[Scott, C.J:— The words joint family can be taken as one adjective qualifying the noun property: the alternative reading is, the word joint is an adjective referring to the compound noun family-property.]

The expression ‘joint family’ property means property belonging to the members of a family descended from a common ancestor and which has remained undistributed in their hands. It refers to joint property belonging to a family. The word joint is used in the sense of undivided. Does the use of this expression lead to the necessary result that Article 127 applies only to Hindus and only to such Mahomedans as are proved to have adopted Hindu customs of joint property ? I submit no.

First of all, look at the changed phraseology. In the Act of 1871, the Legislature used the word “Hindu”; and in 1877 changed it to “person” It is said that this changed phraseology is due to a desire to include within the section such Mahomedans as were known to have adopted Hindu customs of joint family. Fortunately these very enactments show that where the Legislature alters an enactment on that account it knows how to change its phraseology, e.g, Articles 125 and 141 of the present Limitation Act. Under the Mahomedan law, a female's estate is absolute and not restricted like that of a Hindu widow. These Articles, therefore, cannot and do not apply to all Mahomedan females. In the Punjab under a well-recognised custom existing among Mahomedan land-holders, widows take a restricted estate as in the case of Hindus: Wilson's Anglo-Mahomedan Law, 3rd edition, p. 87, In Act XV of 1877 these Articles bore the same numbers and the same phraseology; but in Act IX of 1871, they were numbered as Articles 124 and 142 and differently worded. They referred in terms to “Hindus” only. When the Legislature wanted to extend their operation in 1877, they did not supplant the term “Hindu” by the term “person” (as they did in Article 127); but they associated the words “or Mahomedan” with “Hindu” wherever it occurred. Thus in one case they take out the word “Hindu” and substitute “person” for it: in another case, they add “Mahomedan” to “Hindu”. The first change is wider in scope than the second one. Compare also the Article 107 where also from the expression “Hindu manager,” the word “Hindu” was omitted in 1877.

[Maeten, J.:— What would have been the effect if the Legislature had put “Hindu or Mahomedan” in place of “Hindu”?]

The Article would have applied to every Mahomedan. As the Legislature has used the word “person” it is necessary to consider the import of the expression “joint family property”

It is said the notion of a joint family is essentially a Hindu notion; the expression therefore must be interpreted in the Hindu sense. Let us see what is the Hindu sense of the expression? To begin with, the expression we have got to interpret is “joint family ˜ property” and not merely “joint family”. I would only point out that the argument on the other side would have been more forcible if the expression used were “property belonging to a joint family”. The expression to interpret would, in that event, be “joint family” What is the Hindu sense of the expression “joint family property”? Is it the Hindu family governed by the Mitakshara, or by the Dayabhaga ? For a joint Hindu family governed by the Mitakshara presents features essentially different from those presented by one governed by the Dayabhaga. I am trying to lead to this result: If you include Dayabhaga Hindus in the Article, then you will find it difficult to exclude Mahomedans. Take the case of a Hindu joint family consisting of two brothers, A and B and the sons of A and B. The family owns property which has descended from a common ancestor 0. The position under the Mitakshara is this: (1) The sons of A and the sons of B have an interest in the property by birth, i.e, every single member is the co-owner with the rest. (2) All these members take Over each other by survivorship. The main features therefore are: (1) property by birth; and (2) right of survivorship. Next, take the same family as governed by the Dayabhaga: both these features are absent. (1) As between father and sons, property in the sons arises not by their birth but by the death of the father (Mayne's Hindu Law, sections 37, 248, 324; Dayabhaga, C.I, pll. 11-30). (2) As between brothers and other collaterals, there is no right of survivorship; each holds his own interest in severalty (Mayne's Hindu Law, sections 37, 265, 373; Dayabhaga, C. XL, s. 1, pl. 26). The only features that are common to the two systems are: (1) property is undivided, i.e, there is community of interest; and (2) members derive their interest from a common ancestor. The state of a Dayabhaga joint family is a tenancy-in-common; each one has so to say a distinct share which on his death does not merge into the whole but would pass on to his own heir even to his widow. If then you include Dayabhaga Hindus in Article 127 why are Mahomedans (if they satisfy these two conditions) to be excluded?

The expression “joint family” is a loose expression in every sense of the term. There is no corresponding expression in Sanskrit texts, which use the word (undivided or unpartitioned). The expression is used by English writers to convey a certain idea. We say, it is not used in its technical sense in Article 127 but in common speech. Here, however, the expression we have got to construe is not “joint family” but “joint family property” There is reason to suppose that the Legislature intended “joint” to go more with “property” The expression “with property alleged to be joint” is used in Act XIV of 1859, s. 1, cl. 13 and it means no more than “joint estate of an undivided family,” employed in Article 107. If the Article is narrowly construed, it may be necessary to read into the enactment the words which are not there, e.g, suit by a person governed by the Hindu law of joint family or one who has adopted as a custom such law. Even there, you will have again to make provision for a Dayabhaga family.

What can then be the reason for the Legislature making a special provision in respect of “joint family property” in the Hindu sense. The cases arising in the other High Courts, on which the other side relies, show that in all parts of India, in numerous cases Mahomedan families do not distribute their inheritance for generations and they leave the management of the property to one or other member: see Vellai Mira Ravuttan v. Mira Moidin Ravuttan; Stokes' Anglo-Indian Codes, Vol. II, p. 947. Even a Hindu has to prove (1) the existence of a joint family property-and (2) his known exclusion from it, before he can seek assistance of Article 127. If a Mahomedan satisfies these requirements, why should the assistance of Article 127 be denied to him?

In this Presidency by far the larger proportion of Mahomedans are descendants of Hindus and therefore the Bombay view is more fitted to conditions of things existing here. Moreover, the Legislature from the very start used the expression rather in a loose sense; and never intended it to be confined in its application to Hindus alone. It has also used the expression “undivided family” in section 4 of the Partition Act (IV of 1893) and section 44 of the Transfer of Property Act (IV of 1882) as applicable to Hindus and Mahomedans alike: see Sultan Begam v. Debi Prasad.

The expression “joint family” has, as said above, been invented by English writers on Hindu law. The subject of joint family is treated of as a separate subject first in Mayne's Hindu law. Neither in the Mitakshara nor in the Mayukha, is the subject treated of separately; though ‘partition’ is so treated of None of the early writers, viz., Sir Thomas Strange (1864); Grady (1868); Cowell (Tagore Law Lectures, 1871); Bhattacharya (Tagore Law Lectures, 1865), have any separate chapter dealing with “joint family”.

Let us now briefly examine the history of the enactment and what the different High Courts have held. First, Act XIV of 1859. Cases decided here show that the expression “joint family property” was applied to Mahomedans in cases in which there is apparently no suggestion of a special custom. In Musst. Khy-roonissa v. Salehoonissa Khatoon no special custom was pleaded; and the expression “joint ancestral property” was promiscuously applied to a Mahomedan family. There is no suggestion of a special custom in Achina Bibee v. Ajeejoonissa Bibee; Chunder Monee Debia v. Meharjan Bibee and Bai Janubbi v. Mitha-bhai Gulabbhai. The next Act was Act IX of 1871. It had a very brief existence from April, 1873, to October 1877. It introduced the word “Hindu”. During this brief interval there were no reported cases; and there could be none because of the word “Hindu”

[Scott, C.J referred to Kali Kishore Roy v. Dhununjoy Roy.]

[Merten, J. referred to Devapa v. Ganpaya.]

Then came Act XV of 1877. It gave rise to a series of decisions in our High Court. The applicability of Article 127 to Mahomedans is fully discussed in Sayad Gulam Hussein v. Bibi Anvarnisa. Here, the family_ was a Sayed family; there was no allegation of a special custom; and no other Article was referred to. In Bavasha v. Masumsha the Judges do not give any reasons but they adopt the reasoning in Sayad Gulam's case. The principle is accepted tacitly in Faki Abas v. Faki Nurudin which relies on Bhaudin v. Shekh Ismal. The case of Sayadalli v. Aminbi is entirely in our favour. Then there is the case of Mulla Mahomood v. Khadizabibi decided by Starling J. on the 13th January, 1902.

In Abdul v. Mahomed, Batty J. does not say that custom must be proved before the Article can be applied to Mahomedans. No reference at all is made to the Dayabhaga doctrine. He says that previous joint possession is a condition necessary to its applicability — a point which has not been taken in any of the earlier Bombay cases. Those cases do not go on Article 144. The cases of Fatma Boo v. Ghisan Boo and Nyazgoolkhan v. Bibi Bu decided by Batchelor and Shah JJ., are recent instances of the application of Article 127 to Mahomedans. Thus for over thirty years our High Court has uniformly applied the Article to Mahomedans. The Bombay view is upheld in Tagore Law Lectures for 1882.

As for the cases of the other High Courts, I have already dealt with the early Calcutta cases. In Mahomed Akram Shaha v. Anarbi Chowdhrani, there is no reference to the Dayabhaga view and the judgment dwells only on the alteration in law: see Poyran Bibi v. Lakhu Khan Bepari

In Allahabad, in earlier cases at any rate, the High Court applied Article 127 to Mahomedans: see Sahib-un-nissa Bibi v. Hafiza Bibi; Hashmat Begam v. Mazhar Husain. Two years later came the Full Bench case of Amme Raham v. Zia Ahmad where that Court came to a contrary conclusion. In the last mentioned case, however, the property in dispute had already passed out of the family and Article 127 could not apply in any case.

The early Madras case of Khatija v. Ismail was governed by Act XV of 1877 and was a case of self-acquisition: Patcha v. Mohidin and Kasmi v. Ayishamma are against me; and so is Abdul Kader v. Aishamma. The case of Commercial Bank of India v. Allavoodeen Saheb has no application, for there the allegation was that the uncle and nephew had agreed to put their earnings into a common fund — which could be regarded as joint family property.

To sum up: the expression used in Article 127 is “joint family property” and not “property of a Hindu joint family” or “of any other joint family” In section 1, Clause 13 of Act XIV of 1859, the Legislature use a suggestive expression “the property alleged to be joint”. If we read the words rightly it seems the Legislature do not intend to restrict them to any technical meaning. Then, in Article 107, “joint family” and “undivided family” are used in the same sense. In other enactments also (e.g, Partition Act, section 4: Transfer of Property Act, section 44) the expression “undivided family” is used without restricting it to Hindus. Even among Hindus, the expression “joint family” does not always present the same features. A Mahomedan family, as set out in Sayad Gulam Hussein's case, does not materially differ from a Dayabhaga family. Lastly, the other view imputes to the Legislature either (1) ignorance of alleged well known facts, or (2) an impossible omission to provide for the entire Mahomedan population which may be counted not by tens of thousands but by millions. Assumptions like these will have to be made to support the other view. The Legislature have used the expression “joint family” as in common speech. This they have to do for an obvious reason. For, in the very nature of things they have got to make a few words go a long way to reach a variety of conditions. This Court, if it subverts its view now, can only legislate: see Dalton v. Angus. Thakor, in reply:— If the two tests of (1) community of interest, and (2) descent from a common interest were successfully applied, the Article would apply not only to Hindus and Mahomedans but to other communities like Parsis and Christians. It would also apply to a joint gift made by a Christian father to his sons; and to a partnership which brought together property inherited from a common ancestor. Article 123 applies to cases governed by the Indian Succession Act; but sections 187 and 190 of the Act are quite general. Section 190 applied only when the right of a dead person is put forward as the basis of the property: see Khadersa Hajee Bappu v. Puthen Veettil Ayissa Unmah

The Hindus are a community with whom the institution of joint family is unique. Such an institution is not known to exist so far as the Mahomedans are concerned.

[Marten, J.:— Among Mahomedans, the estate of the father descends and what the sons take on shares is the father's property. This does not hold true of a Hindu family.]

To constitute a joint family property there must be property of a joint family: see Obhoy Churn Ghose v. Gobind Chunder Dey. It is not suggested even in Bombay that the term ‘property’ is governed by “joint”.

[Marten, J.:— If you read “joint family property” as joint property of a joint family, that would be really in harmony with Article 107,]

Among Hindus there were texts special to each School of Hindu law; but the concept of joint family was common to both the Schools, viz., the, Mitakshara and the Dayabhaga. Hindus settling in Bengal accepted what was written in books peculiar to their School. But the institution of a joint family did remain; only one of its incident, survivorship, disappeared. On account of this difference, Hindus of Bengal did not become Mahomedans of Gujarat.

Mahomedans have, it appears, not infrequently lived together in a family and have their affairs carried on by a manager: see Mala Din v. Ahmad Ali; Baba v. Shivappa; Fakiruddin v. Abdul Hussein; Shakul Kameed Alim Sahib v. Syed Ebrahim Sahib; Durgosi Row v. Fakeer' Sahib; Nizam-ud-din Shah v. Anandi Prasad. But no presumption can on that account be made in their case that they form a “joint family”; Hakim Khan v. Gool Khan; and accretions to property held by them do not become joint family property: Abdool Adood v. Mahomed Makmil.

Scott, C.J (Batchelor, Macleod and Marten JJ. concurring):— The following question has been referred to a Full Bench.

Whether Article 127 of the Second Schedule of Act XV of 1877 can apply to the property of a Mahom-medan (or any other person not being a Hindu), and not having been proved to have adopted as a custom the Hindu law of the joint family?

The question referred is expressed to relate to the application of Article 127 of the Second Schedule of Act XV of 1877: the reference is, however, made in a suit brought when the Limitation Act of 1908 had come into operation. Therefore we take the question as relating to Article 127 of the later Act.

The exact point for determination is whether the three words “joint family property” as occurring in Article 127 of the Act of 1908 mean anything but property of a joint family in the sense in which the expression ‘joint family’ is understood amongst Hindus.

Of the three words so arranged two constructions are possible.

Either ‘joint family’ must be a compound adjective qualifying ‘property’ or ‘family property’ must be a compound substantive qualified by the adjective ‘joint’.

The word ‘joint’ may be construed in a technical legal sense or in a loose popular sense. In India ‘joint’ in the Article if used in a technical legal sense would attract to itself the word ‘family’ and form a compound adjective connoting “appertaining to a joint family” living as an undivided Mitakshara or an undivided Dayabhaga family lives. In a loose popular sense ‘joint family property’ might mean undivided property of a family.

The expression ‘joint family property’ is first to be found in the Indian Statute Book in clause 13 of section 1 of Act XIV of 1859. In Bengal (see Musst. Khyroonissa v. Salehoonissa Khatoon, Achina Bibee v. Ajeejoonissa Bibee and Chunder Monee Debia v. Meharjan Bibee and in Bombay (see Bai Janubbi v. Mithabhai Gulabbhai) effect was given to the expression in the Act of 1859 as embracing undivided property of members of Mahomedan families not shown to be joint families in the Hindu sense. That construction was only possible if ‘family property’ was read as a compound substantive. In the Act of 1871 the Legislature not only confined the words to an Article expressly dealing with Hindus' alone but coupled ‘joint’ to ‘family’ by the use of a hyphen. This hyphen appears in all copies and editions of the Act to which we have had access. In the limitation act of 1877 the expression ‘joint family property’ again appears in the same connection but the Article' 127 is no longer confined to Hindus but applies to any ‘person’ The intention to couple ‘joint’ with ‘family’ is again emphasised by a hyphen. This we have verified by reference to the first publication of the Act for general information in the Gazette of India for 21st July, 1877 after it had received the Viceroy's assent and also by reference to the official copies issued from the Government Press in Calcutta in 1878. It follows, we think, that ‘joint-family’ in Article 127 of the Act of 1877 must be read as a compound adjective and the expression ‘joint-family property’ must be read as property appertaining to a joint family.

This being so, we think Sayad Gulam Hussein v. Bibi Anvarnisa upon which a long series of Bombay cases is based was wrongly decided. There is good reason to believe that Sir Charles Sargent was misled by the Edition of West's Bombay Code then before him. That was the Edition of the Indian Statutes in general use by Judges of this Court in 1885. In West's Code Article 127 of the Act of 1871 is correctly printed with the hyphen but the corresponding Article of the Act of 1877 is incorrectly printed in that it omits the hyphen. The copy of the Act of 1877 then in Sir Charles Sargent's Court has the words ‘joint family property’ printed exactly as they were in the Act of 1859. Sir Charles Sargent's use of decisions under the Act of 1859 for the construction of the Act of 1877 is therefore quite intelligible; we do not think it would be intelligible if he had had the hyphened phrase before him.

In Article 127 of the Act of 1908 the wording is precisely the same as in the corresponding Article of the Act of 1877 but the hyphen is omitted. We have, however, for the period of thirty-one years from 1877 to 1908 the definite indication of the Legislature, as we understand it, that the words in the connection in which they appear should be used in a particular sense and we do not think the omission of the indicative hyphen after so many years should be taken as importing an intention that a changed construction should be adopted.

It only remains to consider the argument based on the working rule ‘constat stare decisis’. On the application of this rule we have two recent decisions of the Judicial Committee to guide us, namely, Pate v. Pate and Privy Council case (Tricomdas Cooverji Bhoja v. Sri Sri Gopinath Jiu Thakur), decided in 1916, to the effect that a long series of decisions based upon a clearly erroneous construction of an Act is not to be followed, while a long series of decisions based upon a construction not free from doubt should not be disregarded. There is no substantial reason for thinking that Mahomedan co-sharers who have not-yet sued to establish their right to share in family property relying upon the previous Bombay decisions under Article 127 of the Act of 1877 will be prejudiced by the narrower construction of the Act of 1908 now adopted, for Article 144 sets up for an excluding co-sharer practically the same period for prescription in the case of immoveable property as Article 127, while cases where undistributed moveable family property has existed since the death of the last owner for more than six years are so rare as to make the reduction of the prescriptive period of twelve years under Article 127 to six years under Article 120 a hardship which may safely be regarded as negligible.

We answer the question referred in the negative.

Beaman and Heaton, JJ.:— We only wish to add that certain passages in our referring judgment were written with special reference to the Mitakshara School which is predominant in this, as over the greater, part of India.

Under the Dayabhaga School of Bengal, rules, differing in some important points from those of the Mitakshara, regulate the constitution of the joint-family, more particularly in its legal relations to property. But it is true that the joint Hindu family, as a definite legal entity is as well known and recognized under the Dayabhaga as under the Mitakshara School, and so generally under the Hindu, as contra-distinguished from all other great systems of law. It is equally true, that the “joint family” as a legal entity involving definite legal notions, incidents and consequences is unknown to any other great system of law prevailing in the empire. So that where property is spoken of as belonging to a joint family, it must primarily be referred to a Hindu joint family, and only secondarily to such groups of non-Hindus as can prove that they have by custom adopted the Hindu law of the joint family.

Shah, J.:— Treating the question referred to the Full Bench as relating to the Limitation Act IX of 1908, I am of opinion that Article 127 of the First Schedule of the Act can apply to Mahomedans even though they may not be proved to have adopted as a custom the Hindu law of the joint family. I express no opinion — as it is not necessary to express any opinion — as to whether the Article can apply to persons other than Hindus and Mahomedans, who are not proved to have adopted as a custom the Hindu law of the joint family.

The Article in terms applies to a person excluded from joint family property to enforce a right to share therein. The answer to the question depends upon the construction to be placed upon the expression ‘joint family property’.

This expression was used in the Limitation Act (XIV of 1859), section 1, Clause 13; and the decisions under that clause show that it was held to apply to Mahomedans. The corresponding article in the act of 1871 was in terms confined to Hindus.

In the limitation act of 1877 the word ‘person’ was substituted for the word ‘Hindu’; and the Article as thus altered was held applicable to Mahomedans by Sargent C.J and Birdwood J. in the case of Sayad Gulam Hussein v. Bibi Anvarnisa. This view was followed in subsequent decisions: see Bavasha v. Masumsha; Sayadally v. Aminbi and Boo Fatma v. Boo Ghisanboo. The same view has been acted upon in other unreported cases in spite of the observations of Batty J. in Abdul v. Mahomed and of Beaman J. in Jan Mahomed v. Datu Jaffer.

After a careful consideration of the arguments on both sides, it seems to me that though there may be ground to prefer the view that the expression ‘joint family property’ in Article 127 is used in a technical and Hindu sense, as interpreted by the other Indian High Courts, and not in the broad and non-technical sense, in which it has been interpreted by different Judges of this Court, the construction hitherto put upon it by this Court is a reasonably possible construction of the expression. At any rate I am not satisfied that it is wrong. As this point relates, to a rule of limitation affecting title to property, I am of opinion that the view taken by Sargent C.J in Sayad Gulam Hussein's case in 1885 should be adhered to. It seems to me that any departure from this long course of decisions is likely to result in injustice and hardship.

In the Act of 1859 there was no hyphen between the words ‘joint’ and ‘family’. In the Act of 1871 there was a hyphen between these words. In 1877 though the Article was altered as already stated the hyphen was retained. In the Act of 1908 the hyphen is omitted. This account of the use of a hyphen is interesting; but in my opinion it is not safe to base any conclusion upon it. I see no sufficient reason to assume or to infer that the learned Judges who decided Sayad Gulam Hussein's case had not noticed the hyphen between the words ‘joint’ and ‘family’ in the Act of 1877. The hyphen was certainly brought to the notice of the Court in Sayadalli's case in 1898, when Eanade J. observed that the use of the hyphen by itself did not justify any such restrictive interpretation. If the Article with the hyphen in the Act of 1877 could be applied — and I am not prepared to hold that the learned Judges were wrong in applying it — to Mahomedans generally in this Presidency, I see no reason why the Article without the hyphen in the Act of 1908 should be held to be inapplicable to them.

I see no insuperable difficulty in interpreting the word ‘joint’ as meaning ‘undivided’; and if the word be understood in that sense the view taken in Sayad Gulam Hussein's case and followed in subsequent cases becomes easily intelligible and acceptable.

In view of the judgment of my Lord the Chief Justice, which I have had the privilege of reading, I do not consider it necessary to state my reasons in detail for the conclusion that the expression ‘joint family property’ is susceptible of the construction put upon it in Sayad Gulam Hussein's case.

I sincerely regret that I am unable to agree with my Lord the Chief Justice and my other learned colleagues on this question.

Answer accordingly.

R.R