Chapter XIV. Remarriage of Widows

Volume II: Part I pp. 608–623

Chapter XIV. Remarriage of Widows

Remarriage of widowṣ—The word punarbhū is generally applied to a widow that has remarrieḍ Before going into the detailed history of the remarriage of widows it is desirable to probe into the meaning of the word ‘punarbhū’.¹ Nārada (strīpuṁsa, v. 45) says that there are seven sorts of wives (mentioned in order) who have been previously married to another man (parapūrvā); among them, the punarbhū is of three kinds² and the svairiṇī (wanton woman) is of four kindṣ The three punarbhūs are : (1) a maiden whose hand was taken in marriage but whose marriage was not consummated; in her case the marriage ceremony has to be performed once more; (2) a woman who first deserts the husband of her youth, betakes herself to another man and then returns to the house of her husband; (3) a woman who is given by the husband’s relatives (when the husband dies) to a sapiṇḍa of the deceased husband or a person of the same caste, on failure of brothers-in-law (this is niyoga and no ceremony is to be performed). The four svairiṇīs are: (1) a woman, whether childless or not, who goes to live with another man through love while the husband is alive; (2) a woman who rejects after her husband’s death his brothers and the like and unites herself with another through passion for him; (3) a woman, coming from a foreign country or purchased with money or oppressed by hunger or thirst, gives herself to a man saying ‘I am yours’; (4) a woman who is given to a stranger by the elders relying on the usages of the country, but who incurs the blame of wantonness³ (as the smṛti rules about niyoga are not observed by them or her). Nārada says that each preceding one of the punarbhūs and svairiṇīs enumerated is inferior to the next in order. Yāj. (I. 67) does not give this elaborate classification; all he says is that a punarbhū is of two kinds, one whose marriage had not been consummated and another who has had sexual intercourse and that both have the marriage ceremony performed again (i. e. punarbhū is one who is ‘ punaḥ saṁskṛtā ); a svairiṇī is one who forsakes the husband whom she married when a maiden and lives with another man of the same caste through love for the latter. Viśvarūpa on Yāj. I. 67 remarks that the elaborate classification of Nārada and Śaṅkha (3 punarbhūs and 4 svairiṇīs) is not of much use, that it only indicates the various degrees of blame ( or sin ) attaching to them and is also meant to discriminate among prāyaścittas to be performed by those womeṇ It is the second husband and the son of the second marriage that bear the appellation ‘ paunarhbhava’ ( pati or putra respectively ) and not the first husbanḍ Vide Sṃ Pr. pp. 740–741. The Sṃ C. ( I. p. 75 ) quotes a passage from Baudhāyana and certain verses of Kaśyapa. According to Kaśyapa ⁴ the seven kinds of punarbhū are : (1) the girl who had been promised in marriage, (2) one who was intended to be given; (3) one on whose wrist the auspicious band was tied by the bride-groom, (4) whose gift had been made with water ( by the father ), (5) whose hand was held by the bridegroom, (6) who went round the fire, (7) who had given birth to a child after marriage. In the first five cases it is to be supposed that the bride-groom either immediately died or left the further prosecution of the marriage riteṣ Even such girls would be styled punarbhūs, when they married another person later on, though the first marriage was not complete because the saptapadī had not been performeḍ The sixth case is one of completed marriage (though it refers to only going round fire). Baudhāyana's seven varieties are slightly different, the first two being the same as Kaśyapa's: the others are (3) one who went round the fire (with the bridegroom); (4) one who took the seventh step; (5) one who has had sexual intercourse (either after marriage or even without it); (6) one who has conceived; (7) one who has borne ⁵ a chilḍ These meanings of the word punarbhū must be borne in mind when one meets with the word punarbhū in Vedic textṣ That even the promise to give in marriage without the performance of any cere- monies was looked upon as tantamount in its consequences to marriage follows from the words of Sukanyā quoted above (in note 1306) from the Sat. Br., where she had been only offered to the sage Cyavana by her father, but where no ceremonies had been perform- ed as none are described or referred to thereiṇ Manu (IX. 69–70) confines the rules of niyoga to a girl who was only promised (vāgdattā); while Vaṣ Dh. S. XVII. 72 speaks of the vāgdattā and one given with water as still a maiden, if no Vedic mantras have been repeateḍ⁶ Vaṣ Dh. S. XVII. 74 refers to the 4th variety of Baudhāyana. Yāj. I. 67 when he speaks of akṣatā refers to all the six varieties of Kaśyapa or the first four varieties of Baudhāyana and when he speaks of ‘kṣatā’ he refers to the seventh variety of Kaśyapa and the last three of Baudhāyana. Vaṣ Dh. S.⁷ 17 19-20 describes paunarbhava as the son of a woman, who leaves the husband of her youth and after having lived with another person, re-enters the house of the husband or as the son of a woman who takes another husband after leaving an impotent, outcast or lunatic husband or after the death of the first husbanḍ Bauḍ Dh. S. II. 2. 31 describes paunarbhava as the son of a woman who after abandoning an impotent or outcast husband goes to another husbanḍ Nārada (stripuṃsa, v. 97), Parāśara IV. 30 and Agnipurāṇa 154. 5-6 have the same verse ‘another husband is ordained for women in five calamities viz. when the husband is lost (unheard of), is dead, has become a saṃnyāsin, is impotent or is patita.⁸ Great controversies have raged round this verse. Some like the Par. M. ( II. part I. p. 53 ) give the easy explanation (always given about inconvenient texts) that this verse refers to the state of society in another yuga (age) and has no application to the Kali age. Others like Medhātithi (on Manu V. 157) explain that the word pati means only ‘pālaka’ (guardian). Medhātithi⁹ (on Manu III. 10 and V. 163 ) is not dead against niyoga, but he is opposed to the remarriage of widowṣ Even the Smṛtyarthasāra ( which belongs to about 1150 to 1200 A. D. ) mentions several views viz. some hold that a girl may be married to another if the bride-groom dies before saptapadī, others hold that she may be remarried if the first husband dies before consummating the marriage, while still others are of opinion that if after marriage the husband dies before she begins to menstruate she may be remarried and some hold that remarriage is allowed before conceptioṇ¹⁰

Āp. Dh. S. II. 6. 13. 3-4 condemns remarriage 'if one has intercourse with a woman who had already another husband, or with a woman on whom no marriage saṁskāra has taken place or who is of a different varṇa, then sin is incurred; in that case the son also is sinful'. Haradatta quotes Manu III. 174 and says that the son procreated on another's wife is called kuṇḍa if the husband is living, and golaka if the husband is deaḍ Manu V. 162 is opposed to the remarriage of widows 'nowhere is a second husband declared for virtuous women'; so also Manu IX.¹¹ 65 ( 'in the procedure of marriage there is no declaration about the remarriage of widows ), Manu IX. 47 ' a maiden can be given only once' and Manu VIII. 226 (the Vedic mantras used in pānigrahana are applicable to maidens only ) are opposed to the remarriage of widowṣ The Brahmapurāna and other purāṇas forbid remarriage of widows in the Kali age ( vide note 1427 a above ). The Saṃ Pr. quotes a text of Kātyāyana to the effect that a girl, who has gone through the ceremony of marriage with a sagotra, may be married again and remarks that the text refers to the state of society in another yuga. This is the view of all the commentators and nibandha writerṣ Manu himself ( in IX. 176 ) expressly allows the saṁskāra of remarriage in the case of a girl, whose first marriage has not been consummated or who left the husband of her youth, went to live with another and returned to the first husbanḍ In this the author of the Manusmṛti probably only reiterates popular usage which was too much for him in spite of his own view ( in V. 162 ) denouncing remarriageṣ So it may be taken that Manu does not forbid the use of mantras in remarriage, but holds that even after the mantras are recited the remarriage of a widow is not dharmya ( approved ). It is said in the Mahābhārata ( Ādiparva 104. 34-37 ) that Dirghatamas forbade remarriage and also niyoga.¹² Manu himself speaks of the saṁskāra of a pregnant girl ( IX. 172-173 ). Bauḍ Dh. S. IV. 1. 18, Vaṣ Dh. S. 17. 74, Yāj. I. 167 speak of the saṁskāra of remarriage ( paunarbhava saṁskāra ). Manu III. 155 and Yāj. I. 222 include the paunarbhava ( the son of a punarbhū ) among brāhmaṇas that are not to be invited at a śrāddha. Aparārka ( p. 97 ) quotes a passage from the Brahmapurāna itself which speaks of a fresh saṁskāra of marriage for a child widow or for one who was forcibly abandoned or carried away by somebody.¹³

Several smṛtis contain certain rules about what the wife was to do when the husband had gone abroad for many years after marriage. Nārada ( strīpuṃsa, verses 98-101 ) gives the following directionṣ¹⁴ ‘If the husband has gone to a foreign country a brāhmaṇa wife should wait for eight years, but four years if she has not given birth to a child; after that period (of 8 or 4 years) she may resort to another man (then Nārada lays down lesser number of years for kṣatriya and vaiśya wives). If the husband is known to be living then the periods are double of those stated above; this is the view of Prajāpati when no news can be had of persons and hence there is no sin if a woman resorts to another man (in such cases).’ Manu (IX. 76) says ‘If a man has gone to a foreign land for doing some religious duty the wife should wait for him for eight years, six years if he has gone for acquiring knowledge or fame, or three years if he has gone on a love affair (or for another wife).’ Manu does not state what the wife is to do after these years of waiting. Vaṣ (17. 75-76) requires that ‘the wife of one who has gone to a foreign land should wait for five years and after five years she should go near her husbanḍ’ This may be all right as far as it will go. But if the husband is unheard of, how is she to reach him? Vasiṣṭha says nothing on that point. Viśvarūpa on Yāj. I. 69 says that the periods of waiting prescribed for the wife of an emigrant are meant not for permitting niyoga thereafter, but for calling upon her to repair to her husbanḍ Kauṭilya (III. 4) prescribes some interesting rules: ‘the wife of one, who has long gone abroad, or who has become a recluse or who is dead, should wait for seven menstrual periods and for a year if she has a child already. Thereafter she may marry the full brother of her husbanḍ If there be many brothers she should marry one who is near in age (to the first husband), who is virtuous, capable of maintaining her or who is the youngest or unmarrieḍ If no such brother exists she may marry a sapiṇḍa of the husband or one of the same caste.’ The story of Damayanti suggests that when the husband was not heard of for many years, a wife could marry agaiṇ Damayanti is said to have sent a message to Ṛtuparṇa that, as Nala was not heard of for many years, Damayanti was going to celebrate a svayaṁvara and Ṛtuparṇa hurries for it and does not think it a strange thing (Vanaparva 70. 24). One question¹⁵ raised by Dr. Banerjee is: what is to be regarded as the gotra of a widow when she is to be remarried (is it to be her father’s gotra or of the first husband’s?). There are hardly any indications in the ancient smṛtis or commenta-ries on this point. Viśvarūpa¹⁶ commenting on Yāj. I. 63 (on the word ‘kanyāprada’) observes that according to some the father gives away the bride even if she is not a virgiṇ So it appears that the father’s gotra should be looked to in the remarriage of a widow. Vidyasagar, whom Dr. Banerjee follows, held the same opinioṇ

Certain passages of the Atharvaveda may be considered in connection with the question of the remarriage of widowṣ Atharva-Veda V. 17. 8–9 are¹⁷ ‘when a woman has at first even ten husbands, who are not brāhmaṇas, if a brāhmaṇa takes hold of her hand (i. e. marries her), he alone is her (real) husbanḍ A brāhmaṇa alone is (a real) husband, not a kṣatriya or a vaiśya—the sun goes proclaiming this to the five (tribes of) men’. The first verse is not to be taken literally in the sense that a woman married ten persons in succession and that the 11th was a brāhmaṇa; the first verse contains rather what is called ‘praudhīvāda’ (pompous assertion or boast) and this is indi-cated by the word ‘uta’. The verse may at the most mean that if a woman has first a kṣatriya or vaiśya as husband, and she marries on his death a brāhmaṇa, then the brāhmaṇa is the real husbanḍ The word ‘pati’ may also have been used loosely and all that is meant may be that if a girl is promised to ten persons one after another and then lastly to a brāhmaṇa, the latter is to be accepted as the best. Another passage¹⁸ of the Atharvaveda (IX. 5. 27–28) is ‘whatever woman, having first married one husband, marries another, if they (two) offer a goat with five rice dishes they would not be separated (from each other)’. The second husband secures the same world with his remarried wife, when he offers a goat accompanied with five rice dishes and with the light of fees’. Here the word punarbhū occurṣ It is possible to hold that this may refer to the promise of a girl in marriage, subsequent death of the intended bridegroom before the marriage ceremonies take place and then the bestowal of her on another. Whatever the meaning of punarbhū here may be, it is clear that some sort of sin or inferiority attached to her and that such sin or opprobrium had to be removed by sacrificeṣ Other passages are dis- cussed later oṇ But it must be admitted that remarriage of women was not prohibited in the times of the Atharvaveda. In the Tai. S. III. 2. 4. 4 ‘daidhīṣavya’ (widow’s son) occurṣ The gṛhyasūtras are silent about remarriage; so probably by that time it had come to be prohibited generally, though sporadic instances might have occurreḍ Among the brāhmaṇas and castes similar to them and holding or endeavouring to hold a high place in the hierarchy of castes widow remarriage has been forbidden for centurieṣ One of the earliest historical instances is the remarriage of Dhruvadevī, queen of Rāmagupta, who was after Rāmagupta’s death, married by her brother-in-law Candragupta (vide Journal Asiatique for 1923, pp. 201–208, Sanjan Plates in E. I. vol. 18, p. 255, ‘Indian Culture’ vol. 4, p. 216, Harṣacarita VI, penultimate para). Among śūdras and other lower castes widow remarriage has been allowed by custom, though it is held to be somewhat inferior to the marriage of a maideṇ¹⁹ Among these castes remarriage is allowed after the death of the husband or during his life-time with the consent of the husband who gives a writing called farkhat or sod-chuṭṭī (a deed of release). Such marriages are called pat in Mahārāṣṭra, natra in Gujarat, uḍki in the Canarese districtṣ In some cases the caste in a meeting assembled takes upon itself to bring to an end a marriage and allows the wife to remarry. But the Bombay High Court does not recognize the authority of a caste to declare a marriage void or to permit a woman to remarry another person without the consent of the husband and have convicted of bigamy women re-marrying without the consent of the first husband but with the permission of the caste.²⁰ In modern times, the Hindu Widows' Remarriage Act (XV of 1856) has been passed mainly through the efforts of Pandit Ishwar Chandra Vidyasagar. That Act legalises the marriages of widows notwithstanding any custom or interpretation of the Hindu Law to the contrary and declares that the children of such marriages are legitimate. Owing to the sentiment of centuries widow remarriage is still looked down upon and during more than eighty years since the Act was passed not many widows have taken advantage of it. The Census²¹ of India for 1931 discloses certain appalling figureṣ Among infants of less than one year old there were in the whole of India as many as 796 males who were widowers and 1515 females who were widows, among children up to 5 years there were 12799 widowers and 30880 widows (out of these last the Hindus contributed 10266 as widowers and 23667 as widows). It is true that amongst the castes to which most of these must have belonged widow remarriage is allowed by custom, but there must be a certain number of females who are widows before five, who cannot remarry by the custom of the caste and who would not dare to take advantage of the Hindu Widows' Remarriage Act. Between the ages of 5–10, 11–15, 16–20 the numbers of Hindu widows for the whole of India are respectively 83920, 145449, 404167. The age of marriage for girls is rapidly rising owing to economic causes, the spread of literacy among the masses, and the opera- tion of the Child Marriage Restraint Act (XIX of 1929 ) and it may be hoped that these high figures of child widows will be substantially reduced in the near future. Persons finding fault with Hindu society for large numbers of child widows should not forget one thing. According to Hindu notions every girl must be married at any cost. Hence people rush into child marriageṣ But there has been no problem of old spinsters in Hindu society, as there is in Western countries, though it appears that in the near future that problem will arise even in India. Apart from the considerations as to religious texts prohibiting widow remarriage and the great concern among women for preserving a high ideal of chastity and single-minded devotion to the husband, many Hindus feel that, each girl having been given one chance of marriage, if she becomes a widow, that is her ill-luck, but she should not be allowed to compete with unmarried girls in the marriage market, which is already overcrowded with girls waiting to be marrieḍ

Certain verses of the Ṛgveda and the Atharvaveda have given rise to various explanations and opinion is divided as to whether they refer to niyoga, to the remarriage of widows or to the practice of the immolation of widowṣ First the two verses Rg. X. 18. 7-8 ²² which are part of a funeral hymn will be set out : ‘Let these women, who are not widows and who have good husbands, sit down with clarified butter used as collyrium ; may the wives who are tearless, free from disease and wearing fine jewels (or clothes) occupy the seat in front (first). O woman! raise yourself towards the world of the living; you lie down near this departed (husband) ; come, this your wifehood of the husband who (formerly) held your hand and who loved you has (now) been fulfilled’. It is somewhat strange that Sāyana understands the latter half as an invitation by the husband’s brother to the wife of the departed to remarry hiṃ But that meaning is far-fetched and does not bring out the proper force of ‘hastagrābhasya’, ‘patyuh’ and ‘babhūtha’. The Āśv. gr. ²³ (IV. 6. 11-12) prescribes that (in the expiatory rite performed on the death of an elder) the young women relatives should salve their eyes with butter with their thumb and fourth finger and with tender darbha blades and then throw the blades away, when the performer of the rite should look at them while they are salving their eyes with the verse ‘imā nārīr &c.’. The same sūtra (IV. 2. 16–18) says that “(when a dead body is to be cremated ) they should place the wife to the north of the body and a bow for a kṣatriya ( if he be the dead person ) and then her brother-in-law or some one else who can be the representative of her husband or a pupil or an old servant should make her rise up with the mantra 'udirṣva'". The Śāṁ. Śr. S. IV. 16.16 also directs that the verse Rg. X. 18. 7 is repeated when the eyes of women are touched with blades of kuśa grass and XVI 13. 13 says that Rg. X. 18. 8 and Rg. X. 85. 21-22 are called 'utthāpinyah' (verses addressed for making one rise). There is another mantra in the Atharvaveda (18. 3. 1) and Tai. Ār. VI. 1 ²⁴ 'Oh (dead) man! this woman choosing the world of the husband lies down by thee, the deceased, observing the old universal custom; bestow on her in this world offspring and wealth'. The first verse (i. e. Rg. X. 18. 7) contains nothing about niyoga or remarriage. It has been made use of for giving Vedic authority to the practice of satī (which will be dealt with later on ). That verse refers to the practice that young women of the household of the departed used to go to the cemetery and applied clarified butter to their eyes ( by way of purification and śānti ); this practice continued to the days of the Āśv. gr, the Bṛhad-devatā (VII. 12 ) and the Bauḍ pitrmedhasūtra (I.21.11). The two verses 'iyam nārī' and Rg. X. 18. 8 are employed by the Baudhāyana-Pitrmedhasūtra ²⁵ in the funeral rites, the first to be repeated when the wife is made to sit near the corpse and the next for making her rise. It is to be noted that Bauḍ directs that the corpse is placed on the funeral pile after the wife is made to rise from the vicinity of the corpse; while the Bṛhad-devatā appears to suggest that the wife ascends the funeral pile after the corpse is placed thereon and then the younger brother forbids her with the verse 'udirṣva &c.'. But the Bṛhad-devatā ²⁶ does not mean that the wife burns herself on the funeral pyre and the brother-in-law contents himself with only repeating a verse to dissuade her. The Rgvidhāna ²⁷ (III. 8. 4) says that the brother-in-law should call back the wife of his sonless brother when she is about to ascend the funeral pyre for procreating a son on her with Rg. X. 18. 8. It appears that the verse Rg. X. 18. 8 symbolically describes what even in the days of the Ṛgveda was probably only a tradition viz. that in hoary antiquity a wife burnt herself with her husbanḍ In the times of the Ṛgveda this practice had altogether ceased, but a symbolical imitation of it had come into vogue, viz. that the wife lay near the corpse in the cemetery and then she was asked to get up and was told that by following her husband to the very doors of death she had fulfilled all that was expected of her and that she should returṇ The same idea is referred to by the verse 'iyam nārī', but the latter half appears to refer to the practice of niyoga when it calls upon the departed to bestow on the wife offspring and wealth. It is possible to argue that Rg. X. 18. 8 also impliedly has niyoga in view. But both these verses do not expressly refer to the parctice of sati at all. They at the most might lead to the inference that the woman had either to marry the deceased husband's brother or go in for niyoga because the husband died sonlesṣ The latter appears to me to be the more probable of the two hypotheses, if those are the only explanationṣ The symbolic use made in the grhyasūtras and in the Brhad-devatā of Rg. X. 18. 8 shows that they practically are against the burning of widowṣ It is probable that these authors knew of the custom of niyoga, that it was not then much approved of and so they are silent about it, while the Rgvidhāna which appears to be comparatively a late work takes the verse 'udirşva' as referring to niyoga (which is most probably referred to in Rg. X. 40. 2 ko vām śayutrā vidhaveva devaram). The consideration of the practice of widow burning naturally arises here from the above discussion and will be dealt with in a separate chapter.

Divorce

In the Vedic literature there are at least some texts capable of being interpreted as relating to the remarriage of widows and we have the word 'punarbhū'. But as regards divorce there is absolutely nothing in the Vedic texts nor in post-vedic literature. The theory of dharma-śāstra writers is that marriage when completed by homa and saptapadī is indissoluble. Manu IX. 101 says 'Let mutual fidelity (between husband and wife) continue till death; this in brief may be understood to be the highest dharma of man and wife.' In another place Manu (IX. 46) declares 'neither by sale nor by desertion is the wife released from the husband; we understand that this is the law ordained by the Creator in former timeṣ' The position of the writers on dharmaśāstra is that marriage is a saṁskāra, that the status of wife-hood arises from that saṁskāra, that even if the husband or wife became patita, the saṁskāra already performed is not annulled by that fact, that even if a wife committed adultery she still remains a wife and that when she performs a penance for her lapse, it is not necessary to have a fresh saṁskāra of marriage performed on her.²⁸ We have seen (pp. 552–553) that a man was allowed to supersede a wife and marry another or others or to abandon his wife altogether in certain circumstanceṣ But that does not amount to divorce (i. e. dissolution of the marriage tie); the marriage is still there intact. It was also shown (at pp. 610–611) that according to Nārada, Parāśara and a few others a woman was allowed to remarry in case the husband died, or was unheard of &c.; but according to the digests and commentaries these rules apply to a former yuga (age). Therefore divorce in the ordinary sense of the word (i. e. divorce a vinculo matrimonii) has been unknown to the dharmaśāstras and to Hindu society for about two thousand years (except on the ground of custom among the lower castes). Even when the husband was allowed to abandon the wife for her lapse, still she was in most cases entitled to at least starving maintenance. Therefore tyāga (abandonment) was not only no divorce a vinculo at all but was not even a divorce a mensa et thoro (divorce from board and bed). Later smṛtis and medieval digests could hardly conceive of any ground for which the wife could desert her husband altogether, though Nārada and a few others allowed her to desert one husband and marry another if he was impotent, or became a saṁnyāsin or an outcast. The Mit. on Yāj. I. 77 says that a wife is not under the control of her husband as long as he remains patita (outcast or excommunicated) and that she should wait till he is purified by penance and restored to caste and that thereafter she again becomes dependent on hiṃ The gravest sins can be expiated by penance ( vide Manu XI. 89, 92, 101, 105–106 ); therefore it follows that a wife could not for ever desert even a patita husbanḍ In modern times the Indian courts²⁹ have held that mere change of faith or apostasy by a Hindu wife or husband does not ipso fact dissolve the marriage and that if the wife changes her religion and then marries another while the first husband is still living, she would be guilty of bigamy. Mere desertion or separation for many years or even adultery does not dissolve a Hindu marriage ( vide 42 Madras Law Journal 276 ). Only in cases where the husband or wife becomes a convert to Christianity, a special procedure is prescribed for the dissolution of the original Hindu marriage by the Native Converts' Marriage Dissolution Act (XXI of 1866), which is not set out here as it is beyond the scope of this work to do so. Further, those Hindus who marry under the Special Marriage Act (III of 1872 as amended by Act XXX of 1923 ) can secure divorce under the Indian Divorce Act (IV of 1869 ). The Malabar Marriage Act ( Madras Act IV of 1896 ) forbids polygamy if the first marriage is registered and allows divorce ( sec. 19 ) among those who are governed by the Marumakkata- yam or Aliyasantana Law prevalent in Malabar.

The Arthaśāstra of Kauṭilya contains some interesting observations bearing on divorce.³⁰ 'A wife hating her husband cannot be released from the husband if he is unwilling ( to let her go), nor can the husband release himself from the wife ( if she is unwilling ) ; but if there is mutual hatred then release is possible. If a man fearing danger (or injury) from his wife desires release from her, he shall return to her whatever was given to her (at the time of marriage). If a woman out of fear of danger (or injury) from the husband desires release, the latter need not return to her what was given to her (at the time of marriage) ; marriages in the approved form cannot be dissolveḍ' Kauṭilya himself says (in III. 2) that the first four forms viz. brāhma, prājāpatya, ārṣa and daiva are dharmya (approved), since they are brought about under the authority of the father.³¹ Therefore according to Kauṭilya there can be no dissolution of the marriage tie if the marriage was in the gāndharva, āsura or rākṣasa form, then the tie may be dissolved by mutual consent, if both have come to hate each other. But he seems to hold that there can be no release at the instance of only one party to the marriage who has begun to feel aversion to the other party in whatever form the marriage may have been performeḍ Even in this latter case he apparently makes an exception, where physical danger is apprehended by one party from the other.

It is beyond the scope of this work to compare the law of divorce in other countries or under other religious systemṣ It may be stated, however, that according to the strict theory of the Roman Catholic Church the marriage tie is indissoluble,³² though decrees of nullity of marriage were sometimes granted by that Church to those who could pay for theṃ In England after the Restoration divorce could be secured through the Parliament by a private Bill where a divorce a vinculo matrimonii was desireḍ But this method could be resorted to only by the rich, as the passage of a private Bill for divorce cost at least 500 £. The Ecclesiastical courts in England granted divorces a mensa et thoro on the ground of adultery, cruelty or unnatural offences, though such a divorce did not dissolve the marriage. But this procedure also was costly, as even an un-defended suit for divorce a mensa et thoro would ordinarily cost from 300 to 500 £. Then came the Matrimonial Causes Act of 1857 (20 and 21 Vic. chap. 85) by sec. 27 of which a wife could petition for divorce only if she proved that since the celebration of marriage the husband was guilty of incestuous adultery, or of bigamy with adultery, or of adultery coup-led with such cruelty as, without adultery, would have entitled her to a divorce a mensa et thoro or of adultery coupled with desertioṇ The Matrimonial Causes Act of 1923 (13 and 14 Geo. 5 chap. 19) placed women on an equality with men by allowing them to apply for divorce merely on the ground of adultery by the husband without having to prove anything more. Then the Act of 1937 known as A. P. Herbert's Act (1 Edw. 8 and 1 Geo. 6 chap. 57) allows the husband or wife to petition for divorce on four groundṣ It will be noticed from this that the absence of divorce and rigorous restrictions thereon are not a peculiarity of the Brahmanical religion or of the caste system, but existed even in so-called progressive, casteless and Christian countries up to very recent timeṣ In modern times even in Roman Catholic countries divorce is allowed on various grounds; and in India too efforts are being made to enact legislation permitting divorce among Hindus for various reasonṣ And it may be admitted that some legal provision is necessary for securing divorce in hard cases even as to marriages celebrated under the ancient śāstric system on grounds similar to those in the English Act of 1937.