Axon v Axon

[1937] HCA 80

59 CLR 395

COMPLAINANT, AXON

ON APPEAL FROM THE SUPREME COURT OF

SOUTH AUSTRALIA. MaintenanceHusband and wife-Marriage-Validity-Burden of proof-Pre-

sumption of validity-Previous marriage by wife-Previous husband not heard of for over seven years-Presumption of death-Maintenance Act 1926-1936 (S.A.) (No. 1780-No. 2279), sec. 66.

A and B went through the ceremony of marriage on 6th January 1932. On 22nd August 1937 A lodged a complaint against B for maintenance under the Maintenance Act 1926-1936 (S.A.), and an order was made in her favour. At the hearing A proved her marriage by the production of the marriage cer- tificate. On 15th July 1911 A had married C, and had lived with him until 1923, when he left her and went to Broken Hill. She obtained an order for maintenance against him, but no payment was ever made, and she did not see him again. She heard that he was dead. Before her subsequent marriage, she inserted advertisements in several newspapers and made inquiries from the police at Broken Hill, but without result. B claimed that his marriage to

A was not valid. He called evidence to show that C had been alive in 1936, but there was a doubt as to the identity of the man who was said to be C. The magistrates held that strict proof should be given by B to show that C was alive at the date of A's second marriage and that he had not discharged

Held, by the whole court, that the burden was on A to prove that she was the lawful wife of B: but, by Dixon and Exatt JJ., that the burden was sup- ported by the presumption of the validity of the marriage of A to B, and, by

59 CLR 396

Evatt J., by the further presumption, based on the law of South Australia relating to bigamy, and arising from C's continued absence for seven years before A's second marriage, that C had died before that marriage; and, further, by Diron J., that, if the magistrates were not satisfied, on the balance of probabilities, after giving due consideration to the presumption of the continu- ance of C's life, that C was alive at the date of A's second marriage, that mar- riage must be upheld as valid, and, by Evatt J., that, A being entitled to rely on a presumption that C was dead at the date of A's second marriage, it was for B to adduce evidence to rebut that presumption.

Held, further, by the whole court, that in the circumstances the case should be remitted to the magistrates for rehearing.

Decision of the Supreme Court of South Australia (Richards J.) affirmed, but (by Dixon and Evatt JJ.) on different grounds.

APPEAL from the Supreme Court of South Australia.

Mary Ann Axon took proceedings against Edwin Lewis Axon, claiming maintenance under the provisions of the Maintenance Act 1926-1936 (S.A.). The complaint, which was laid on 22nd August 1937, alleged that the defendant (hereinafter called " the husband ") had deserted the complainant (hereinafter called " the wife ") on and since 16th March 1937, and had wilfully neglected to provide reasonable maintenance for her on and since 25th March 1937. It was held by the court of summary jurisdiction hearing the complaint that these grounds were established. The wife proved, by the production of the marriage certificate, that she had gone through the ceremony of marriage with the husband on 6th January 1932. In cross-examination she admitted that she had been married on 15th July 1911 to one Mauro Herzich and that she had lived with him at Port Pirie until 1923. In that year he left her and went to Broken Hill. She took proceedings against him for maintenance on the ground of desertion and obtained an order against him. Herzich was brought down by the police to Port Pirie from Broken Hill, but left the next morning by train. He made no payment under the order, and the wife had not seen him since, nor had she heard of his being alive. She was informed by a Broken-Hill resident (since dead) that he was dead. She placed advertise- ments in the newspapers The Chronicle, The Advertiser and Truth, and made inquiries from the police at Broken Hill, but she obtained no information. These inquiries were made before her marriage to

59 CLR 397

the defendant. No certificate of Herzich's death and no other witness who might have been expected to hear of or from Herzich was called to say that he had not SO heard. The husband produced a witness, Chesson, who swore that, in December 1936, Herzich called at his home at Port Pirie in a car. Another witness, Edwards, said that he was the driver of the car which conveyed the visitor to Chesson's but that he did not know him as Herzich, but only as Herzog, and that the man lived at Birrell Street, Broken Hill. Before judgment was delivered, the magistrate hearing the case received a letter purporting to come from Beryl Street, Broken Hill, and purporting to be signed by Herzog. This letter stated that the writer was not Herzich, and that he was Herzog and no one else, and stated that Chesson's evidence was embarrassing to his wife and himself. In the judgment no attention was paid to this com- munication, but it was made known to both parties. The court found that the complainant was the wife of the defendant by virtue of the marriage ceremony performed in 1932, and stated:- After full consideration of the whole of the evidence a substantial doubt remains in the minds of the court as to whether the man Herzog was in fact identical with Herzich. In the opinion of the court strict proof is required of the existence of Herzich if the defence set up in this case is to succeed, and the defendant has failed to discharge the onus which lies upon him in this regard to the satisfaction of the court." An order was therefore made that the husband pay certain sums by way of maintenance for the wife.

The husband appealed to the Supreme Court of South Australia. It appeared to Richards J. that the matter had been dealt with on a mistaken assumption as to the onus and degree of proof required. He therefore set aside the order and remitted the complaint for hearing in order that the wife should have an opportunity to give further evidence, if she could, on all the elements necessary to estab- lish her case.

From that decision the wife, by leave, appealed to the High Court. Newman, for the appellant. The onus is on the defendant to show that the complainant's former husband was alive at the date of her marriage with him (the defendant). There is a difference between

59 CLR 398

the onus of proof required from a woman suing for maintenance

and that in other proceedings in which valid marriage must be proved (Spivack v. Spivack 1 ). There is a strong presumption law as to the validity of a proved marriage (See Phipson on Evidence, 7th ed. (1930), pp. 32, 653, 667). The particular reason why it may not be valid cannot alter the presumption. The validity of the marriage is not sufficiently impugned by the defendant's evidence (Kempton v. Public Trustee 2, applying Piers v. Piers 3 Harrod V. Harrod 4; Ousley v. Ousley 5 Mackay v. Mackay 6 ). The onus of proving survivorship rests upon the party asserting it (Re Green's Settlement 7; Re Benajmin Neville v. Benjamin 8 ). There is no absolute presumption of law as to the continuance of life (Lapsley v. Grierson 9 ). It would be too much against the weight of the evidence to presume that the former husband was alive at the date of the marriage between the complainant and defendant. [Counsel also referred to Carlin v. Carlin 10, comparing it with Haviland v. Mortiboy 11; Irish Society v. Derry 12, and distinguished the following cases Lapsley v. Grierson 13 R. V. Harborne 14 Hogton v. Hogton 15; Ivett v. Ivett 16; Deakin V. Deakin 17.] Even if there was an onus cast on the complainant of showing that on the balance of probabilities her former husband was dead at the date of her marriage to the defendant, her evidence is sufficient to throw the onus of proof of showing that he was alive back on the defendant (In re Phené's Trusts 18 ).

Alderman, for the respondent. A party who affirms must prove the allegation. The relevant allegation here is that the defendant is the husband of the complainant. Although the complainant is able to call in aid certain presumptions of fact, in the result she

1(1930) 99 L.J.P. 52. 2(1932) N.Z.L.R. 1380. 3(1849) 2 H.L.C. 331 ; 9 E.R. 4(1854) 1 K. &J. 4 69 E.R. 344. 5(1912) V.L.R. 32 33 A.L.T. 155. 6(1901) 18 W.N. (N.S.W.) 266. 7(1865) L.R. 1 Eq. 288. 8(1902) 1 Ch. 723. 9(1848) 1 H.L.C. 498 9 E.R. 10(1906) 70 J.P. 143. 11(1859) 32 L.T. (O.S.) 343. 12(1846) 12 Cl. &Fin. 641 ; 8 E.R. 13(1848) 1 H.L.C. 498 ; 9 E.R. 14(1835) 2 A. &E. 540 ; 111 E.R. 15(1933) 150 L.T. 80. 16(1930) 143 L.T. 680. 17(1869) 33 J.P. 805. 18(1870) 5 Ch. App. 139. 59 CLR 399

must satisfy the court with the proper degree of certainty of the fact she alleges (See Thayer on Evidence at Common Law (1898), pp. 313 et seq., and Apothecaries Co. v. Bentley 1 ). The presumption of death from unexplained absence is only a presumption of fact (See Roscoe, Evidence in Civil Actions, 20th ed. (1934), vol. I., pp. 35, 44 Taylor on Evidence, 11th ed. (1920), vol. I., p. 114; R. v. Harborne 2 In re Phene's Trusts 3 ). The presumption does no more than affect the shifting onus. There is no case in which the pre- sumption has been held to outweigh any acceptable evidence of continued existence. The defendant does not rely on the fact that the previous husband is alive, but tenders his evidence to disprove the complainant's allegation that she is his wife. The court of summary jurisdiction never directed its attention to the question whether the complainant had proved that she was the defendant's wife it regarded the issue only as a question whether the defendant had proved beyond reasonable doubt that he was not her husband. The degree of proof required by the court was too great. The evidence called by the complainant does not prove circumstances from which it is proper to draw the inference of death. It is sub- mitted that a new trial should be ordered in any event because the evidence is SO unsatisfactory.

Newman, in reply. The following written judgments were delivered :-

LATHAM C.J. Mary Ann Axon took proceedings against Edwin Lewis Axon claiming maintenance under the provisions of the Maintenance Act 1926-1936 (S.A.). Sec. 66 of that Act provides, inter alia, that any married woman whose husband during the preceding six months has been guilty of desertion or of wilful neglect to provide for her may apply for summary protection under the Act and that the court may order such protection. The complaint, which was laid on 22nd August 1937, alleged that the defendant

1(1824) Ry. &M. 159 ; 171 E.R. 2(1835) 2 A. &E. 540 ; 111 E.R. 3(1870) 5 Ch. App. 139. 59 CLR 400

A. had deserted the complainant on and since 16th March and had

wilfully neglected to provide reasonable maintenance for her on and since 25th March 1937. It was held by the justices that these grounds were established. The question which arises is whether the justices were right in holding that the complainant had estab- lished that she was the wife of the defendant.

The complainant proved that she was married to the defendant on 6th January 1932. If no further evidence had been given, she was entitled to an order. But in cross-examination she admitted that she had been married on 15th July 1911 to one Mauro Herzich. She lived with him at Port Pirie until 1923. In that year he left her and went to Broken Hill. She took proceedings for maintenance on the ground of desertion and obtained an order for maintenance against him at the rate of £4 per fortnight. She received no payment under the order, and she has never seen him since. She heard from some unnamed person that he was dead. She placed advertisements in the newspapers The Chronicle, The Advertiser and Truth. and made inquiries from the police at Broken Hill. She obtained no information as the result of these advertisements and inquiries. She did not produce any certificate of Herzich's death. No other witness who might have been expected to hear of or from Herzich was called to say that he had not SO heard. On the other hand, the defendant called evidence directed to show that Herzich had been seen at Port Pirie in 1936. If this evidence had been accepted, it would have concluded the matter. The justices, however, said that doubts arose as to whether the person seen in December was in fact Herzich. One witness said that this man's name was Herzog. The justices said After full consideration of the whole of the evidence a substantial doubt remains in the minds of the court as to whether the man Herzog was in fact identical with Herzich. In the opinion of the court strict proof is required of the existence of Herzich, if the defence set up in the case is to succeed, and the defendant has failed to discharge the onus which lies upon him in this regard to the satisfaction of the court." They said: 06 It has not been proved to the satisfaction of the court that Mauro Herzich, the former husband of the complainant, was alive at the date when her marriage to the defendant Axon was celebrated."

59 CLR 401

There can be no doubt that a wife claiming maintenance against a defendant as her husband, must prove that she is his wife. In this case the complainant gave evidence which, if unchallenged by other evidence, would have established this proposition. She admitted, however, that she had been married to another man and she was not able to produce ordinary evidence as to his death. If it were left uncertain whether he was alive or dead, the complainant would necessarily fail-the onus being upon her to establish her claim. She relied upon the rule that, if a person has not been heard of for seven years by persons who in the ordinary course would have been expected to hear of him if he were still alive, it may be presumed that he is dead (Taylor on Evidence, 11th ed. (1920), vol. 1, par. 200). The complainant had not heard of her former husband for fourteen years and contended, therefore, that it should be presumed that he was dead. But the application of the rule does not establish death at any particular time (In re Phené's Trusts 1 ). It only produces the result that, if a person has not been heard of by persons who might have been expected to hear of him for a period of not less than seven years, he may be presumed to be dead at the time when the question arises in legal proceedings. The rule does not bring about the result that the person is deemed to be dead at the end of a seven-years' period (cases cited in Halsbury's Laws of England, 2nd ed., vol. 13, pp- 630, 631). Thus, even if the rule is applied in the present case, it could do no more than establish that the com- plainant's former husband had died at some time before 5th May 1937, when the complaint against the defendant was heard at Port Pirie. The rule cannot be relied upon as establishing that her husband had died before 6th January 1932, when she married Axon.

Further, the rule only applies where it is shown that persons who might have been expected to hear of the continued existence of the person whose death is in question have failed to hear of him. It is clear, as a matter of common sense, that no presumption as to the death of a particular person can arise from the fact that he has not been heard of by persons who had no connection or association with him, or between whom and himself no mutual interest had existed. The same principle must apply where the person who gives evidence

1(1870) 5 Ch. App. 139. 59 CLR 402

A, that he or she has not heard of him is a person from whom he would

naturally conceal his whereabouts. In this case the complainant had obtained a maintenance order against Herzich, and he would have every reason for not making his continued existence known to her and for endeavouring to prevent her from learning where he was. For this reason, therefore, the rule cannot be applied in favour of the complainant for the purpose of conclusively establishing the death of Herzich.

The justices, as is shown by the quotations which I have made from the reasons which they gave for their decision, considered that, the complainant having proved her marriage to the defendant, it was for the defendant to satisfy the court that the former husband was still alive. But it was for the complainant to establish upon a balance of probabilities, after all the evidence had been duly considered, that she was the wife of the defendant. If the justices were left in a state of doubt as to whether or not she was the wife of the defendant, no order should have been made in favour of the complainant.

The defendant appealed to the Supreme Court, and Richards J. set aside the order and remitted the complaint for hearing at Port Pirie. This course was adopted in order that the complainant should have an opportunity to give further evidence, if she could, on all the elements necessary to establish her cause. In my opinion this was a proper order to make and the appeal to this court should, therefore, be dismissed. As the appellant was permitted to appeal to this court in forma pauperis, there will be no order as to the costs of this appeal.

DIXON J. This is an appeal by special leave from an order of the Supreme Court of South Australia setting aside and remitting for rehearing an order of a court of summary jurisdiction made under the Maintenance Act 1926-1936 (S.A.). The order directed the now respondent to pay maintenance to the now appellant as his wife. Before the court of summary jurisdiction he denied that she was his lawful wife. The ceremony of marriage was duly performed between them on 6th January 1932, and they lived together as man and wife for the next five years. But on 15th

59 CLR 403

July 1911 she had been married to one Mauro Herzich, a man then twenty-eight years of age. After nearly twelve years of cohabitation Mauro Herzich appears to have deserted the appellant. In 1923 she obtained an order against him for maintenance in proceedings which began with his arrest.

But immediately after these proceed- ings he departed, and she has not seen him or heard from him since. He made no payments under the order. In these circumstances the respondent denied that the appellant was free to marry him in 1932.

The issue upon which the validity of the marriage depended was whether Mauro Herzich was alive on 6th January 1932. The parties proved some further facts on which one or other of them relied as relevant to this issue, but the finding of the court was based upon the burden of proof. It was, in effect, that the evidence did not establish satisfactorily that Mauro Herzich survived until 6th January 1932. The order was set aside in the Supreme Court by Richards J. on the ground that the burden of proof of the survival of Mauro Herzich had been erroneously placed upon the respondent, or, at all events, too high a degree of proof had been demanded of him.

There can be no doubt that the appellant in proceeding against the respondent for maintenance undertook the burden of establishing that she was his lawful wife. The burden of proof upon this issue lay upon her throughout the case. She could not succeed unless upon the whole evidence the court, after giving due effect to any appropriate presumption, was left reasonably satisfied that she was lawfully married to the respondent. The degree of proof required is proof upon a preponderance of probabilities, and it may be obtained by direct testimony, by circumstantial evidence, or presumptively. But whatever mode or combination of modes of proof she might adopt, she must sustain the issue upon which her title to a main- tenance order primarily depends.

The important question raised by the case is how far presumptions operated in her favour to support the burden thus placed upon her. Upon proof that a marriage ceremony had been duly performed between herself and the respondent a presumption arose in favour of the validity of the marriage. It is said that the presumption is confined to the regularity and efficacy of the ceremony as a lawful mode of marriage. This, in my opinion, is not correct. The

59 CLR 404

presumption in favour of the validity of a marriage duly celebrated casts upon those who deny it the burden of producing reasonable evidence of the fact which renders the marriage void, whether that fact is an impediment consisting in a prior marriage or a prohibited degree of relationship or the failure to fulfil some condition indispens- able to the efficacy of the ceremony. It may be true that, until the decision in the Court of Appeal in Monckton v. Tarr 1, it was not easy to find English authority directly deciding that the presumption extended to the freedom of each spouse from the impediment of a subsisting prior marriage. But the rule semper presumitur pro matrimonio was not, I think, understood as limited to ceremonial regularity, and there was direct Dominion and American authority. Indeed, in America, the presumption against invalidation by a prior subsisting marriage has been pressed to very great lengths. It has been considered to impose upon those seeking to invalidate the later marriage the burden of proving not only the performance of the earlier ceremony and the survival of the parties to it, but also that the marriage has not been dissolved by judicial decree-a burden which, apparently, has proved no light one where SO many jurisdictions may be invoked (See Harvard Law Review, vol. 30, pp. 500-503). Thus, when it appeared that between the appellant and the respondent a ceremony of marriage had been duly performed, enough had been shown to sustain the burden of proving that she was his lawful wife and she became entitled to a finding upon that issue in her favour unless circumstances were proved justifying an inference of fact sufficient to invalidate the marriage between them. The facts relied upon by the respondent for this purpose were two, namely, that she had already married Mauro Herzich and that on the date of her marriage to the appellant Mauro Herzich was still alive. The first of the two facts was fully proved, and it is the second that raises the question on which the case depends. When it is proved that a human being exists at a specified time the proof will support the inference that he was alive at a later time to which, having regard to the circumstances, it is reasonably likely that in the ordinary course of affairs he would survive. It is not a rigid presumption of law. The greater the length of time the weaker

1(1930) 23 B.W.C.C. 504 (1931) W.C. &I. Rep. 24. 59 CLR 405

the support for the inference. If it appears that there were circum- stances of danger to the life in question, such as illness, enlistment for active service or participation in a perilous enterprise, the presumption will be overturned, at all events when reasonable inquiries have been made into the man's fate or whereabouts and without result. The presumption of life is but a deduction from probabilities and must always depend on the accompanying facts. "In England it is only a general supposition of continuance, applic- able to everything which has once been proved to exist-to an orange as well as a man :-a presumption which serves, in reasoning, to relieve from the necessity of constantly re-proving, from minute to minute, this once-proved fact of existence" (The late Professor J. B. Thayer, Preliminary Treatise on Evidence at Common Law (1898), p. 348). As time increases, the inference of survivorship may become inadmissible, and after a period arbitrarily fixed at seven years, if certain conditions are fulfilled, a presumption of law arises under which a court must treat the life as having ended before the proceed- ings in which the question arises. If, at the time when the issue whether a man is alive or dead must be judicially determined, at least seven years have elapsed since he was last seen or heard of by those who in the circumstances of the case would according to the common course of affairs be likely to have received communica- tions from him or to have learned of his whereabouts, were he living, then, in the absence of evidence to the contrary, it should be found that he is dead. But the presumption authorizes no finding that he died at or before a given date. It is limited to a presumptive conclusion that at the time of the proceedings the man no longer lives. In Lal Chand Marwari v. Mahaut Ramrup Gir 1 Lord Blanesburgh, speaking for the Privy Council, said that there is only one presumption and that is that at the time when the suit was instituted the man there in question was no longer alive. "There is no presumption at all as to when he died. That like any other fact is a matter of proof." His Lordship observed as not a little remarkable that the contrary theory was still widely held, although

often shown to be mistaken. After stating how it reappeared in the case before the board, he continued: Searching for an

1(1925) L.R. 53 Ind. App. 24, at p. 31 ; 42 T.L.R. 159, at p. 160. 59 CLR 406

explanation of this very persistent heresy their Lordships find it words in which the rule both in India and in England is usually expressed. These words, taken originally from In re Phené's Trusts 1, run as follows 'If a person has not been heard of for seven years, there is a presumption of law that he is dead: but at what time within that period he died is not a matter of presumption but of evidence and the onus of proving that the death took place at any particular time within the seven years lies upon the person who claims a right to the establishment of which that fact is essential.' Following these words, it is constantly assumed-not perhaps unnaturally-that where the period of disappearance exceeds seven years. death, which may not be presumed at any time during the period of seven years, may be presumed to have taken place at its close. This of course is not SO. The presumption is the same if the period exceeds seven years. The period is one and continuous, though it may be divisible into three or four periods of seven years. Probably the true rule would be less liable to be missed, and would itself be stated more accurately, if, instead of speaking of a person who had not been heard of for seven years, it described the period of disappearance as one of not less than seven years. follows that in the present case the disappearance in 1923 of Mauro Herzich gives rise to no presumption that he was dead on 6th January 1932. In fact the conditions were not fulfilled for presuming his death at the hearing before the court of summary jurisdiction when the order now in question was made. For, in the circumstances in which he left his wife, she was not a person with whom he would be likely to communicate or who would be likely to hear of his whereabouts. He was, in effect, a fugitive from her.

But the question is not whether a positive finding that he was dead on 6th January 1932 is justified by proof or legal presumption. It was for the respondent to overcome the presumption in favour of the marriage celebrated on that date between himself and the appellant. He failed to obtain from the court of summary juris- diction an affirmative finding that Mauro Herzich was then alive, and unless such a finding is made the marriage must be treated as

1(1870) 5 Ch. App. 139. 59 CLR 407

valid. Richards J., to whom Monckton v. Tarr 1 does not appear to have been cited, set aside the decision because he considered that the court had erroneously placed upon the alleged husband, the now respondent, the burden of proving that Mauro Herzich survived until 6th January 1932. But, in my opinion, it was not erroneous to place that onus upon him, and, if in the end the court of summary jurisdiction were not satisfied upon a balance of probabilities that her previous husband was alive on that day, the marriage with the now respondent, the celebration of which was proved by his alleged wife, the now appellant, must be upheld as valid.

At the same time. in considering whether the higher degree of probability was in favour of the inference that Mauro Herzich was then living, the court was bound to weigh with the other circumstances of the case the presumption arising from the fact that in 1923 he was thirty-nine years of age and, SO far as appears, was in good health. He had already deserted his wife, and the order for maintenance would afford a powerful motive for his hiding his identity and suppressing his whereabouts from her. The husband attempted to prove that a man bearing the name Herzog was in fact Mauro Herzich, and the reasons of the court of summary juris- diction contain the statement that, after a full consideration of the evidence, a substantial doubt remained in the minds of the court as to whether the man Herzog was in fact identical with Herzich. The reasons proceed to say that "strict proof is required of the existence of Herzich." These observations suggest that, in arriving at its conclusion, the court did not give consideration, or, at all events, full weight, to the presumption of life as affording, in the general circumstances of the case, presumptive proof of Mauro Herzich's existence on 6th January 1932 on which the court was at liberty to act if, on all the facts, a sufficient degree of probability arose to produce a reasonable satisfaction of his survival to that date. The question must be treated as one of fact, and the court was not bound to draw such an inference. But it seems likely that the court demanded a stricter degree or heavier burden of proof than the law requires and treated the failure to prove to its complete

1(1930) 23 B.W.C.C. 504 ; (1931) W.C. &I. Rep. 24. 59 CLR 408

A. satisfaction that Herzog and Herzich were one man and not two

men as decisive against the husband.

On the whole, I think, for this reason, that the matter should go back for rehearing and the appeal should accordingly be dismissed.

EVATT J. In these proceedings it has been assumed that sec. 66 of the Maintenance Act (S.A.), under which the appellant applied for summary protection against her husband, based on desertion and wilful neglect to provide reasonable maintenance, permits a magistrate to determine the question of status whether or not the appellant was the wife of the respondent. I do not ques- tion the assumption. But as a general rule it is very unsatisfactory to allow a respondent who is failing to maintain the woman with whom he has entered into the bonds of matrimony to challenge the validity of that ceremony except before the highest court of the State.

The magistrate found in the wife's favour on all questions. The only one upon which Richards J. differed from him concerns the question whether Mrs. Axon proved that she was the wife of the respondent.

It appeared that the parties to the proceedings went through the ceremony of marriage on January 6th, 1932, at Port Pirie, South Australia. The case of the husband was that, at the time of this ceremony, the first husband of the present appellant was still alive. It was proved by the husband that, on July 15th, 1911, also at Port Pirie, South Australia, the wife had been married to one Herzich. It is proved by the wife that in February 1923 she charged her then husband, Herzich, who had remained with her at Port Pirie from the time of the marriage until 1923, with desertion and obtained an order against him; and that subsequently, early in 1924, Herzich was brought down by the police to Port Pirie from Broken Hill. It also appeared that Herzich left Port Pirie the next morning by train, and that, during the intervening eight years or more, his wife had never heard of his being alive, that she had been positively informed by a Broken-Hill resident that he was dead, that she advertised for him in the newspapers unsuccessfully and communicated with the Broken-Hill police to find him without any result, and that, prior to

59 CLR 409

her marriage with Axon, she informed him of all these facts. The information she had obtained from the Broken-Hill resident was from a person of Austrian descent, the same as that of her husband Herzich, and was to the effect that Herzich had died at Broken Hill shortly after the time of the court proceedings in 1923-1924.

In answer to all this evidence Axon produced a witness, Chesson, who swore that, in December 1936, a man called at his home at Port Pirie, having been driven there in a car, that the man was Herzich, and that he talked with him about old times at Port Pirie, each being pleased to see the other again. It appears from the evidence of the witness Edwards that he was the driver of the car which conveyed the visitor to Chesson's residence. Edwards did not know the visitor as Herzich, but only as Herzog. Edwards swore that he showed him over the works and around the wharves at Port Pirie during a stay of several days, and that the man was accompanied by his wife and a sister of Edwards, and lived at Birrell Street, Broken Hill.

An extraordinary feature of Chesson's evidence is this. Although he says that he knew Herzich at the Port Pirie Smelters before his disappearance in 1923, and that he knew the present Mrs. Axon as having been Herzich's wife, and although he (Chesson) had heard the rumours to the effect that Herzich was dead, he did not say a word of all this to the man whom he positively identifies as Herzich. It is particularly surprising to observe that, although Chesson knew that Herzich's former wife was then married to another man, he failed to tell his visitor of this interesting and exciting event.

Before the magistrate delivered his judgment he received a letter purporting to come from Beryl (not Birrell), Street, Broken Hill, and purporting to be signed by Herzog. This letter stated that the writer was not Herzich, and that he was Herzog and no one else it then made the statement, understatement I should say, that the evidence of Chesson as reported was rather embarrassing to his wife and himself. In his judgment the magistrate paid no attention to this communication, but it was made known to both parties.

The magistrate found that Mrs. Axon was the wife of Axon by virtue of the marriage ceremony performed in 1932. He doubted whether the person driven by Edwards to Chesson's place was

59 CLR 410

A. Herzich, the former husband of Mrs. Axon, and pointed out that,

during the stay of this man in Port Pirie, although he was about there for some time, no one else had been called to identify him as Herzich.

The magistrate said that after considering the evidence a sub- stantial doubt remained in his mind as to whether the man Herzog was in fact identical with Herzich. He added: " In the the court strict proof is required of the existence of Herzich if the defence set up in this case is to succeed, and the defendant has failed to discharge the onus which lies upon him in this regard to the satis- faction of the court."

In my opinion this statement by the magistrate was not intended by him to express an opinion that the general onus of establishing her status as Axon's wife was ever shifted from the shoulders of the appellant. All he meant to suggest was that the only fact which introduced any element of doubt into the case was the fact, con- clusive if established, of Herzich's recent appearance at Port Pirie. Then, without condemning Chesson as being utterly unreliable, he thought that his doubt as to Chesson's evidence of identity should be resolved in favour of the appellant. In my view the course the magistrate took was correct.

Before the Supreme Court and also before us the language of the magistrate has been subjected to a meticulous verbal criticism. But I think that it is answered, for the most part, by the fact that the magistrate never intended to hold more than this-that the general onus of establishing the appellant's case was affirmatively established by reason of certain evidence combined with a presump- tion of fact.

At the time of Mrs. Axon's marriage to Axon in 1932, the law in force as to bigamy was sec. 77 of the Criminal Law Consolidation Act 1876. A similar provision is now contained in sec. 79 of the Consolidation Act of 1935. After defining the crime of bigamy, the section provides that nothing therein shall apply to any person marrying a second time " whose husband or wife shall have been continually absent from such person for the space of seven years then last past, and shall not have been known by such person to be living within that time."

59 CLR 411

This section reproduces the relevant portion of sec. 57 of the English Act 24 &25 Vict. c. 100, which was referred to in the leading case of R. v. Tolson 1. It will be observed that, both in England and South Australia. no question of bigamy can arise in relation to the act of marrying a second time provided that (a) during seven years prior to the remarriage, the defendant's former wife has been continually absent from the defendant, and (b) during the space of seven years terminating at the time of remarriage, the former wife was not known to be living.

In R. v. Tolson 2, Cave J. recites the terms of the Act of James I. which first made bigamy an offence punishable by the courts of common law. The enactment is substantially the same as 24 &25 Vict.

57. Further, as Cave J., emphasizes, when the Act of James I. was passed, the presumption of a man's death after he had not been heard of for seven years had not been established. Indeed, the general presumption was brought into force by analogy to the statute of James I., and to the subsequent statute 18 &19 Car. II. c. 11, which provided that, after the absence for seven years of a person for whose life an estate had been granted with no sufficient and evident proof that such person was living, " the person or persons upon whose life or lives such estate depended shall be accounted as naturally dead, and in every action brought for the recovery of the said tenements by the lessors or reversioners. their heirs or assigns, the judges before whom such action shall be brought shall direct the jury to give their verdict as if the person SO remaining beyond the seas or otherwise absenting himself were dead."

In my opinion, the terms of the South Australian Act in relation to the crime of bigamy make it necessary that, in proceedings in which such a presumption can be applied, the court should presume that, if at the time of a second marriage of a person, his spouse has been continually absent from him for the space of seven years pre- viously, and at no time during such period has been known to be alive, the absentee spouse has died prior to the time of the remarriage. It is true that, apart altogether from the presumption of death prior to remarriage which, in my opinion, is required by the bigamy enactment, there exists the presumption which Stephen calls " 'the

1(1889) 23 Q.B.D. 168. 2(1889) 23 Q.B.D. at p. 184. 59 CLR 412

presumption of death from seven years' absence." Such presump- tion is of general application, and its nature and history are fully discussed in Re Phene's Trusts 1. In that case, Giffard L.J., in a judgment which has since won frequent approval, quotes the case of Doe v. Nepean 2, where the Court of Exchequer Chamber had laid emphasis on the fact that the Act 18 &19 Car. II. C. 11 (misquoted as C. 6) distinctly points to the presumption of the fact of death, but not the time of death. As is pointed out in Stephen's Digest of the Law of Evidence, Art. 99, the general presumption of death carries with it no presumption as to the time of death, and the burden of proving death at any particular time is on the person who asserts it. Such general presumption operates SO as to prove the fact of death at the time of the institution of the legal proceedings where the fact giving rise to the presumption is proved. Of course, in many cases, such presumption is sufficient to carry the person who relies upon it the necessary distance, e.g., in cases under an insurance policy, where the fact of the termination of the life is sufficiently proved if death can be presumed as at the time when the writ is issued (Prudential Assurance Co. v. Edmonds 3 ). In many cases, however, where death must be shown to have occurred at some point of time anterior to the curial proceedings, the presumption may carry the party relying upon it only a certain distance, or no distance at all (Re Phené's Trusts (1) ).

But, if any presumption can properly be drawn by virtue of the terms of the enactment of South Australia dealing with bigamy, it must be drawn as to the fact of the death of the absent spouse prior to the remarriage of the other, and it necessarily must relate to the fact of death at some point of time prior to the legal proceedings where the presumption is relied upon. Then why should any pre- sumption be drawn by virtue of the terms of the bigamy enactment The reason is this-the meaning of the Act, SO far as material for present purposes is, not that it is excusable for a spouse to remarry merely because the other has been absent for seven years prior to the remarriage and has not been known to have been alive but rather that, if such an absence and the fact of not being heard of

1(1870) 5 Ch. App. 139. 2(1833) 5 B. &Ad. 86 ; 110 E.R. 724. 3(1877) 2 App. Cas. 487. 59 CLR 413

are proved, the reasonable inference which everyone should draw is that the absent spouse is dead. The inference is SO reasonable that the law permits or perhaps even invites the non-absentee spouse to remarry, irrespective of his or her belief. Why ? Because no reasonable person would fail to assume the fact of death. But that is only another way of saying that the statute says that, in the conditions described, there shall be a presumption of death and of death prior to the second marriage.

Although the original bigamy Act led to the general rule as to presumption of death after an unaccounted-for absence of at least seven years, it had an independent existence in relation to, and as. reinforcing, the somewhat vaguer presumption in favour of the validity of the second marriage. Whereas the general presumption of death is by no means limited to the case of husband and wife, and is not directed to the fact of death at any given time, the par- ticular presumption required by the bigamy enactment relates solely to the case of husband and wife, and is directed solely to the point of time at which remarriage takes place. In each case, however, the presumption of death is a presumption of fact, which presumption is, of course, rebuttable. Indeed, in the usual instance of a prosecu- tion for bigamy, the presumption is immediately rebutted because the Crown proves beyond doubt that, at the time of the second marriage, the first spouse was alive.

In R. v. Lumley 1, the court pointed out that, even in a prosecution for bigamy, the fact that the first spouse was alive at the time of the second marriage could be proved by circumstantial evidence. But the court explained the proviso to the bigamy Act as to absence for seven years prior to the second marriage, &., as follows: "The legislature, by this proviso, sanctions a presumption that a person who has not been heard of for seven years is dead; but the proviso affords no ground for the converse proposition, viz., that when a party has been seen or heard of within seven years, a presumption arises that he is still living. That, as we have said, is always a question of fact ' 2.

The first sentence in this quotation from Lush J. shows that the proviso to the bigamy enactment of South Australia is based on

1(1869) L.R. I C.C.R. 196. 2(1869) L.R. 1 C.C.R., at p. 199. 59 CLR 414

the principle that, if the conditions therein mentioned are satisfied, it is SO reasonable to presume death prior to the marriage that criminal liability can never attach in relation to the spouse remarry- ing.

Tennyson puts the same argument into the mouth of Philip:

It is beyond all hope, against all chance That he who left you ten long years ago Should still be living." In my opinion, the presumption which derives from the bigamy enactment is applicable to such proceedings in summary jurisdiction as were had in the present case. Thus, in Ousley v. Ousley 1 a'Beckett J. had to consider an application for an order to review a complaint for maintenance brought in the Court of Petty Sessions at Richmond, Victoria. The question was whether a man who married a woman in August 1909 was entitled to refuse to support her because she was unable to prove the death of her first husband, who deserted her in 1896 and had not been heard of by her since that time. a'Beckett J. came to the conclusion that the justices were wrong in holding against the woman's claim. They had said that, although she had not heard from her first husband for seven years, " the presumption of law thus arising as to his death was merely a safeguard against a prosecution for bigamy, and dismissed her complaint" 2. a'Beckett J. held that the justices were wrong, distinguishing the case from one where some positive right was being asserted which was dependent on the fact of the death of the husband. He said :-

"I think the justices were wrong in declaring the second marriage bad, not merely because the evidence raised a presumption of the first husband's death before the second marriage, but that even if it did not, and the matter was left open, they should not have treated the second marriage as bad until it was proved to be so. Whether the first husband was living at the time of the second marriage was left in uncertainty. It would be a great hardship on a deserted wife if she could not contract a marriage which would hold good in a court of law unless in a position to establish by proper legal evidence the death of the deserter" 3. In Kempton v. Public Trustee 4, Reed J. said:

But a marriage, prima facie valid, being established, evidence of cohabita- tion with a general reputation of marriage affords a strong presumption that

1(1912) V.L.R. 32 33 A.L.T. 155. 2(1912) V.L.R., at p. 34. 3(1912) V.L.R., at pp. 34, 35. 4(1932) N.Z.L.R., at p. 1387. 59 CLR 415

there was no legal impediment to the marriage. Although directed to different circumstances, the rule enunciated by Lord Lyndhurst in Morris v. Davies 1 and approved by the House of Lords in Piers v. Piers 2 is of general applica- tion. His Lordship said that the presumption of a lawful marriage is not lightly to be repelled. It is not to be broken in upon or shaken by a mere balance of probability. The evidence for the purpose of repelling it must be strong, distinct, satisfactory and conclusive.' Lord Brougham, in Piers V. Piers, expressed a doubt as to whether the rule should not stop at the word

' satisfactory,' and later authorities would appear to support the view of that learned judge-the evidence need not necessarily be conclusive, but must be strong, distinct and satisfactory. The intention to contract a valid marriage, as in the present case, strengthens the presumption- the courts are astute to presume marriage where the matrimonial intent can be shown to exist' (Eversley, Domestic Relations, 4th ed., p. 6)." These two decisions show that in the present case the appellant was entitled to call in aid two different but concurrent presumptions (1) the presumption based on the language of the bigamy enactment that, prior to the remarriage, the first spouse had died, and (2) the presumption in favour of the validity of the second marriage. As to the second presumption, Wigmore has said :-

Whether the successive shiftings of the burdens should be worked out with mathematical nicety according to the various presumptions applicable, or whether all should be merged in a general presumption in favor of the later marriage, is a knotty question; and no successful generalization is yet accepted. But it may be noted that the peculiar force of a presumption as merely affecting the opponent's duty to produce some evidence is not always observed in the judicial discussion of the problem' " (Canadian ed., sec. 2507). The complications to which Wigmore refers have not yet been removed from the United States cases (Harvard Law Review, vol. 46, pp. 1143, 1144).

Therefore, although the general onus of proving her marriage status lay upon the present appellant, I am of opinion that the magistrate intended to hold that in view of all the other circum- stances proved-absence for nine years prior to the remarriage, enquiries by public advertisement, enquiries by the police, reports of death prior to marriage and the ceremony of remarriage-the onus had been sufficiently discharged unless the respondent satisfied him that Herzich was alive at the date of the remarriage.

But I am also of opinion that Mrs. Axon was entitled to rely upon a presumption that Herzich was dead at the time of her mar- riage with Axon, and that, in order to rebut such presumption,

1(1837) 5 Cl. &Fin. 163, at p. 2(1849) 2 H.L.C. 331, at p. 381 9 59 CLR 416

evidence had to be adduced by Axon to satisfy the magistrate that, at the time of the Axon marriage, Herzich was still alive. The case as I have put it is much stronger than one where the remarriage has taken place without any absence of seven years during which the first spouse has not been heard of.

The result of these considerations is that the magistrate's decision in favour of Mrs. Axon was sound both on the law and on the facts. In ordinary circumstances the appeal ought to be allowed and the order in her favour restored. But I strongly favour the order for rehearing for this reason it is contrary to the true interests both of the appellant and of the respondent that the question whether the man Herzog is identical with Herzich, Mrs. Axon's first husband, should be left in doubt. Admitting fully that the present evidence in favour of the fact of identity is very unsatisfactory, and perhaps open to suspicion, the letter subsequently sent to the magistrate and purporting to come from Herzog shows that there is an imperative necessity for clearing up the question once and for all in Herzog's interests and those of his present wife, as well as in the interests of both parties to the present appeal.

For these reasons I agree that a rehearing is desirable, and my only regret is that the question of identity was not finally disposed of before the application to this court to grant special leave. At the same time, I think that the general legal merits SO strongly favour Mrs. Axon that, in the absence of evidence satisfying the magis- trate on a rehearing that Herzog is identical with Herzich, she is practically bound to succeed. Therefore I consider that the order of the Supreme Court should be varied in relation to the question of

Subject to such a variation, I would dismiss the appeal.

Appeal dismissed without costs. Costs of hearing

and of appeal to Supreme Court to abide result of rehearing. Solicitor for the appellant, S. J. Warren, by Newman, Gillman &Sparrow.

Solicitor for the respondent, C. R. Hannan, by Alderman, Reid &Brazel.