Lawrence Jenkins, C.J.
1. By this suit the plaintiff seeks a declaration that the 1st defendant is not the validly adopted son ofDamodardas Tapidas, and that his alleged adoption by Harkor, the widow of Damodardas, is void and of no effect. Damodardas and the plaintiff were the sons by different wives of Tapidas Varajdas, who died in 1886.
2. It is common ground that they were separate in estate.
3. On the 8th of August 1895 Damodardas executed a will in which there was the following clause :
Should I after my death have no male issue of my (loins) and should I in my life time have not adopted any one as my son, then my first wife Harkor and should she be not alive then my second wife Amba Lakshimi, shall after my death adopt any boy out of the descendants of my father (ancestors) and if amongst them there be no boy who can according to the Shastras (i. e. religious texts) be adopted, then out of my caste, by performing the ceremony prescribed in the shastras (i. e. religious texts) (sick).
4. He also executed two codicils to his will dated the 22nd of February 1897 and the 15th of August 1898. On the 5th of July 1890 the testator died, leaving two widows Harkor and Ambalaxmi, but no issue. He was also survived by the plaintiff and the plaintiffs son Carsondas who is still alive.
5. Probate of the will and codicils was on the 12th of March 1900 granted to Harkor and 2nd defendant, Hargovandas Mulchand.
6. On the 5th of December 1904 Ambalaxmi died.
7. On the 11th September 1905 Harkor purported to adopt the 1st defendant, the son of the 2nd defendant who was her sister's son, and it is this adoption that is questioned in this suit. A deed of confirmation was executed on the 16th of the same month. On a preliminary issue Beaman J. decided in favour of the adoption, but his decree, dismissing the suit was not passed until after this appeal was preferred. All objections however on this score have been waived, and by consent of all parties this is to be treated for all purposes as an appeal from the decree of dismissal passed by Beaman J.
8. The appeal in the first instance came before Heaton J. and myself, but as it appeared to us that this was eminently a case where familiarity with the vernacular was desirable, Chandavarkar J. has joined us with the consent of both sides.
9. There is no dispute as to the fact of adoption, and for the defendants it is claimed that it was properly made in exercise of the power contained in Clause 6 of the will. The plaintiff contends that this clause does not justify what has been done.
10. He argued first that under it Harkor could only have adopted the 1st defendant in case there was no descendant of the testator's father, Tapidas, and, secondly, that even if the words authorized an adoption in the first instance from among the testator's sapindas, it is not proved that the first defendant answers that description.
11. First, then, let us consider whether the power is of the limited scope for which the plaintiff contends. The will is in Gujarati and the words descriptive of the class out of which the selection was in the first instance to be made are mara bapika vanshmanthi and Beaman J. would render them 'from amongst my paternal agnates'.
12. Mr. Jayakar, who has argued the case with considerable force and ability on behalf of the appellant, contends that this rendering is erroneous and that the words mean 'from among the descendants of my father'.
13. Chandavarkar J. will deal more critically than I can with this part of the case, and I will do no more than state briefly the grounds on which I think the view of Beaman J. should be upheld.
14. Now mara is a possessive pronoun, bapika according to the evidence of the plaintiff's witnesses is an adjective, and vanshmanthi is a noun.
15. Both sides are agreed as to this. But if bapika is an adjective, then according to the ordinary rules of grammar the possessive pronoun is not descriptive of it but of the noun vanshmanthi. Even the plaintiff's evidence shows that vansh may mean, clan, race, lineage or family, as well as descendants, and if mara be read as descriptive of vanshmanthi it cannot, in the context in which it is used, be reasonably read as ''from among my descendants', for adoption from among them would be impossible. Therefore vansh here must mean clan, race, lineage, or family.
16. And if bapika is an adjective and is not qualified by mara, it must in my opinion have the force of paternal, or related on the father's side.
17. The expression therefore means from among my family on the father's side, or, as Beaman J. has rendered it, from amongst my paternal agnates.
18. And now I will deal with the argument that the first defendant is not proved to have been a sapinda of the testator.
19. The pedigree on which the defendant relies is annexed to his written statement, and, if that be proved, the necessary relationship is established.
20. First then we have Ex. No. 2, which is a Gujarati statement on the face of an envelope in the handwriting of Damodardas, and it is there written, 'My funeral ceremonies are to be got performed by Bhai Hargovindas Mulchand Kasidas. The same are not to be got performed by any other person'.
21. The Gujarati, thus rendered, makes it clear that Hargovandas was to perform the ceremonies, and this is conceded by counsel for the appellant.
22. Exhibit No. 3 is another statement to the like effect. The Hargovind here mentioned is the 2nd defendant, and it is established that the funeral ceremonies were in fact performed by him. This affords a strong indication that the testator regarded Hargovind as his Sapinda, and Ex. No. 2 is important as showing that Hargovind's father was Mulchand, and his grandfather Kasidas.
23. And next we have the plaintiff's own admission that he observed Sutak for the mother of Mulchand Kashidas, and that is a clear indication of Sapinda relationship.
24. A suggestion was made before us in argument that among Gujaratis this well established indication would not have the same force; but this there is nothing to support.
25. Mulchand Kashidas declares that he observed Sutak for Damodar, and that Dayabhai performed the ceremonies for one day and after that Hargovind performed.
26. Of the truth of their statements I feel no doubt and they strongly favour the Sapinda relationship for which the 1st defendant contends.
27. Chunilal Nagindas, a comparatively near relative of Dayabhai on the father's side, says that Hargovan is his cousin on the paternal side, and I cannot find that this statement is in any way shaken in cross-examination.
28. All this clearly points to the 1st defendant's being an agnate of the testator.
29. And now I come to the proofs adduced in favour of the several relationships shown in the pedigree.
30. It will be seen that the common ancestor is Chakasa, and I will first deal with Dayabhai's side of the pedigree.
31. The pedigree up to Atmaram and Parbhudas is admitted by the plaintiff, and it is in my opinion satisfactorily proved by Chunilal Nagindas and Exs.4, 11 and 12 that Parbhudas was the son of Purshotamdas and that Purshotamdas was the son of Chakasa. For this purpose I discard the latter part of Ex. 4, as its genuineness appears to me to be open to question.
32. On the other side of the pedigree there can be no doubt as to its accuracy as far as Laldas, and the only question is whether the relationship between Laldas Pitamber and Chakvasa is made good.
33. In considering whether this relationship is proved I will not rely on Mulchand's deposition as from the terms in which it is recorded it is doubtful whether his only source of information was what he learnt from the priests at Nasik.
34. But in the circumstances I think it is sufficiently established by Exs. 11 and 12 and Exs. 8. 9 and 10 taken with the evidence of agnatic relationship to which I have already alluded.
35. As against this body of evidence no proof is adduced by Dayabhai, nor is any explanation furnished by him of his own conduct, which unquestionably points to the existence of the relationship indicated in the pedigree.
36. It only remains for me to consider the objection that as the testator's will gave rise to the difficulty Beaman J. should not have ordered the costs to be borne by the plaintiff. But I can see no sufficient reason for interfering with the exercise by the learned Judge of his discretion as to costs and the result is that we must confirm his decree with costs.
37. The question whether the adoption of the first respondent, Chunilal, by Harkore, one of the widows of the testator, Damodar Tapidas, is valid, depends upon the construction of certain Gujarati words in the 6th clause of the will, which are 'mara bapika, vamshamanthi.' The official translation of these words is 'out of the descendants of my father (ancestors.)'
38. The case for the appellant, which has been placed before us in appeal by his learned Counsel in an exceedingly able and clear argument, is that the word bapika, which is an adjective, is rightly translated as 'of my father' but that the explanation, added in parenthesis by the Official Translator, that it means 'ancestors,' is wrong. Shortly summarised, the learned Counsel's argument is as follows. The word bapika is a derivative adjective, which retains its power, and the possessive pronoun mara which immediately precedes, refers to it and not to the word vamsa. Therefore, it is urged, the correct translation is: ' The vamsha of my father'.
39. This argument loses sight, however, of the fact that in the Gujarati, as in the Marathi, language, the possessive pronoun, when used with a noun substantive, has always an affix denoting gender, which is the same as that of the noun which it governs. And when the noun has an adjective preceding it but following the possessive pronoun, the adjective too takes its gender from the noun. There is no rule of grammar in either language, according to which the possessive pronoun can be solely connected with the adjectival word. The learned Counsel relied upon the Sanskrit expression 'mama matrikam dhanam' [my maternal wealth] in support of his argument, because the Gujarati language has borrowed its grammar and words more or less from Sanskrit. But there is this difference between the two that in Sanskrit-as in Canarese-the possessive pronoun mama does not vary as to gender with the noun substantive, which it governs, but remains the same. It is otherwise in Marathi and Gujarati. But even in Sanskrit, in such expressions as mama matrikam dhanam (my maternal wealth) and mama paitrikam dhanam (my paternal wealth,), the words matrika (maternal) and paitrika (paternal) are adjectives of secondary derivation from the words matri (mother) and pitra (father) respectively; and, according to the rules of Sanskrit syntax, the possessive pronoun mama (my) cannot be grammatically construed or connected with either adjective, matrika (maternal) or pitrika (paternal).
40. When, therefore, the testator has used the words in Gujarati, 'my paternal vamsha', the correct translation is : 'my vamsha on my paternal side '.
41. It is admitted before us by Mr. Jayakar, and indeed most of the appellant's witnesses concede, that the adjectival word bapika (paternal), when used of an estate, means 'ancestral', though literally the word means 'paternal', i. e. 'of father'. But it is contended that when the same adjective is used to qualify the noun vamsha, it has the literal meaning. And in support of the contention we have been referred to the meanings of the word bapika given in some of the Gujarati-Eng-lish dictionaries and Gujarati books. In one of them 'paternal; partimonial (as an estate)' are given as the English equivalents of the word. The word 'estate' seems to be mentioned there as merely illustrative, not exhaustive or exclusive. The word bapika is the same in meaning as the Sanskrit pitrika. Both literally mean paternal; and paternal is used both in the Hindu law books and. common language to mean, 'ancestral', referring not merely to the father but also the grand-father and great-grandfather. To cite a familiar illustration, 'ancestral property' is called pitrasthi or pitrudhana or paitriha dhana. If bapika or paitrika means 'ancestral' when it is used with the noun 'estate', there is no reason why it should lose that meaning when it is used so as to qualify any other noun.
42. No doubt in some of the Gujarati books of reference cited in support of his argument by Mr. Jayakar it is stated that the adjectival word bapikoo is the same as bapnoo, the genitive case of bapa (father), meaning 'of the father' But it must be remembered that it is one short way of giving the meaning of the word bapikoo when it is used singly in a dictionary. But it would be idle to argue that the expression' my paternal wealth' (mari bapiki milkat) conveys generally the same meaning as the -expression 'the wealth of my father' (mara bapni milkat). The former prima facie means that it is wealth which belongs to the speaker, having descended to him from his paternal side, and would include not only the father but his ancestors unless there is anything in the context to confine its origin to the father. The latter, on the other hand, prima facie implies that the wealth still belongs to the speaker's father, unless there is anything in the context to the contrary, in which case it may mean wealth which has come to the speaker from his father. This is illustrated by the 21st clause of the will and the 2nd clause of the first codicil in the present case. In the 21st clause the testator at the outset mentions the property derived by him and his step-brother Dayabhai from their father under a will as 'property acquired by us under the will of our father'. And in Clause 2 of the codicil, after referring to the same property 'as the property and effects of my father' he says it has come to his share. And having thus made it clear that the property belonged to and has come to him from his father, he describes it towards the end of the clause as mari bapki mil/cat ('my paternal property') and sarve bapiki milkat ('all paternal property'). There bapiki milkat which means anestral property is described in specific terms as a particular species of paternal, i. e, ancestral property-i. e. property which has come to him from the father and no other ancestor.
43. Mr. Jayakar has strongly relied upon the use of the adjectival words bapki and bapiki in Clause 2 of the 1st codicil in support of his argument that the same adjective, used in 'clause 6 of the will, was intended by the testator to mean not ancestral but 'of the father.' Mr. Jayakar contends that the testator by using the adjectival word in the sense 'of my father' in the former clause has given us a clue to his meaning of the word. But in Clause 2 of the codicil the testator describes in so many words the property to which he refers as being property which has come to him from his father under a will; and it is after' taking care to give that description that he classes it under or rather describes it as ' bapiki milkat', in the same clause. What that shows is that he included the said property in that class or description, i. e. in the class of paternal property. But it does not show that on that account he intended the phrase bapiki itself to be confined, wherever he has used it, to that which has come to a son from his father, and not to include that which may have come from any of his other ancestors. And when, after giving particulars of the property, referred to in Clause 2 of the codicil, the testator calls it bapiki milkat, whereas in Clause 6 of the will he uses the expression mdrd bapikd vamsha without giving a similar explanation or description as to it, it is reasonable to infer that he intended the latter expression to retain its larger meaning of his ancestral vamsha, including his ancestors not merely the father. In other words, where he intends the adjectival word bapika to refer to the father and no other ancestor, he says so or rather takes care to make his meaning clear by giving a specific description of the source of the property. Where he does not intend that restricted meaning, he simply uses the expression and says no more. The testator has no doubt made a dictionary for himself ; but such as the dictionary is, so far from supporting the appellant's construction of the 6th clause of the will, it shows that construction to be contrary to the testator's intention.
44. Assuming, however, that the adjectival word bapika means 'father's ' simply and not ' ancestral' and that the possessive pronoun mara refers to it and not to the noun vamsha, the question even on that view of the case remains whether the word vamsha means the line of the father's descendants and does not include the line of his ascendants. Mr. Jayakar's contention is that the strict meaning of the word is descendants; and that generally it should be construed in that sense unless the context in any particular case shows that the ascendants were also intended to be included. The word is of Sanskrit origin. In Professor Monier Williams's Sanskrit Dictionary its different meanings are given, and these include 'race,' 'family,' So also in the St. Petersburg Sanskrit' Dictionary. In the two Sanskrit Lexicons (Koshas)-Vishwa and Medini-it is stated to be a synonym for the word ' Kula,' which means ''clan' or 'family'. The word has been adopted from Sanskrit into all the three vernaculars of this Presidency-Marathi, Gujarati, and Canarese, and is commonly used in the same sense in which it is used in Sanskrit. On page 11 of Vaahistha Dharma Shastram (Bombay Sanskrit Series, No. XXIII, page 11) we find the expression matripitru vamshaha, which is translated into English in para 19 at pages 19 and 20 of Vol. XIV of the Sacred Books of the East, as follows :-'One whose ancestors both on the mother's and on the father's side'. So also Nilakantha in the Vyavahara Mdyukha referring to ' royal writings,' quotes a text of Yajnyavalkya in which the word vamshyan occurs. The text is translated by Rao Saheb Mandlik as follows:-
On a piece of cloth or a copper plate marked above with his seal, the king, having written down the names of himself and his ancestors etc' [page 19, lines 5 to 7].
45. The Sanskrit words in the above text for ' his ancestors' are 'atmano vamshyan'. The word atmano is the genitive case of atman and literally the expression means ' vamshyas of his own'. If the appellant's counsel is right, it ought to mean ''descendants of his own'; but he had to admit in the course of urgument that the word vamsha is used in the text above given to mean ' the line of his' (the king's) 'ancestors'. The same text is quoted and explained by Vijnaneshwara in the Mitakshara (See ver.ses 319 and 320 Chapter on 'the Duties of a King 'in the section on Rituals, pages 94 and 95, Mitakshara edited by Bapu Shastri Moghe, 3rd Edition). Vijnane hwara says that atmano vamshyan means ' the great grand father, the grand father, and others pitras (ancestors).' (sick) in the same chapter, a text is quoted from Yajnyavalkya, in which the qualifications of a King's minister are prescribed. One of them is that he must be a mowla, the meaning of which is given by Professor Monier Williams in his Dictionary as follows :-' hereditary,' 'holding office from father and grand-father. ' Vijnaneshwara explains the word to mean swa vamsha parampara yatan-i. e. descended from his own (good) vamsha. ' There vamsha is used to mean 'line of ancestors.' [page 93], Similarly Vijnaneshwara uses the word vamsha samachara to mean 'family custom, 'that is, usage handed down from one's ancestors, the same as kulachara. [See pages 77 and 80 of the same edition.] It is unnecessary to multiply further illustrations to show that the word-whether in the Hindu Law books or in ordinary parlance-generally means 'pedigree,' 'family,' or 'clan,' including in fact the line of ascendants as well as decsendants. Accordingly when a Hindu, whether his Vernacular be Marathi, Gujarati, or Canarese, speaks of preparing his pedigree or genealogy, he uses the word vamshavali, giving the line of his ascendants as well as descendants. This larger meaning of the word vamsha is admitted by all the witnesses of the appellant in cross-examination.
46. The word vamsha which occurs in the 6th clause must, therefore, in my opinion, be interpreted in the same sense-that is, as meaning ' the line of ancestors, ' or agnates, even if the expression mara bapika (my paternal)' governing it be translated as 'of my father's.' and not as ' my ancestral. ' Mara bapika vamshamanbhi would then run in English as follows : ' from the line of the ancestors of my father. '
47. Had the testator intended that his widow's power of adoption should be restricted to boys in the line of his father's descendants, he would have, I think, having regard to the surrounding facts, used another expression in the 6th clause of his will. As a matter of fact when the will was executed, that line consisted merely of the testator himself, his step-brother Dayabhai, and Dayabhai's only son. Upon the more limited construction contended for by the appellant, Dayabhai was the only descendant of the testator's father of whom the testator could possibly think as the only one from whose line the adoption could be made. If that was his intention, instead of saying, (which was a very simple thing to say) that his widow should adopt one of the sons of his brother Dayabhai, why did he use the expression paternal vams (sick), which he as a Hindu must be presumed to have known was capable of a wider meaning and could be so construed as to give the widow the power of selecting from a wider circle than that of his brother Dayabhai's sons? It was not a question here of selecting a boy from among several of the descendants of the testator's father. There was but only one descendant to whom the field of choice by the widow could be restricted and he was the testator's brother. The testator's father being dead, there was no possibility of any other line of descent coming into existence thereafter than that of Dayabhai. The testator could have plainly said, if he had intended it, that his widow should adopt a boy from that line, And he would have said that, had he intended it, because where a Hindu intends that the adoption to him by his widow should be of a boy from among his brother's sons, he generally says so and is not likely to use such an expression as mara bapika vamsha. This is a consideration which, I think, ought not to be lost sight of in construing the expression occurring in a Hindu will.
48. I agree, therefore, that Beaman J. is right and that his decree should be confirmed with costs.