Arthur Strachey, C.J.
1. This is a, reference to the Court under Section 9 of the Indian Divorce Act (IV of 1869), by the District Judge of Agra, of a question arising in a suit for dissolution of marriage pending in this Court. The suit was brought by the husband as petitioner against his wife, the respondent, and against a co-respondent. In the course of the hearing, counsel for the co-respondent contended that the petition must be dismissed on the ground that the respondent had never been lawfully married to the petitioner. It appears that the respondent had formerly gone through a ceremony of marriage with one Elloy. On the 18th July 1888, she obtained a decree declaring that marriage null and void in the Court of the Judicial Commissioner of Oudh, who, until the passing of Section 42 of Act XX of 1890, was 'a District Judge,' within the meaning of Section 3, Clause 2 of the Indian Divorce Act, for Oudh. On the 7th December 1888, the decree of nullity was confirmed by this Court, which had jurisdiction over Oudh under Section 3, Clause 1 of the Act. On the 21st December 1888, the respondent was married to the present petitioner. The contention now raised by the counsel for the corespondent is that under Section 20, read with the proviso in Section 17, the High Court was not competent to confirm the Judicial Commissioner's decree of nullity until after six months from the pronouncing thereof; that the order of confirmation, having been made less than six months from the date of the decree, must be held to be illegal and void; that, therefore, the decree must be treated as not having been validly confirmed; and that, consequently, the subsequent marriage of the respondent with the petitioner was also illegal, and cannot be made the subject of a decree for dissolution of marriage. The District Judge has accordingly stayed the proceedings pending a reference under Section 9 of the question whether, having regard to the facts just stated, the marriage sought to be dissolved was a valid marriage.
2. There are two questions to be considered. The first is, whether the High Court's decree of the 7th December 1888, was in contravention of Section 20, read with Section 17 of the Act. The second is whether, if so, it follows that that decree was void and inoperative as a confirmation of the Judicial Commissioner's decree of the 28th July 1888. In regard to the first point, reliance is placed on the decision of the High Court of Bombay in A v. B (1898) I.L.R. 23 Bom. 460. In that case no question arose as to the effect of a confirmation made by the High Court before the time, if any, prescribed by the Divorce Act. It was a submission by a District Judge of a decree for nullity for confirmation under Section 20, upon which the petitioner applied to the High Court for immediate confirmation. The Court held that it could not confirm the decree before the expiration of six months from the pronouncing thereof, and so rejected the application with leave to renew it when the six months' period had expired. We have first to consider whether we agree with the construction placed by the Bombay High Court upon Sections 17 and 20 of the Act. After the fullest consideration I am unable to agree with it. Section 20 provides that 'every decree of nullity of marriage made by a District Judge shall be subject to confirmation by the High Court, and the provisions of Section 17, Clauses 1, 2, 3 and 4 shall, mutatis mutandis, apply to such decrees.' Section 17 provides for the confirmation by the High Court of decrees for dissolution of marriage made by a District Judge. It consists of six paragraphs. The fifth paragraph is as follows: 'Provided that no decree shall be confirmed under this suction till after the expiration of such time, not less than six months from the pronouncing thereof, as the High Court by general or special order from time to time directs.' If this fifth: paragraph is the fifth 'Clause' of Section 17 within the meaning of Section 20, then Section 20 does not make it applicable to decrees of nullity of marriage made by a District Judge, and such decrees, therefore, need not wait for six months, but may be confirmed at once. If, though the fifth paragraph, it is to be regarded as the fourth 'Clause' of Section 17 within the meaning of Section 20, then Section 20 makes it applicable to decrees of nullity of marriage, and such decrees, like decrees for dissolution of marriage, cannot be confirmed till after the expiration of six months from the pronouncing thereof.
3. Now the word 'Clause' used in Section 20 is nowhere defined in the Act. The paragraphs into which Section 17 is divided are not numbered, and so far as the form of the Section is concerned there is nothing to suggest that one paragraph is more or less a 'clause' than another, or that the fifth paragraph is not a clause. It is rather difficult to gather from the judgments in the Bombay case what the learned Judges considered to be the exact relation between the proviso and the other paragraphs of Section 17. Mr. Justice Parsons says: 'The fifth paragraph is not, in my opinion, a Clause of the section. It is a proviso to the Clause which precedes it, joined to it as printed in the Government of India (Legislative Department) edition 1887 of the Acts, by a colon, and must be considered to be a part and parcel of the foregoing clauses, governing and controlling them, and not forming itself a separate clause.' Mr. Justice Ranade says: 'The proviso appears as a separate paragraph, but it is clear from its context that it cannot be read as a separate Clause from paragraph 4, which it qualifies. It does not, as the preceding four paragraphs, or the succeeding sixth paragraph, relate to distinct subject-matters.' Mr. Justice Fulton holds that 'the proviso governs and forms part, of the fourth clause.' Now if it is correct to say that the fifth paragraph is 'a proviso to the Clause which precedes it' and 'forms part of' that Clause and cannot be read as separate from that clause, it seems contradictory to say, as Mr. Justice Parsons goes on to say, that it 'must be considered to be a part and parcel of the foregoing Clauses governing and controlling them.' Apart from this, it appears to me quite impossible to hold that the proviso merely forms part of the Clause immediately preceding it. That Clause relates only to oases in which the District Judge has, upon the direction of the High Court, made further inquiry or taken additional evidence. That is clearly shown by the word 'thereupon.' If the proviso merely formed part of that clause, it would follow that it did not apply to the far more numerous cases in which no further inquiry or additional evidence is required, and the result would be that, contrary to the obvious intention of Section 17, the vast majority of decrees for dissolution of marriage might be confirmed at once. As then the proviso clearly applies to case3 not falling within the preceding clause, it cannot merely form part of that clause; and if it does not merely form part of any one of the previous clauses, but governs and controls each, there can be no reason for not regarding it as itself a clause. Upon similar reasoning to that of the Bombay High Court, it would be logical to hold that paragraph 4 also was not a separate clause, as, notwithstanding the observation of Mr. Justice Ranade to the contrary, it also does not 'relate to distinct subject-matters:' it is merely consequential to Clause 3; and there is, in my opinion, more reason to hold that paragraph 4 forms part of Clause 3, and is therefore not a separate clause, than to hold that the proviso forms part of Clause 4. In regard to Mr. Justice Parsons' argument based on the colon at the end of paragraph 4, the Privy Council in The Maharani of Burdwan v. Krishna Kamini Dasi (1886) I.L.R. 14 Cal. 365, at page 372 of the report say (in accordance with many English authorities) that 'it is an error to rely on punctuation in construing Acts of the Legislature.' The soundness of this principle is well illustrated in the present instance by the fact that in the original edition of the Indian Divorce Act (see that Gazette of India, March 6th, 1869, p. 375), there is not a colon at the end of the fourth paragraph of Section 17, but a full stop. Mr. Justice PARSONS proceeds to give illustrations from the Code of Criminal Procedure in support of the proposition that if in Section 17 of the Indian Divorce Act the Clauses had been numbered, the proviso would not have been numbered as a clause. When the Sections of the Code to which he refers--Sections 33, 35, 48, 57 and 123--are looked at, I think it clearly appears that they establish no such proposition. In every one of those Sections it is obvious from the context that the proviso was intended to apply to, and govern the immediately preceding proposition only, and that to mark this the proviso was not separately numbered. But, for the reasons which I have just given, it is impossible to hold that the proviso in Section 17 was intended to apply to and govern the fourth paragraph only. I agree with Mr. Justice Ranade, that from the point of view of considerations of expediency or public policy, such as the interests of children, the prevention of collusion, and so forth, decrees for dissolution and decrees of nullity should stand on the same footing. But the question is whether that was the view of the Legislature in 1869 when the Indian Divorce Act was passed. So far as collusion is concerned, it certainly was not. It is obvious from Section 20 that the Legislature deliberately excluded from the case of decrees of nullity the last paragraph of Section 17, authorizing intervention on the ground of collusion during the progress of a suit for divorce in the District Court. Further, in regard to suits tried by the High Court in its original jurisdiction, whereas under Section 16 a decree for dissolution must, in the first instance, be a decree nisi, not to be made absolute for at least six months, during which period any person may show cause why the decree should not be made absolute by reason of collusion, or concealment of material facts; on the other hand, a decree for nullity under Section 18 is made absolute at once, and there is no provision for intervention. Again, in England in 1869 the same distinction obtained, and it was not until the passing of the Matrimonial Causes Act, 1873, that decrees for dissolution and decrees of nullity were assimilated in respect of confirmation and intervention. It cannot therefore be argued that there was in 1869 any a priori probability or presumption that because a decree for dissolution made by a District Judge had to wait for confirmation for six months, therefore the Legislature considered a similar delay as appropriate for the confirmation of decrees of nullity. Mr. Justice RANADE in connection with the Matrimonial Causes Act, 1873, relies on Section 7 of the Indian Divorce Act, which provides that 'subject to the provisions contained in this Act, the High Courts and District Courts shall, in all suits and proceedings hereunder, act and give relief on principles and rules which, in the opinion of the said Courts, are as nearly as may be conformable to the principles and rules on which the Court for Divorce and Matrimonial causes in England for the time being acts and gives relief.' In the first place, that Section is 'subject to the provisions contained in this Act.' This shows, I think, that the principles mentioned in the Section are only applicable in the absence of express provisions in the Act: they cannot be applied to construe the provisions contained in the Act, such as Sections 17 and 20, or to extend or restrict the operations of those provisions. In Abbott v. Abbott (1869) 1 B.L.R. 51, Mr. Justice Macpherson held that 'Section 7 of the Divorce Act applies not to points of procedure, but to the general principles and rules on which the Court is to act and give relief.' Sections 17 and 20 relate to 'points of procedure' only. In A. v. B. (1898) I.L.R. 22 Bom. 612, it was held by Sir Charles Farran C.J., and Mr. Justice Tyabji, that the principles and rules referred to in Section 7 were not mere rules of procedure, such as the rules which regulate appeals: and I think that the same may be said in reference to the rules which regulate confirmation, especially when it is remembered that in England there is nothing which precisely corresponds to the matrimonial jurisdiction of a District Court in India, or the confirmation of the decrees of those Courts by the High Court. I understand the practice of this Court to have been in accordance with the view that Section 20 of the Divorce Act does not make the proviso in Section 17 applicable to the confirmation of decrees of nullity made by a District Judge. I see no reason to think that this practice is wrong, and I am therefore of opinion that this Court had power, on the 7th December 1888, to confirm the Judicial Commissioner's decree of nullity of the 28th July 1888.
4. The next question is, assuming that by reason of the proviso in Section 17 the High Court ought not to have confirmed the Judicial Commissioner's decree until after the expiration of six months from the pronouncing thereof, does it follow that the confirmation was null and void, and the subsequent marriage of the respondent with the petitioner invalid? The District Judge in his reference assumes that the answer to this question must be in the affirmative; but he gives no reasons, and I cannot agree with him. The decree of the High Court of the 7th December 1888, was a decree of the kind specified in Section 41 of the Indian Evidence Act, 1872. It was a final decree made in the exercise of matrimonial jurisdiction, declaring the present respondent not to be the wife of the then respondent. If it was the decree of 'a competent Court,' then, however erroneous or irregular it may have been, it is under the Section conclusive proof that the respondent's previous marriage was a nullity. The effect of such conclusive proof can only be avoided by showing that the High Court was not 'a competent Court' within the meaning of Section 41, or was 'a Court not competent to deliver' the decree within the meaning of Section 44. Unless that can be shown, the decree is conclusive, as no fraud or collusion is suggested. The question then is, was the High Court's decree of the 7th December 1888, 'delivered by a Court not competent to deliver it? It appears to me that this question must be answered in the negative. The High Court had undoubted jurisdiction in the suit for nullity of marriage. As regards place, it possessed the local jurisdiction defined by the Act. It possessed personal jurisdiction over the parties to that suit who were persons governed by the Divorce Act; and it had jurisdiction over the subject-matter, or the class of suit as disclosed in the petition for declaration of nullity. It was properly seised of the case, which was duly transmitted to it by the Court of the Judicial Commissioner, and notice of the date fixed for confirmation was duly served upon the parties, of whom the petitioner was represented at the hearing by a pleader. There was no appeal to Her Majesty in Council from the decree of confirmation, as there might have been under Section 58. Since the High Court had jurisdiction in the suit, it follows that it had jurisdiction to consider and determine every question of law or fact arising in the suit. This would of course include any question of procedure, such as a question of the construction of Sections 17 and 20 of the Indian Divorce Act. To illustrate this, let us suppose that at the hearing either the petitioner or the respondent had formally taken the objection that an adjournment was necessary, as under the proviso in Section 17 the decree could not be confirmed until the six months' period had expired. Suppose further that, after full argument on the point, the High Court had taken a view of Section 17 different from that expressed in the Bombay case, and had confirmed the decree of the Judicial Commissioner accordingly. In such a case surely the Court would not only be competent but bound to decide the question thus raised and argued. If competent to consider and decide the question, it cannot be supposed that the Court was 'competent' to decide it in one particular way only. This shows that even if the decision was erroneous or irregular, the Court was nevertheless 'competent to deliver' it. If not, what is the alternative? Could any Court, however subordinate, in any subsequent suit, at any distance of time, treat the High Court's decree as a nullity and the parties still husband and wife? For instance, could a creditor successfully sue the former husband in a Small Cause Court for the price of necessaries supplied to the wife after the decree, on the ground that the decree was void, as the High Court had taken an erroneous view of the proviso in Section 17? Again, after the High Court's decree, could either of the parties re-marrying be prosecuted for bigamy and the children of the subsequent marriage be held illegitimate? If these conclusions would be absurd where the High Court decided the question of the construction of Section 17 after argument, they must equally be so in a case like the present. The competency or jurisdiction of the Court cannot possibly depend on whether a point which it decides has been raised or argued by a party or counsel. An express decision upon the construction of Sections 17 and 20 and an implied decision must stand on the same footing. The view that the decree was a nullity by reason of the proviso in Section 17 could only be supported on the principle that wherever a decision is wrong in law, or violated a rule of procedure, the Court must be held incompetent to deliver it. Such a principle is obviously unsustainable. In the first place, it is opposed to the language of Sections 41 and 44 of the Evidence Act, which were undoubtedly meant to make the decrees which they refer to conclusive except in a very restricted class of cases. If the intention had been to make such decrees questionable on the ground of any legal defect or irregularity, very different expressions would have been used, and it would be inaccurate to describe such decrees as constituting 'conclusive proof.' In the second place, If the principle were sound, any judgment might be collaterally attacked by contending: that it was in violation of such rules of procedure as the rule of res judicata contained in Section 13 of the Code of Civil Procedure, or the rule of limitation contained in Section 4 of the Limitation Act, 1877. These rules are expressed in language as peremptory as that of the proviso in Section 17 of the Divorce Act; but it has never been held, and it could not be held, that a Court which erroneously decrees a suit which it should have dismissed as time-barred, or as barred by the rule of res judicata, acts without jurisdiction and is not competent to deliver its decree. The insecurity of titles and of status arising from the adoption of such a principle is just what Sections 41 and 44 of the Evidence Act were intended to prevent. The Sections recognize that, given the competency of the Court, even error or irregularity in the decision is a less evil than the total absence of finality which would be the only alternative. In the third place, the judgment of the Privy Council in Amir Hasan Khan v. Sheo Bakhsh Singh (1884) I.L.R. 11 Cal. p. 6 : L.R. 11 I.A. 237, shows that, even for the purposes of direct attack in revision under Section 622 of the Code of Civil Procedure, a decree cannot be held to have been made without jurisdiction or illegally, merely because it is wrong in law or alleged to be in violation of such rules of procedure as those contained in Sections 13 and 43 of the Code. If so, then a fortiori such a decree could not be regarded as made without jurisdiction for the purposes, not of direct but merely collateral attack in a subsequent suit. In Sardarmal Jagonath v. Aranvayal Sabhapathy Moodliar (1896) I.L.R. 21 Bom. 205, a judgment-creditor sought to maintain an attachment on the property of his debtor who had previously been adjudicated an insolvent by the Madras Insolvent Court, and to resist a claim by the Official Assignee, under Section 278 of the Code, for the release of the property from attachment, on the ground that the order of adjudication and the vesting order were null and void, and gave no title to the Official Assignee, inasmuch as the original petition to the Insolvent Court disclosed no act of insolvency on which an order of adjudication could legally be passed under the Statute. I held that as the Madras Court was undoubtedly competent to deal with the petition, and was both competent and bound to consider whether the acts alleged in the petition constituted acts of insolvency within the meaning of the Statute, the order, even if wrong in law, was not one which the Madras Court was not competent to deliver within the meaning of Section 44 of the Evidence Act, and that therefore it could not be treated in collateral proceedings as null and void, but was conclusive of the insolvency and of the Official Assignee's title. At page 214 of the Report, I said: 'Once recognize that a Court is competent to decide a suit or a petition in insolvency or any other matter, and it follows that it is competent to decide all questions which arise in that matter, whether they are questions of fact or of law, and whether they appear on the face of the plaint or petition or arise subsequently. If it decides them wrongly, its decision may be subject to reversal on appeal or otherwise, but cannot be treated as a nullity.' The same principle is, I think, recognized in the judgment of Mr. Justice Knox and Mr. Justice Aikman in Durga Prasad v. Mahabir Prasad Weekly Notes 1899 p. 199. The English, Indian and American authorities collected in Mr. Hukm Chand's learned Treatise on the Law of Res Judicata, Chap. VII, Sections 186, 187, 189,190 and 192, establish that for the purpose of showing in collateral proceedings that a judgment is void for want of jurisdiction or competency in the Court, it is not sufficient to show error in law, irregularity in practice, or departure from the provisions of the law of procedure, as for instance, by taking the proceedings at a wrong or unauthorized time. In one American case cited at p. 475, it was said 'the principle is so well settled that it is said to be an axiom ofthe Law, that when a Court has jurisdiction over the subject-matter and the parties, its judgment cannot be impeached collaterally for errors of law, or irregularity in practice.' In another American case cited at page 476, it was said: 'Jurisdiction having been obtained, the fact that the judgment was rendered sooner than it should have been, does not make the judgment void: a judgment thus rendered is irregular only.' The whole subject is elaborately discussed by a learned American author, Mr. Vanfleet, in his work 'The Law of Collateral Attack on Judicial Proceedings' (see especially Chapter VIII), In Chapter XIV, Sections 710, 711, 712 and 713, the author gives instances to show that a 'premature judgment,' that is, a judgment given before it ought to have been given according to the law of procedure, cannot therefore be treated in collateral proceedings as void and given by a Court without jurisdiction. 'An administrator's order to sell land could not be granted lawfully until after the final account of the personal assets had been settled; but an order granted before that had been done is not void. The Missouri Statute required the Court to delay the approval of an administrator's or guardian's sale of land until the next term after it was made, but such a sale is not void because approved at the same term or an adjourned term.'
5. For these reasons I would answer the reference by saying that, in our opinion, the marriage of the respondent with the petitioner was not invalid by reason of any want of jurisdiction in the High Court's decree of the 7th December 1888.
6. I desire to repeat what I stated at the hearing, that the Court is much indebted to Mr. Porter and Mr. Sorabji, who appeared as amici curiae for the co-respondent and the petitioner, respectively, for the assistance rendered to the Bench by their very able argument.
7. I fully concur both in the reasons and in the conclusions arrived at by the learned Chief Justice, and have nothing further to add.
8. I also entirely concur in the conclusions arrived at by the learned Chief Justice and in the reasoning on which those conclusions are based. I have only one addition to make. It is that, in my opinion, the judgment of a Bench of this Court confirming the decree for nullity of marriage is an authority on the question of law whether for the validity of such a confirming order a delay of six months is necessary. The Bench which implicitly decided that the six months' delay imposed in cases of dissolution of marriage was not necessary in cases of nullity was a Bench similarly constituted to the present, and of co-ordinate authority; and, if not by strict law, by the comity of the Courts, the law in such a decision ought to be taken as authoritative until declared to be erroneous by a Full Bench of the Court. A fortiori it was not open to an inferior Court to question the decision of any Bench of this Court. It is impossible to draw the inference which appears to be suggested by the District Judge that the matter was not considered and decided by the Bench of this Court which confirmed the decree of nullity. It was necessary as a foundation for the order which it made that it should have adjudicated on that question and decided that the six months' delay was not in that case imposed by the law. Therefore on authority as well as on the reasoning set forth in detail in the learned Chief Justice's judgment I would make the same answer to this reference.