as powerful means of religious merit, blessings, and safe advice by its new influential adherents among merchants and rulers.
Particularly as a result of King Aśoka’s use of Buddhist ethics for his imperial ideology (see Olivelle 2005f : 127–9), Brahmins felt challenged but were unable to compete with the celibate and ascetic lifestyle, since family life was crucial to the priestly kinship-based ethos. 3 The idea of the Buddhist pabbajjā —leaving the home to become homeless—was basically against these values. Here I will propose the hypothesis that the Brahmin authors of the Dharmasūtras stressed the preoccupation with ritual purity as an essential priestly competence, in an attempt to counter the ascetic charisma of celibate Buddhist monks. In order to make this relevant for their own lifestyle, which was rooted in the family, they gradually extended the notion of ritual to include all the activities in the normal life of a family. In addition to lifecycle rituals (saṃskāras ), like the initiation of boys, marriage, and cremation and ancestral rites known from the domestic ritual manuals (Gṛhyasūtras), daily activities like eating, bathing, and sex, became ritualized and thereby also subject to purity rules, 4 allegedly on the basis of existing “customs” (ācāra ) among an elite of learned Brahmins who had mastered the Vedic knowledge (Olivelle 2005f : 133–4).
Under the influence of the Vedic theology of the mīmāṃsā school of thought, these rules of daily behavior were seen by their Brahmin authors as a continuation of Vedic injunctions (vidhi ). The nature of these injunctions was such that to follow them would ensure salvific or beneficial results, while to transgress against them would be demeritorious. The Brahmin authors of Dharmaśāstra texts accordingly reformulated the notion of dharma to include the extended rules for family life. To follow these rules would be an act of dharma , and as such, meritorious like Vedic vidhis , whereas to go against or neglect them would be adharma (unrighteousness) and would therefore produce bad karmic results. 5 In short, this re-appropriation and reformulation of dharma achieved both the inclusion of Brahmin family values in an inclusive, but compartmentalized social ideology and also the maintenance of the semantic associations with kingship, the latter emphasized by making the duties of kings (rājadharma ) a standard subject of Dharmaśāstra. 6
From early on, purification was divided into two broad types, external and internal. According to Baudhāyana:
The Veda points out two types of purification which are practised by cultured people: external purification is the elimination of stains and smell, whereas internal purification is the abstention from injuring living beings. The body is purified by water, and the intellect by knowledge; the inner self is purified by abstaining from injuring living beings, and the mind by truth.
(BDh 3.1.26–3.1.27 in Olivelle 2000 : 305).
Here dirt and ill-smelling bodily products like sweat, urine, and feces are considered outer impurities, and the impulse to violence is considered an inner impurity.
Outer purification was further divided into three: purification related to one’s family (purifications after birth and death), purification of one’s property (tools, cloths, and articles of various materials), and purification of one’s body (the “elimination of stains and smell”). Inner purification was likewise divided, but into five categories: purification of mind and speech, as well as purification of eyes (vision), nose (smell), and tongue (taste). 7
It may be helpful to correlate these different categories of impurity with specific purification practices. Impurities of one’s body (śarīra ) and property (artha ) are removed according to rules of cleanliness (śauca ); the personal purifications necessary after a birth or death in the family (kula ) are regulated according to the rules of death impurity (āśauca ); and the inner (āntara ) impurities related to mind, speech, and eyes are purified by various (mostly minor) penances (prāyaścitta ) like mantra recitation, speaking to a Brahmin and looking on the sun for a moment respectively. I analyze below each of these occasions for impurity and purification.
Apart from the considerations of personal comfort and social interaction, a dirty and unclean body could never be accepted as a proper agent of ritual acts. Based on a large selection of texts from various periods of the development of Dharmaśāstra, Kane (1962–75, vol. 2: 648–68) offers a general description of the various elements of bodily cleanliness that form part of the daily routine (āhnika ) of a Brahmin. Having got up from sleep before sunrise, he has to go through a schedule of four items in preparation of his daily morning worship: washing after answering the call of nature, sipping water (ācamana ), brushing the teeth, and bathing (snāna ).
However, early texts do not waste words on repeating the same procedures performed at different times of the day, but collect them together as one. Āpastamba summarizes:
When he has washed away the stains of urine or excrement after going to the toilet, the stains of food, the stains from eating, the stains of semen, and then washed his feet and sipped water, he becomes pure.
(ĀpDh 1.15.23 in Olivelle 2000 : 51).
Baudhāyana is a bit more detailed:
After voiding urine, he should first wash the organ with earth and water and then the left hand three times. He should do the same after voiding excrement, but three times in turn for the anus and the left hand. After a seminal discharge, he should do the same as after voiding urine. (BDh 1.10.11–1.10.15 in Olivelle 2000 : 217).
Sipping water, which is done regularly in the morning, but which is also required during the day after having food, is prepared for by washing the hands (and feet according to some texts):
Seated on a clean spot, placing his right arm between his knees, and wearing his upper garment over his left shoulder and under his right arm, he should wash both his hands up to the wrists. Then he should silently sip three or four times an amount of water sufficient to reach his heart, wipe his lips twice, sprinkle water on his feet, rub water on the cavities of his head [eyes, ears, and nostrils], and place his hand on the crown of his head. After sleeping, eating, and sneezing, he should sip water over again. (GDh 1.36–1.37 in Olivelle 2000 : 123).
The instructions for the obligatory daily bath became increasingly complex, particularly in adding a large repertoire of mantra recitations that should accompany the bath, which is taken as repeated dips in cold water, preferably in a river, having first smeared the body with loose earth for the first dip and with cow dung for a second dip (Kane 1962–75, vol. 2: 660–3; Glucklich 1994 : 71–5).
Manu (5.135) lists twelve bodily impurities, the stains of which should be removed: “Body oil, semen, blood, marrow [or perhaps ‘oily residue on the skin after sweating’], urine, faeces, ear-wax, nails, phlegm, tears, discharge of the eyes, and sweat” (Olivelle 2005a : 145 and 287). Note that saliva is missing from this list. This may seem surprising seen from the perspective of social anthropological studies of South Asian food culture. Ravindra Khare talks of the “exclusivity of the saliva” as the central principle regulating Indian commensal systems (Khare 1976b : 8). The idea is that food left over from someone’s meal cannot be shared with others, except in specific circumstances governed by particular rules. In general, such food can only be shared with socially inferior persons (Malamoud 1972 : 9). But outside the food context, saliva does not cause impurity. On the contrary, Parāśara declared, “drops of saliva which pass between one person and another, as well as leavings of oil which remain after the meal, these are not impure.” The commentator, Mādhavācārya, explains the drops of saliva as “drops which come from the mouths of people during conversation and which fall on the body” (Par /PāM 7.32 in Tarkālaṅkāra 1883–99, vol. 2: 142). 8
That such drops of saliva are not seen as demanding purification is part of a larger set of rules that limit the occasions for impurity in contexts where such limits are too unpractical. Some specific rules:
The hand of the artisan is always clean, as also goods displayed for sale.
(BDh 1.9.1 in Olivelle 2000 : 215)
No taint is created by the hair of the beard getting into the mouth. Bits of food sticking between the teeth are like the teeth themselves, as also whatever is in the mouth and what remains after sipping. One becomes purified by simply swallowing them. Even when someone pours water for others to sip and drops splash on his feet, they do not make him unclean; they are said to be the same as the ground (VaDh 3.40–3.42 in Olivelle 2000 : 367).
A woman’s mouth is always pure; 9 so is a bird when it makes a fruit to fall, a calf when it makes the milk to flow, and a dog when it catches a deer.
(MDh 5.130 in Olivelle 2005a : 145).
These exemptions seem rooted in pragmatic considerations, but the fact that they need to be mentioned explicitly also emphasizes the general preoccupation with bodily purification.
The mentioning of “the hand of the artisan” and “goods displayed for sale” is part of a larger theme on purification of “things” (dravya ) and “property” (artha ). This theme involves general cleaning methods for various types of material, but more specifically, it addresses things that have become impure by being handled by persons whose touch generally should be avoided. The general cleaning methods are rather straightforward. Various metal objects made of gold or silver, as well as jewels, are washed with water, but combined with ash and earth if they are stained; objects made of copper and cheaper metals may need water with acid or alkali (MDh 5.111–5.114 in Olivelle 2005a : 144). Similar practical methods are applied for other materials, like bone, ivory, clothes, silk, wool, skin, grain, vegetables, fruit, and plots of land.
The rules regarding objects that have been in physical contact with people whose touch is generally avoided are more complex. The prototype of such people in early texts is the Caṇḍāla , an untouchable caste whose living places were segregated in the outskirts of villages and cities. 10 In the Dharmasūtras it is taken for granted that public spaces, such as roads, boats, grass, seats, and couches, are polluted by the physical presence of Caṇḍālas , but again pragmatic considerations exempt these occasions of causing impurity by the idea that these spaces are naturally and constantly purified by the wind (BDh 1.9.7 in Olivelle 2000 : 215).
Things get more complicated in later texts. Parāśara seems to add time as a parameter. If people stay at these spaces at the same time as a Caṇḍāla , they will need some sort of purification: a fast for three days if they have slept on the same couch, and muttering the gāyatrī prayer if they have walked on the same road (Par 6.23 in Tarkālaṅkāra 1883–99, vol. 2: 82). The same text also makes a distinction between, on the one hand, public spaces and goods that are for sale, and on the other, personal belongings, including goods that have already been purchased. The latter is much more serious than the former, and Par addresses that situation in a long section on the purification of a house, its inmates, and articles, in the event that a Caṇḍāla has stayed there. I shall return to that case later in this chapter.
However, everything related to the Caṇḍāla is a special case. The purifications after contact with such a person, whether through direct touch or indirectly through spaces or things, do not really belong to the class that has been dealt with in this section, that is, purification (śauca ) of stains on the body and on things. This is implied by the fact that rules about purification after direct contact with a Caṇḍāla are placed together with rules about death impurity. This, I think, is not accidental.
Death impurity (āśauca ), as we saw, belongs to the category of outer impurity related to the family (kula ) in Harīta’s typology. Death in the family renders the relatives impure for a certain amount of time, according to all four Dharmasūtras. However, the length of the period depends on the age (less for small children), marriage status (less for unmarried women), and family remoteness of the deceased relative (less for remote relatives), as well as on the social class (varṇa ) of the family in which the death has occurred (compare the four Dharmasūtras in Olivelle 2005f : 142–5).The general rule that was preserved in later texts is: if the deceased belonged to the common ancestry (sapiṇḍa ), the period is ten days for a Brāhmaṇa, twelve days for a Kṣatriya, fifteen days for a Vaiśya, and one month for a Śūdra (MDh 5.83).
During this period, the surviving relatives were subject to a series of taboos. The Dharmasūtras mention remaining passive, fasting, extra bathing, visible marks of mourning like shaving or throwing dust on the hair, as well as the inability to distribute food to others. Later, the commentator on GDh , Haradatta (twelfth–fourteenth century), defined āśauca according to four specific taboos: people who are subject to death impurity cannot participate in rituals (other than those related to the dead person) or serve food to others or distribute gifts, and they are untouchable to others (Har-G 14.1 in Pandeya 1969b 141 as 2.5.1). Clearly, these taboos are meant to isolate the mourners and protect their social surroundings from contact with them, and ultimately with death impurity.
The unequal duration of āśauca for the four classes is not necessarily an expression of any inherent degree of impurity, presumably lower for Brāhmaṇas and higher for Śūdras. Most often it is explained in terms of the vital necessity of the sacrifice and, with that, the need for Brahmins as the priestly class to be able to get back to business as soon as possible. The sacrifices performed by the Brahmins are for the good of the other three classes, indeed for the good of the whole world. Therefore, the Brahmins’ greater duties and entitlement (adhikāra ) in relation to the sacrifice and the gradually lesser degrees of the same for the three lower classes are likely the reason why the Brahmin authors of this literature settled on durations of death impurity that would release themselves from this impurity relatively faster than the other classes would be.
In principle, the same rules apply when a child is born, but here the texts express divergent views. The main rule is that āśauca should be observed after birth as it is after death, that is, for the same period as prescribed when someone has died. The alternative rule is that only the parents need to observe āśauca after birth, not the rest of the close family. And the second alternative is that only the mother must observe the full period of birth impurity, while the father is purified merely by bathing (GDh 14.14–14.15; MDh 5.61–5.62).
The impurity caused by birth is most obviously explained by the rule, mentioned earlier, that makes blood an impure substance (MDh 5.135). This would also explain the second alternative rule, whereby only the mother needs to observe the full duration of āśauca . But it may also be an expression of the perils of childbirth for the mother.
Another alternative rule, found in some later texts, specifically reduces the duration of the taboo of untouchability, which, as we saw, was regarded as one of four taboos connected with death impurity, to the number of days after which the bones could be collected from the cremation ground. This again depends on the class of the family, four days for Brāhmaṇas, six days for Kṣatriyas, and eight days for Vaiśyas and Śūdras (e.g. SS 38–40b in Āpṭe 1905 : 413; Medh 5.60 in Jhā 1920–39, reprint 1999, vol. 1: 449). 11 The principle here seems almost universal: The strongest degree of pollution coincides with the time it takes for the corpse to reach a state where there can be no more decay. The same principle seems to be behind the rule that a bath is needed for purification after having touched a human bone that was “greasy,” but only “sipping water, touching a cow, or gazing at the sun, if the bone was dry” (MDh 5.87 in Olivelle 2005a : 142).
Grouped together with the rules of āśauca , we find rules for purification in case one has been in physical contact with persons whose mere touch is polluting. This context is not surprising, since a corpse, someone who has touched a corpse (and has not yet undergone purification), and a woman who has just given birth are all counted among these persons. In addition to these three, the older lists include a woman during menstruation, 12 a person who has committed a grievous sin (like murder or theft), and a Caṇḍāla , the untouchable caste (GDh 14:30; MDh 5.85).
It should be noted that the latter, the Caṇḍāla , is the only person in these older lists who is permanently untouchable and, therefore, permanently impure. 13 Impurity is rarely a static state (Olivelle 2005e : 240). Occasions for impurity are normally occasions for purification. The exception is with regard to persons who are impure because of their prescribed caste-bound occupations, such as for the Caṇḍāla : carrying away corpses of persons without relatives, executing persons sentenced to death, and hunting (MDh 10.49, 10.55–10.56; VaikhDh 16.11). The connection with death impurity is obvious.
With growing social complexity and interaction, later texts added other occupational groups to these lists, such as washermen (handling cloths stained with menstruation blood) and leather workers (handling skin of dead animals). 14 Also, cripples, fools, temple priests, as well as village animals like fowl and pigs, are included (PāM 6.24).
A parallel proliferation of the occasions for impurity can be witnessed. Just to give one example: Parāśara (Par 7.11–7.15) deals with purifications for menstruating women of different class (varṇa ) who happen to touch each other. The commentator, Mādhavācārya, quotes other texts that account for the following situations: a menstruating woman who is touched by a dog while she is eating (eating always aggravates an occasion for impurity); mutual touch between two menstruating Brāhmaṇa and Śūdra women who are at the same time impure after their meal (i.e., who have not yet sipped water for purification); a menstruating woman who touches a Brāhmaṇa man who is still impure after his meal; a menstruating woman who touches a corpse or a woman who has just given birth; and, finally, the latter case aggravated by the fact that the menstruating woman was having a meal while the incident happened (PāM 7.11–7.15 in Tarkālaṅkāra 1883–99, vol. 2: 124–5). Such details may seem absurd. But I think they should be seen not only as a natural accumulation within a scholastic literary tradition through the centuries (remember that Mādhavācārya’s commentary was written in the middle of the fourteenth century), but also as a response to growing social complexity. This is also indicated by the following example.
In the eighth century, when the Par might have been composed (Olivelle 2010b: 48), untouchable castes had become a necessary labor force within village and city life, and along with this increased interaction, boundary markings had become narrower and more detailed. It is, therefore, not a far-fetched situation that is envisaged by the text when it gives detailed rules about how to purify a home in the event that a Caṇḍāla has stayed there unnoticed. The purifications include three days of fasting on barley grain cooked in cow urine (gomūtrayāvaka ) with milk, curd, and ghee for the inhabitants of the house, and the cleaning of the utensils of the house using the specific method: rubbing with ashes for metal and washing for cloths. Earthenware has to be thrown away. Foodstuff and other inflammable goods must be set aside at the door, while a fire is lit on the floor in order to purify walls, floor, and ceiling. Added to this, the floor needs to be dug anew, the walls need to be plastered, and a homa (offerings in the household fire) should be performed with sacred mantras. Finally, Brahmins should be invited to stay in the purified house and receive food and a donation of cows and a bull (Par 6.34–6.42 in Tarkālaṇkāra 1883–99, vol. 2: 88–91).
The pollution of the soil that is implied in this example is further discussed by Mādhavācārya with a quote from Devala:
The soil where a woman has given birth or a man has died or been cremated, which is inhabited by caṇḍālas or where faeces and urine have been passed—that soil which is full of foul things like these is declared to be ‘impure’ [amedhya ]. When it is touched by animals like dogs, pigs, donkeys and camels, it becomes “defiled” [duṣṭa ]. It becomes “dirty” [malina ] through charcoal, husk, hair, bones, ashes and so forth. The soil which is “impure” is purified in five or four ways. 15 The soil which is “defiled” is purified in three or two ways, whereas that which is “dirty” is purified in one way. (PāM 7.35 in Tarkālaṅkāra 1883–99, vol. 2: 147) 16
Impurity is here divided according to degree, and the necessary purifications are likewise divided in degrees from more to less. The worst is associated with death, birth, and Caṇḍālas , which, as we have seen, are all connected with death impurity. But here, excrements and urine belong to the same filthy category. According to Vij (1.191 and 2.214), this amedhya type of impurity is explained as “dirt originating from the body,” which only becomes impure when coming out of the body (Olivelle 2005e : 237–8)—like other bodily impurities from the list of MDh 5.135. But for those elements that are related to death impurity, it is characteristic that the purifications must involve ritually active ingredients or actions like fire, cow dung, homa , and mantras, as in the case of the house, whereas the milder degrees of impurity can be overcome by simple cleaning.
Internal impurity was related to violent inclinations according to BDh 3.1.26–3.1.27 and further divided in impurity of mind, speech, eye, nose, and tongue by Harīta (Kane 1962–75, vol. 2: 651). I will present just a few examples. MDh 12.3 divides action into three categories: mental, verbal, and bodily. Internal impurity is associated with the first two of these categories. The text specifies:
Coveting the property of others, reflecting on undesirable things in one’s mind, and adhering to false doctrines are the three kinds of [impure] mental action. Harshness, falsehood, slander of every sort, and idle chatter are the four kinds of [impure] verbal action.…On account of faults resulting from verbal actions, he becomes a bird or an animal; and on account of faults resulting from mental actions, he becomes a man of the lowest caste [in a coming life].
(MDh 12.5–12.6, 12.9c–d in Olivelle 2005a : 230)
Smelling liquor or other things that should not be smelled is a sin causing some sort of exclusion within the caste community, and it must be expiated by a penance—a major penance (sāṃtapana ) if it was done deliberately, and a minor but still quite severe penance (prājāpatya ), if done unintentionally (MDh 11.68, 11.125). Telling a lie makes a man unworthy of receiving gifts, and it must be expiated through another major penance (cāndrāyaṇa or lunar penance) (MDh 11.70, 11.126). 17 Similarly, there are minor penances for talking to or looking at a Caṇḍāla : speaking with a Brāhmaṇa for talking with him, glancing at the “heavenly lights” for looking at him (ĀpDh 2.2.8–2.2.9 in Olivelle 2000 : 77).
The impurities of the tongue refer to the rules about unfit and forbidden food (see Chapter 14 ). This is a major subject. 18 Impure food is divided into foodstuffs that are inherently impure and cannot be eaten (abhakṣya ) and food that is impure only owing to the influence of wrong treatment, dirt, or the person from whom it originates and should not be eaten (abhojya ) for such reasons. 19
I shall give only one example: If a Brāhmaṇa swallows something impure (amedhya ), such as semen, or if he eats beef or the food of a Caṇḍāla , he should observe the lunar penance (cāndrāyaṇa ) lasting one month. For the same transgression, a Kṣatriya or a Vaiśya must observe half of that penance, whereas a Śūdra has only to perform a Prajāpati Penance lasting twelve days. In addition, each has to offer a donation (dakṣiṇā ) to the council of Brahmins that imposed these penances: the Brāhmaṇa must give one pair of cows; the Kṣatriya two; the Vaiśya three; and the Śūdra a donation of four pairs of cows (Par 11.1–11.3 in Tarkālaṅkāra 1883–99, vol. 2: 294, 301).
To swallow semen is forbidden and thus abhakṣya ; likewise to eat beef according to Parāśara’s text, which has a long chapter (Chapter 8 ) on penances for killing a cow. Food prepared by a Caṇḍāla is unfit for eating and thus is a common example of unfit food (ahojyānna ). However, the commentator, Mādhavācārya, connects beef and Caṇḍāla food, suggesting that eating these items may occur if a man is kept as a prisoner where Caṇḍālas are employed as prison guards. He also explains what else should be understood by the phrase “impure food.” His example is food contaminated by feces or urine. This, he says, no one eats intentionally, but it may happen “since it appears that eating together with one’s small children has become quite frequent among ordinary people.” And swallowing semen, he points out, happens in the situation of homosexual oral sex, or as he phrases it, it happens to men who are “afflicted with the disease of using the throat as vagina” (PāM 11.1 in Tarkālaṅkāra 1883–99, vol. 2: 294). 20
Purity systems became a major topic within social anthropology during the 1960s and 1970s. The British anthropologists Edmund Leach and Mary Douglas, both inspired by the ideas of Émile Durkheim and Marcel Mauss on classification systems being rooted in social structures, did parallel studies of the notions of taboo, which they saw as a product of the cognitive ordering of experience: elements that somehow overlap between different cognitive domains, and for that reason are cognitively ambiguous, are marked by taboos. The corpse in its transition from what looks like a living body to a skeleton is a clear example. Mary Douglas, in her 1966 book Purity and Danger , expanded these ideas, seeing rules concerning bodily boundaries and their protection, such as rules about food and sex, as a parallel to or an expression of the existence of social boundaries and the concern to protect them. As discussed at the start of this chapter, much of the Dharmaśāstra material provides a case for the idea that the priestly emphasis on purification rules served to strengthen group boundaries, both vis-à-vis Buddhists and other heretics and internally toward Śūdras (the three “twice-born” versus the “once-born” Śūdra), as well as toward the two lower classes (Brāhmaṇa versus Kṣatriya and Vaiśya).
Primarily through the works of Louis Dumont, other anthropologists became aware of the rich material from the Sanskrit literature, and using the English translations of Dharmaśāstra works that existed then (Bühler, Jolly, Gharpure, and Dutt), they offered new interpretations in a theoretical perspective hitherto unknown (or unapplied) by most indological scholars.
The inspiration from Dumont, however, had the result that most of these contributions saw degrees of purity and impurity as static indices of hierarchical social status, which, as has been argued in this chapter, is mostly a misreading of the texts (Olivelle 2005e ). Some of these contributions will therefore be mentioned in Chapter 4 of this volume, on social classes. However, a few are of relevance also to the dynamic aspects of impurity and purification that have been presented here.
In an attempt to solve some of the problems inherent in Dumont’s theory that the caste system is based on an ideology of relative purity, in particular the problem of accounting for the status of Kṣatriyas over that of Vaiśyas, different scholars (inspired by the works of M. N. Srinivas) developed an alternative theory, according to which the pure-impure dichotomy must be seen in relation to another dichotomy, that of auspiciousness–inauspiciousness (Carman and Marglin 1985 ). These scholars analyzed Indian words for the auspicious, like śubha and maṅgala , and found that such words are mostly used in contexts of time and events like astronomical constellations and lifecycle rituals, while words for purity are used for states of being, for instance with reference to the status of food and materials (Madan 1985 : 12–13, 17, 24).
These analyses point to an interrelation of two sorts of agency. The auspicious and inauspicious are mainly related to cosmic or divine agencies governing, for instance, weather phenomena and biological events, especially pregnancy, birth, disease, and death, and their relation to astrological patterns. The pure and impure are related mostly to human agency, that is, to the domain of dharma , understood as “the right way of behaving oneself” (Rocher 1972b: 86), and as such regarded as a precondition of securing auspicious responses through ritual activity and averting inauspicious consequences. In understanding some of the complexities involved in rules about purification in the Dharmaśāstra, I think it has proved useful to apply these two sets of dichotomies together.
One especially telling example is a series of rules concerning what should happen in the case of a man’s wife being revealed to have had a sexual relation with a Caṇḍāla . If the wife is already pregnant by her husband when she had the affair with the Caṇḍāla , she will not have to do any penance until the child is born, and the penance is the mildest in this series of rules. If she is not already pregnant and the sexual intercourse was interrupted, she will have to perform the penance as soon as the incident is revealed, and the penance is harsher than in the first case. If the intercourse was unintentional but was consummated without any conception, the penance is even harsher. If it was both intentional and consummated, but still without conception, then a more severe penance is needed. Finally, the worst case, if she becomes pregnant by the Caṇḍāla , no penance is possible and she must be exiled to another country (Par /PāM 10.16–10.21, 10.27c–10.28b in Tarkālaṅkāra 1883–99, vol. 2: 277–80, 284).
We may say that in the first of these events, auspiciousness is already at work, because the wife has become pregnant by her husband. The fetus is not influenced by the affair with the Caṇḍāla during pregnancy, but when the child is born, the wife needs to go through some purification in order to be a proper instrument for future conceptions. In the last event, it is quite the opposite: inauspiciousness has planted a bad seed in her womb and she is irreparably spoiled. But in the events in between these two, neither auspiciousness nor inauspiciousness is at work, since there was no conception. This allows the wife to go through hard penances and by this purification process, restore the purity needed for her to become pregnant by her husband after such an event (Aktor 2008 : 171–9).
Another anthropological contribution relates to the rules of death impurity that were presented earlier in this chapter. The American anthropologist Henry Orenstein wanted to find out the principle that might explain the seemingly contradictory situation between the rules that, on the one hand, govern the increasing amounts of penance (prāyaścitta ) enjoined on people from each of the four social classes (Varṇa ), from Śūdra to Brāhmaṇa, and, on the other hand, the decreasing durations of death impurity (āśauca ) for the four classes, again from Śūdra to Brāhmaṇa. Orenstein’s solution took it as basic that a certain level of inherent impurity is an essential property of each social class, lowest for Brāhmaṇas, highest for Śūdras—much in line with Dumont’s theory. His idea was then that in the case of penance, the pollution of a given sin is the same for all four classes, and the amount of purification will then be the level of pollution minus the level of inherent impurity. Since the latter is low for Brāhmaṇas and high for Śūdras, the subtractions will result in higher amounts of purification for Brāhmaṇas and less for Śūdras. This is perfectly in line with the rules about eating forbidden and unfit food (Par 11.1–11.3) that were discussed in the section on internal impurity: More penance for Brāhmaṇas, less for the lower classes.
However, in the case of death impurity, the impurity has to be multiplied since it affects not only the individual but also the family. Death impurity is therefore calculated according to a common multiplication factor, by which the inherent impurity levels must be multiplied. The low inherent impurity for Brāhmana is set to two, and the high one for Śūdra is set to eight. When these numbers are multiplied by a common multiplication factor, say five, the amount of purification is calculated by subtracting the inherent impurity levels from the numbers reached by the multiplication, that is ten minus two for Brāhmaṇa and forty minus eight for Śūdra. This gives the numbers eight and thirty-two, which are close to the prescribed number of āśauca days known from MDh 5.83 of ten and thirty respectively for Brāhmaṇa and Śūdra (Orenstein 1968: 116–117). As already discussed in the section on death impurity in this chapter, a more obvious explanation is the need for Brahmins to be able to perform sacrifices on behalf of the other classes as soon as possible.
In line with the structuralism of his time, Orenstein insisted on a structure, a “grammar of defilement” (the title of his article), that would overcome the empirical diversity of conflicting purity rules. Although his attempt is speculative, it has the merit of insisting on explanations. The problem lies in his method. It turns out that he relied on a few general rules, but neglected to account for the complexities that appear when a broader set of material is treated. For instance, he neglected the significance of the internal varṇa relations in cases where a person from one class has offended a member of another class, whether anuloma (high-class offender, low-class victim) or pratiloma (low-class offender, high-class victim). He also neglected the significance of the donation (dakṣiṇā ) that is offered at the completion of a penance. This counters the amount of penance proper: Brāhmaṇas observe more penance, but give less dakṣiṇā , while Śūdras observe less penance, but have more dakṣiṇā demanded—again, just as we saw in the rules discussed already about eating forbidden and unfit food. He also omits a discussion on the difficulty in comparing penances, the elements of which are very diverse (fasts, diets, durations, ingredients, mantras , etc. For a full discussion, see Aktor 2007 : 13–24).
Since the 1960s, when Orenstein wrote his article, Indologists have become better informed about developments in anthropological scholarship and other branches of the humanities. This more balanced combination of high standards of indological methods, with insights from more theoretical approaches to religious and legal history, can be an enriching enterprise for the future.