Piggott v Piggott

[1938] HCA 62

61 CLR 378

RESPONDENT, PIGGOTT

ON APPEAL FROM THE SUPREME COURT OF Evidence - Divorce - Adultery - Condonation - Birth of child - Admissibility of

evidence by husband of mon-intercourse-Legitimacy of child-Rule in Russell SYDNEY,

Held, by Latham C.J. and Starke J. (Dixon and McTiernan JJ contra), Aug. 18, 19,

that the rule in Russell v. Russell, (1924) A.C. 687, does not apply to render inadmissible evidence of non-intercourse given by a husband to support a denial of a defence of condonation in proceedings for dissolution of marriage although that evidence may show that a child born to the wife is illegitimate.

The court being equally divided, the decision of the Supreme Court of Tas- mania (Crisp C.J.) was affirmed.

APPEAL from the Supreme Court of Tasmania.

A petition was filed on 25th June 1937, in the Supreme Court of Tasmania, by Russell George Jacob Piggott for the dissolution of his marriage with Dorothy Joyce Piggott on the ground that on divers occasions between 12th and 24th April 1937, at Burwood, a suburb of Sydney, New South Wales, she committed adultery with Gordon Whitebrook of Sydney.

The wife filed an answer to the petition in which she admitted that on 23rd April 1937 she committed adultery with Whitebrook

61 CLR 379

but otherwise denied the allegation of adultery. She also alleged that the petitioner condoned the act of adultery and, further, that he had been guilty of wilful neglect and misconduct towards her which had conduced to the adultery.

The petitioner in his reply joined issue upon the allegation of adultery except in SO far as it was admitted, and he denied the condonation and, also, the wilful neglect and misconduct alleged.

The parties were married on 23rd March 1935, and were both domiciled in Tasmania.

The wife gave particulars of the condonation relied upon by her. In these particulars it was alleged that intercourse took place between the husband and wife at specified places in and around Hobart, Tasmania, on the following dates in 1937, namely, June 11th, 12th, 13th, 15th, 16th, 19th, 20th, 21st and 22nd. At the hearing of the petition the wife admitted that, although alleged in the particulars supplied by her, intercourse did not take place on 19th June.

The hearing of the petition was commenced on 10th May 1938. On 1st March 1938 a child was born to the wife. The birth of the child was premature by between two and four weeks. The medical evidence was that the child was conceived in the latter part of June 1937, and it excluded any reasonable possibility of the child being the result of the adultery with the co-respondent in April 1937. Objection was taken on behalf of the wife to the husband giving evidence denying the acts of intercourse alleged. The objection was overruled by the trial judge, Crisp C.J., on the ground that the rule in Russell v. Russell 1 has not been extended to a case where the issue is condonation. The wife then gave evidence that inter- course had occurred on all the occasions, other than on 19th June, mentioned in the particulars given by her. The husband denied the evidence of the wife, saying that intercourse had not taken place. Both parties gave evidence of the surrounding circumstances and of their mutual relations during the important period. The husband's father and another witness gave evidence showing the relations which in June 1937 existed between the husband and the wife. At the time when, it was alleged, the acts of intercourse took place, the position was that the wife, with her husband's consent, left

1(1924) A.C. 687. 61 CLR 380

Hobart on 10th April 1937 for a holiday in New South Wales. She returned to Hobart on 4th May 1937. Prior to her departure the wife had given her husband an undertaking that she would not see the co-respondent and that she was going to stay with a woman friend at Yass. In fact she went straight to the co-respondent's home at Burwood, Sydney, where he was living with his mother, and stayed there for about twelve days. She committed adultery with him there. She admitted only one act of adultery, but the correspondence between her and the co-respondent which was put in evidence made it difficult to believe that adultery did not take place on other occasions. She told her husband untruths about her holiday, and for a long time denied that she had been guilty of misconduct. Eventually, on 11th June 1937, she admitted to her husband that she had committed adultery with Whitebrook at Burwood in April 1937, and it was upon this day, within an hour of the making of the admission, and upon the days immediately succeeding this admission that, as she alleged, the acts of intercourse took place, generally in a motor car, but upon one occasion upon the beach. Throughout this time she was in correspondence with the co-respondent. His letters to her were of the warmest possible character. At that time they were proposing to get married as soon as she was divorced from her husband. The husband was in a condition of anxiety, indecision and uncertainty. This fact was established by the evidence of all the witnesses. The parties were engaged in discussions about their future, and he ultimately decided to seek a divorce upon the ground of adultery. He knew at the time that his wife was proposing to join the co-respondent and to marry him. She agreed to leave Tasmania while he obtained a divorce, upon the understanding that the divorce should be put through quietly. After she left Tasmania on 23rd June 1937 she wrote letters to her husband from Melbourne. On 8th-or possibly 15th- July 1937, she wrote a letter in which she stated that the proposal to marry the co-respondent was off, that her menstrual period had been delayed and that she was very anxious about it, and that as her husband had put her " in an awkward position" he could "hardly go on with the divorce." In effect she told him that she believed or feared that she was pregnant. The husband did not

61 CLR 381

reply to these letters and refused to speak to her when she rang him up on the telephone. He did not take the letters to his solicitor until some time afterwards. The husband admitted a difficulty in describing his state of mind. He said in evidence: 'When I got the letter I thought she was in the family way and I was not going to have anything to do with her." He said that she had promised him that she would go straight until she married the co-respondent and he--the husband-considered that he "had done enough."

The trial judge, who said that the wife was a person capable of great deceit, found that the husband did not have sexual intercourse with his wife after her return to Hobart on 4th May 1937, and that though for some few days after that date the petitioner was undecided what to do in the matter, he never had any fixed intention of resuming marital relations with her. His Honour found against the wife on her allegations of wilful neglect and misconduct on the part of the petitioner.

A decree nisi for the dissolution of the marriage was granted. From that decision the wife appealed to the High Court. R. C. Wright, for the appellant. Upon a consideration of the issue of adultery, if at the material time there was access between husband and wife, and a child is born to the wife, there is a strong presumption of legitimacy (Gaskill v. Gaskill 1 ); a fortiori on an issue of condonation by intercourse. The onus of rebutting that presumption is upon the husband. The evidence of non-access lacked positiveness. The rule in Russell v. Russell 2 applies to the issue of condonation. In cases of impotence and non-access evidence is given because there is not any child. The point in Russell v. Russell was whether evidence of non-access could be given when the proof of intercourse is a child. The child proves inter- course; if non-access is proved then adultery is established: See Halsbury's Laws of England, 2nd ed., vol. 2, pp. 562 et seq. It is not logical to apply the rule where adultery is to be proved and not where condonation is alleged, especially where there is a child (Rayden and Mortimer on Divorce, 3rd ed. (1932), pp. 273, 274;

1(1921) P. 425. 2(1924) A.C., at pp. 697-699, 704-706, 709, 718, 722, 727-729, 731, 744, 753. 61 CLR 382

Browne and Latey on Divorce, 11th ed. (1931), p. 289, n. 0, Joske on Marriage and Divorce in Australia, 2nd ed. (1936), pp. 154, 155) Even if a wife has intercourse with a number of men and a child be born at the relevant time, the child is presumed to be legitimate unless non-access be proved (Gordon v. Gordon 1 ). Where a man has had probable access the law protects the child by preventing the husband from giving direct evidence as to non-access. A wife's admission of adultery and a statement that she believes that another man is the father of the child cannot bastardize the child unless non-access be proved (Warren v. Warren 2; Justice v. Justice 3 Dick v. Dick 4 ). Intercourse with other men does not conclude the question as to the legitimacy of the issue (Roast v. Roast 5 ) The admission of intercourse with another is correct because it does not affect legitimacy. But a statement of intercourse or non-inter- course with the husband is an affirmation or denial of paternity. One the law presumes the other the law forbids. The birth of a child to a wife during wedlock is prima facie evidence that such child is legitimate (Banbury Peerage Case 6 ). In the absence of satis- factory proof of non-access the court will never pronounce against legitimacy. Upon proof by third persons of non-access the rule in Russell v. Russell 7 does not prevent proof of intercourse by a wife with somebody not her husband (Brown v. Leech 8; Pearce V. Kitchin 9; Ex parte Letherbarrow; Re McMurray 10 ).

[DIXON J. referred to In re Hamer's Estate; Public Trustee V. Attorney-General 11.]

That case does not accurately state the full extent of the decision in Russell v. Russell (7). The decisions in Farnham v. Farnham 12 and Burgess v. Burgess 13, which are nullity cases, were based upon a misunderstanding of the rule in Russell v. Russell (7), and should not be followed. This court should follow the decision in G. v. G. 14, where the facts were similar to those in Farnham V.

1(1903) P. 141. 2(1925) P. 107. 3(1925) S.A.S.R. 278. 4(1927) Q.S.R. 365, W.N. (N.S.W.) 78. 5(1938) P. 8. 6(1811) 1 Sim. &St. 153 57 E.R. 7(1924) A.C. 687. 8(1925) 94 L.J. K.B. 48. 9(1931) 26 Tas. L.R. 38. 10(1938) 38 S.R. (N.S.W.) 281 ; 55 11(1936) 53 T.L.R. 275. 12(1937) P. 49. 13(1937) P. 60. 14(1934) N.Z.L.R. 246. 61 CLR 383

Farnham (1). The reasoning in the Poulett Peerage Case (2) forms no basis for the deductions made in Farnham v. Farnham (1). Facts should not be supposed to exist; they must be proved by legal evidence. The principle enunciated in the Poulett Peerage Case (2) was applied in Re D. F. Mackay (3) and McLean v. McLean (4). Although as regards intercourse before marriage the rule in Russell v. Russell (5) does not apply, the presumption of legiti- macy does apply (Halsbury's Laws of England, 2nd ed., vol. 2, p. 560, par. 769). There is not any foundation for the suggestion that the rule does not apply where, as in R. v. Inhabitants of Sourton (6), husband and wife were not living under the same roof. For the application of the rule where the husband and wife are living apart under a decree of separation, see Andrews v. Andrews (7), or under a separation, see Mart v. Mart (8); Stafford v. Kidd (9)-in which In re Bromage Public Trustee v. Cuthbert (10) was disapproved- and Henley v. Henley (11). The principle of those decisions is not applicable where a maintenance order is in existence in respect of a husband and wife who live apart (Bowen v. Norman (12) ). To suggest that the rule does not apply to a case where the child is in the period of gestation is merely evading the rule for that reason Aarnes v. Aarnes (13) was wrongly decided. The trial judge was wrong in rejecting the presumption of legitimacy (Banbury Peerage Case (14) Gaskill v. Gaskill (15); Halsbury's Laws of England, 2nd ed., vol. 2, pp. 559 et seq.). The evidence given for the purpose of rebutting that presumption was not strong, distinct, satisfactory and conclusive (Morris v. Davies (16); Inre Bromage; Public Trustee V. Cuthbert (17) ). The presumption continues despite adultery (Gordon v. Gordon (18) Pryor v. Pryor (19); Pearce v. Kitchin (20); Halsbury's Laws of England, 2nd ed., vol. 2, p. 560, par. 768). The

(11) (1927) S.A.S.R. 364. (2) (1903) A.C. 395.

(12) (1938) 1 K.B. 689, at p. 692. (3) (1928) 28 S.R. (N.S.W.) 404; 45

(13) (1929) 2 D.L.R. 298. W.N. (N.S.W.) 106.

(14) (1811) 1 Sim. &St. 153; 57 E.R. (4) (1931) N.Z.L.R. 167. (5) (1924) A.C. 687.

(15) (1921) P., at p. 434. (6) (1836) 5 Ad. &E. 180 111 E.R.

(16) (1837) 5 Cl. &Fin. 163, at p. 266;

7 E.R. 365, at p. 404. (7) (1924) P. 255.

(17) (1935) Ch., at p. 611. (8) (1926) P. 24.

(18) (1903) P. 141. (9) (1937) 1 K.B. 395.

(19) (1887) 12 P.D. 165. (10) (1935) Ch. 605.

(20) (1931) 26 Tas. L.R. 38.

61 CLR 384

presumption was considered in Lloyd v. Powell Duffryn Steam Coal Co. Ltd. 1: See also Schofield v. Orrell Colliery Co. Ltd. 2. The evidence as a whole is insufficient to rebut the presumption, or, if the presumption be not available, the evidence as a whole affirma- tively proves condonation. The letters written and sent by the appellant to the respondent will not reasonably bear the construction that they were the letters of a designing woman who was trying to entrap her husband. Intercourse subsequent to knowledge is proof of condonation (Halsbury's Laws of England, 2nd ed., vol. 10, p. 681, n. u; Rayden and Mortimer on Divorce, 3rd ed. (1932), p. 121, n. b; Cramp v. Cramp 3; Germany v. Germany 4 ).

[He was stopped on this point.] Doyle, for the respondent. The ultimate issue resolved by the trial judge was a pure question of fact as to whether there had been condonation of the appellant's adultery. The appellant was not a truthful witness. Where evidence of the parties was in conflict the judge accepted the evidence of the husband. The appellant's allega- tion of condonation was not established to the satisfaction of the judge. On the contrary, he found that the allegation had been disproved. The onus is upon the appellant to prove intercourse with the respondent. From the medical evidence it is reasonable to assume that conception did not necessarily take place before her departure from Tasmania. There never was any reconciliation between the parties. Evidence relating to the alleged acts of intercourse was admissible notwithstanding the rule in Russell v. Russell 5. The express terms of the decision in Russell v. Russell (5) do not embrace this case. The rule does not apply to condonation. This view is supported by Farnham v. Farnham 6. Such statements in Russell V. Russell (5) as would seem to extend the application of the rule to a case of condonation were obiter dicta, and were not justified by authority. The rule must be related strictly to matters coming within the precise compass of the decision in Russell v. Russell (5), or, in other words, the rule does not apply and should not apply where evidence of

1(1914) A.C. 733, at pp. 740 et seq. 2(1909) 1 K.B. 178 (1909) A.C. 433. 3(1920) P. 158, at pp. 165-167, 170, 4(1938) P. 202. 5(1924) A.C. 687. 6(1937) P. 49. 61 CLR 385

non-access is not given for the purpose of proving the child illegiti- mate. Here there is not any reference on the pleadings to the birth of the child; it is a mere piece of evidence. Except for a reference in the judgment of Littledale J., R. v. The Inhabitants of Sourton 1 does not seem to be an authority for the proposition for which it is cited in Halsbury's Laws of England, 2nd ed., vol. 2, p. 562. Yates V. Chippindale 2, cited as an authority for the same proposition, dealt with the proposition of the giving of evidence of access, and not of the giving of evidence of non-access. The rule is only applic- able where the object is to bastardize the child (Boston v. Boston 3 In re Hamer's Estate Public Trustee v. Attorney-General 4; Farnham v. Farnham 5 Ex parte Letherbarrow Re McMurray 6 ), and not, as here, to prove or disapprove intercourse of which the bastardizing of the child is merely the possible result (Mart V. Mart 7 )-and see Bowen v. Norman 8. The appellant admitted adultery therefore the respondent should be permitted to prove non-access (Burgess v. Burgess 9 ). The appellant and the respondent were in fact living apart; therefore the evidence should be admitted as a proper extension of the decisions in the judicial separation cases (Dowd v. Dowd 10 ). The trial judge did not rule against a presumption of legitimacy he merely stated that he was not trying such an issue. At most the presumption is only one of access, and, therefore, of intercourse, and is not a presumption of legitimacy. The appellant has not discharged the onus of proving intercourse at the relevant time between herself and her husband. The appropriate presumption, on the birth of the child having been proved, was a presumption relevant to the issues before the court, namely, access and intercourse with the husband, and not the legitimacy of the child, which was not an issue in the case. Those presumptions are separate and distinct one from the other (Mart V. Mart 11 the presumption of intercourse is more readily

1(1836) 5 Ad. &E. 180; 111 E.R. 2(1862) 11 C.B. (N.S.) 512 ; 142 3(1928) 138 L.T. 647. 4(1936) 53 T.L.R., at p. 276. 5(1937) P., at p. 56. 6(1938) 38 S.R. (N.S.W.) 281 : 55 W.N. (N.S.W.) 78. 7(1926) P., at p. 26. 8(1938) 1 K.B. 689. 9(1937) P., at p. 62. 10(1929) Q.W.N. 13 11(1926) P. 24. 61 CLR 386

rebuttable than the presumption of legitimacy (In the Estate of L. 1 ). The pleadings raise only an issue of intercourse, not of legitimacy, therefore the appellant can have only the benefit of the presumption of intercourse and not the benefit of the presumption of legitimacy. The presumption of legitimacy is irrelevant upon an issue of condonation. Gaskill v. Gaskill 2 was the commencement of the erroneous application of the presumption of legitimacy to cases where the birth of the child was relied upon. To permit the appellant to avail herself of the presumption on the mere birth of the child would place a heavy burden upon the husband to prove a negative on the issue of condonation. He would be deprived of the benefit of the particulars furnished by the appellant. The matters raised do not come within those particulars. These proceedings are not proceedings for divorce in the nature of the proceedings in Russell V. Russell 3. The words " object or possible result " 4 must be limited to the precise facts in Russell v. Russell (3). The rigid applica- tion of the rule would prevent the truth from being placed before the court and, in many cases, doubtless, would result in an injustice being done, or would produce curious results (R. v. Seaton 5 ). Even if there are presumptions the findings of the judge are sufficient to determine the matter in favour of the husband. Those findings are fully warranted by the evidence. Condonation is not concluded by intercourse (Spunner v. Spunner 6 ). There cannot be condona- tion without an intention completely to forgive and to resume matrimonial relations. The references in Rayden and Mortimer on Divorce, 3rd ed. (1932), at pp. 120, 121, and Joske on Marriage and Divorce in Australia, 2nd ed. (1936), at p. 221, are based on Cramp V. Cramp 7; that decision, however, proceeds upon a wrong basis in that it excludes from consideration the element of forgiveness in fact. Condonation is a matter of implication from the facts between the parties (Keats v. Keats 8; Hall v. Hall 9; Roberts V. Roberts 10; Hare v. Hare 11; Turnbull v. Turnbull 12; Timms

1(1919) V.L.R. 17, at pp. 26, 30, 2(1921) P. 425. pp. 760, 763, 765. 3(1924) A.C. 687. 4(1924) A.C., at p. 697. 5(1933) N.Z.L.R. 548.

6(1926) V.L.R. 183; 47 A.L.T. 178. 7(1920) P., at p. 162. 36 ; 40 A.L.T. 153, at pp. 157, 8(1859) 1 Sw. &Tr. 334, at pp. 159, 161. 347, 354, 357 ; 164 E.R. 754, at 9(1891) 60 L.J. P. 73. 10(1917) 33 T.L.R. 333, at p. 335. 11(1920) 36 T.L.R. 331. 12(1925) 41 T.L.R. 507. 61 CLR 387

V. Timms 1 ). The evidence does not disclose any indication on the part of the husband of "the complete forgiveness and blotting out of the conjugal offence committed by the appellant (Sneyd V. Sneyd 2 ), on the contrary, it does show an intention to continue the divorce proceedings. The most the appellant is entitled to is a new trial.

R.C. Wright, in reply. One act of marital intercourse is construed as condonation of a known offence (Germany v. Germany 3; Bray V. Bray 4 ). Timms v. Timms (1) did not deal with condonation it is merely a decision that casual acts of intercourse may not break the continuity of desertion. Spunner v. Spunner 5 is distinguish- able on its own particular facts. The presumption of legitimacy continues despite adultery (Cope v. Cope 6 Wright v. Holdgate 7; Phipson on Evidence, 7th ed. (1930), p. 652)-See also Law Quarterly Review, vol. 30, p. 153. For the true meaning of the passage in the judgment of Lord Dunedin in Russell v. Russell 8, see G. v. G. 9. The letters and other communications by the appellant should not have been ignored by the husband on the contrary they called for a reply (Wiedemann v. Walpole 10 ). At the date of the pleadings and the particulars the child had not been born, thus accounting for the absence therein of any mention of the child.

Cur. adv. vult. The following written judgments were delivered :------

LATHAM C.J. This is an appeal from a decree for dissolution of marriage pronounced by the Chief Justice of the Supreme Court of Tasmania upon a husband's petition for divorce upon the ground of adultery.

The petition was presented to the court in June 1937. The respondent wife admitted that she had committed adultery with a named co-respondent on 23rd April 1937, but otherwise denied

1(1925) V.L.R. 597 47 A.L.T. 50. 2(1926) P. 27. 3(1938) P., at p. 208. 4(1919) 36 W.N. (N.S.W.) 169, at 5(1926) V.L.R. 183 47 A.L.T. 6(1833) 5 Car. &P. 604 ; 172 E.R. 7(1850) 3 Car. &K. 158 175 E.R. 8(1924) A.C., at p. 729. 9(1934) N.Z.L.R., at p. 259. 10(1891) 2 Q.B. 534, at p. 539. 61 CLR 388

the allegations of adultery. She relied upon condonation as a defence to the petition and also pleaded that the petitioner had been guilty of wilful neglect and misconduct conducing to adultery. The learned Chief Justice found against the wife on all the defences raised. The appeal to this court relates to the decision of the learned judge upon the issue of condonation.

The respondent gave particulars of the condonation relied upon by her. In these particulars it was alleged that intercourse took place between the husband and wife at specified places on the follow- ing dates in 1937, namely, June 11th, 12th, 13th, 15th, 16th, 19th, 20th, 21st and 22nd. On 11th June the wife admitted to the husband that she had committed adultery with the co-respondent in April. If the husband had intercourse after that admission, he acted with knowledge of her adultery and therefore condoned it (Cramp V. Cramp 1 ).

The particulars mentioned were given in pursuance of an order made on 19th November 1937, by which it was also ordered that the suit be tried at Hobart on a date to be fixed not earlier than April 1938. On 1st March 1938 a child was born to the wife. The birth of the child was premature by between two and four weeks, and the medical evidence was that the child was conceived in the latter part of June 1937-at or about the time fixed by the respondent for the alleged intercourse between her and her husband. The medical evidence excluded any reasonable possibility of the birth of the child being the result of the adultery with the co-respondent in April 1937.

The trial took place in May 1938. Objection was taken on behalf of the wife to the husband giving evidence denying the acts of inter course alleged. The objection was based upon Russell v. Russell 2. The learned judge overruled the objection. The wife then gave evidence that intercourse had occurred on all the occasions (except one) mentioned in the particulars given by her. The husband denied the evidence of the wife, saying that no intercourse had taken place. Both parties gave evidence of the surrounding circumstances and of their mutual relations during the important period. The learned Chief Justice believed the evidence of the husband as against the

1(1920) P. 158. 2(1924) A.C. 687. 61 CLR 389

wife, saying that, even upon the basis of the medical evidence mentioned, his firm opinion was that the husband had no sexual intercourse with his wife on any of the occasions alleged. The result was that a decree nisi for dissolution of the marriage was pronounced. The wife now appeals to this court.

In Russell v. Russell 1 the House of Lords considered the statement of Lord Mansfield in Goodright's Case 2: "It is a rule, founded in decency, morality, and policy that they" (that is, husband and wife) "shall not be permitted to say after marriage, that they have had no connection, and therefore that the offspring is spurious; more especially the mother, who is the offending party." The precise question which arose in Russell v. Russell (1) was whether the rule applied to exclude evidence by the husband of non-intercourse in proceedings for divorce founded upon adultery where (as Lord Finlay said) "the charge of adultery rests solely upon the birth of a child, which is said to be bastardized by the husband's proof of non-access 3. The legitimacy of the child depended upon whether the husband had had intercourse with his wife at the relevant time.

There is a strong presumption that a child born to a married woman is begotten by her husband. But that presumption may be rebutted by evidence of non-intercourse. Such evidence may be given in any case by other persons than husband or wife. But in legitimacy cases, that is, where the issue between the parties was the legitimacy of the child, it had long been settled that neither husband nor wife could give such evidence. In Russell v. Russell (1) it was decided that the rule also applied to evidence of non-access in divorce proceedings where such evidence was given for the purpose of establishing the illegitimacy of a child and thus proving adultery. The illegitimacy of the child was the essential point in the petitioner's case. If the child was legitimate he failed to prove adultery. Thus the evidence of non-access was tendered for the very purpose of proving the illegitimacy of an existing child.

The question which arises in the present case is whether the decision in Russell v. Russell (1) lays down a rule universally applic- able in divorce proceedings, with the result that no evidence of non-

1(1924) A.C. 687. 2(1777) 2 Cowp. 591, at p. 594 ; 98 E.R. 1257, at p. 1258. 3(1924) A.C., at p. 716. 61 CLR 390

access can be given by either husband or wife in relation to an issue of condonation where that evidence would in fact show that a child born to the wife was illegitimate. If, in the present case, the evidence of non-access is excluded, the presumption of legitimacy operates conclusively, there being no possibility of evidence in rebuttal from that presumption the inference of intercourse with the husband follows the intercourse by reason of which the child was conceived is shown by the medical evidence to have taken place late in June, that is, after the proved adultery of April and the wife's confession of adultery to the husband therefore condonation is inferred from intercourse which is itself inferred by the aid of a presumption of legitimacy.

The quite logical contention for the appellant is that no evidence of intercourse or non-intercourse by either husband or wife is admis- sible in this case upon the issue of condonation. Thus the wife could not give evidence to support her particulars of intercourse on 11th June or on any of the other days mentioned in her particulars of condonation' because, upon this contention, the wife cannot by her own evidence seek to establish any particular act of intercourse. Strictly, she should have obtained leave to amend her particulars. Such an amendment would have raised in an interesting form the question now under discussion. The relevant paragraph in her answer would have remained the same "That the petitioner condoned the said act of adultery." The particulars of acts of inter- course on specified dates would have been struck out. What would have been substituted ? Presumably some such statement as "A child was born to the respondent on 1st March 1938. The respondent will contend that therefore the petitioner had intercourse with her at some date later than the adultery which the respondent has admitted in par. 1 of the answer to the petition." The only evidence admissible under such particulars would be evidence of the birth and the date thereof-which could be provided by a birth certificate or the evidence of a doctor-and medical evidence as to the time of conception of the child. Thus the respondent could, if the contention submitted on her behalf is sound, conclusively estab- lish condonation of adultery by intercourse without herself giving any direct evidence of intercourse and, whatever the facts were,

61 CLR 391

the husband himself could give no effective evidence in rebuttal. This appeal raises the question whether particulars such as those suggested would have been good as particulars of condonation and whether condonation can be established by the evidence mentioned.

Does the decision in Russell v. Russell 1 cover such a case as the present ? No question of condonation arose in that case but both Lord Birkenhead 2 and Viscount Finlay 3 use language of wide generality. Viscount Finlay, for example, says that the rule exclud- ing evidence of non-access by either husband or wife "applies in every case in which, for any purpose, it becomes necessary to deter- mine the question whether a child born of the wife during the marriage is the child of the husband." This is, I think, the widest statement of the rule made by any of their Lordships. Lord Sumner and Lord Carson dissented from the judgment of the House and, as I shall seek to show later, Lord Dunedin was apparently of opinion that the rule did not apply in cases where condonation was the issue.

Is it then necessary in the present case to "determine" the question whether the child born to the wife is legitimate ? It was necessary to determine this very question in Russell v. Russell (1). Unless the child was illegitimate there was no evidence whatever of adultery (which was the issue) and it was therefore necessary for the court to determine the question of legitimacy. But the position appears to me to be different in the present case. In Russell V. Russell (1) the decision of the issue of adultery or no adultery depended upon a decision as to the legitimacy of the child. In the present case the issue of condonation or no condonation depends upon a decision as to whether there was or was not intercourse. If the decision that there was no intercourse, there was no condonation. No decision or determination as to the legitimacy of the child is required or involved. It is true that an inference as to the legitimacy of the child would follow from the decision that there was or was not intercourse, but that inference is irrelevant to the issue of condona- tion. In Russell v. Russell (1) the inference of illegitimacy was an essential element in deciding the issue of adultery. It therefore appears to me that the rule as expressly stated in precise terms by

1(1924) A.C. 687. 2(1924) A.C., at p. 704. 3(1924) A.C., at p. 706. 61 CLR 392

OF A. Lord Finlay should not, upon its own terms, be held to apply where,

as in the present case, the issue is one of condonation. The husband's evidence of non-access is not, in this case, "given to prove the illegitimacy of the wife's child born during the marriage," to use Lord Finlay's words 1. His evidence is tendered to prove that he did not have intercourse with his wife on certain days in June 1937. If such evidence is believed, the result may be, as a reason- able inference, that the child is illegitimate. But such an inference is not any part of his case. (Of course no decision given in these proceedings can determine the status of the child as legitimate or illegitimate; but Russell v. Russell 2 clearly decides that this circumstance is irrelevant.)

It thus appears to me that this particular statement of the rule by Lord Finlay is made in such terms that it is not applicable to cases of condonation. But, nevertheless, it must be admitted to be at least doubtful whether Lord Finlay did not state that it was SO applicable. He said We have been referred to the practice of admitting the evidence of the husband and wife in nullity suits, and in suits for cruelty as between husband and wife. Such suits have no relevance to the present question. In them there is no question of paternity, or of bastardizing issue. It is immaterial with what object such evidence is given, but, unless there is a child, and the evidence is to show that that child is not the child of the husband, the rule never comes into play at all. In condonation cases the question of bastardizing issue will, in the nature of things, very rarely, if ever, emerge. If it should, in the course of the consideration of such a defence, become necessary to ascertain the paternity of a child, non-access could no more be proved by the husband or wife than it could in other proceedings" 3. In seeking to ascertain the meaning of these words it must be remem- bered that Lord Finlay is throughout dealing with cases in which a child has been born to the wife. No question as to the admissi- bility of evidence of access or non-access arises in cases where there is no such child. Undoubtedly such evidence can be given by either husband or wife in such cases. Thus Lord Finlay, in the passage quoted, is dealing with nullity suits, cruelty suits and issues of

1(1924) A.C., at p. 712. 2(1924) A.C. 687. 3(1924) A.C., at pp. 718, 719. 61 CLR 393

condonation where there is a child. Otherwise there is no point in the observations made by the learned Lord. The point of his observations is that in the suits mentioned ' there is no question of paternity or of bastardizing issue," and, therefore, even though there may be a child, no question as to the application of the rule arises. "Such suits have no relevance to the present question." But the learned Lord proceeded to say that, if in condonation cases it should 'become necessary to ascertain the paternity of a child," the rule would be applicable. I have already stated the reasons for my view that, in the present case, it is not necessary (though it was necessary in Russell v. Russell 1 to ascertain the paternity of the child. Upon this view, the judgment of Lord Finlay, strictly con- strued, does not require the exclusion in this case of the husband's evidence of non-access. But I admit that I have much difficulty (as will more particularly appear hereafter) in understanding this part of the learned Lord's judgment.

Lord Dunedin, who reached the same conclusion in Russell V. Russell (1) as Lord Birkenhead and Lord Finlay, dealt more precisely with the scope of the rule in the form which he approved. He said

" Now as regards nullity, cruelty and condonation I do not feel the slightest difficulty; the whole point of Lord Mansfield's dictum rests on the concluding words; and to make the issue spurious, in other words, it is when conjugal conduct is used, not as a thing in itself, but as leading to other inferences that the harm comes in. No proof of conduct or want of conduct which shows nullity, no proof of cruelty, such as communicating venereal disease, no proof of connection such as in itself is condonation, has the remotest reference to the point of legitimacy of issue. The evidences [sic] of the spouses in these cases is the only evidence available to the direct fact in issue, and has in the giving of it no evil consequences " 2. In my opinion what Lord Dunedin is condemning in this passage is the use of evidence of conjugal conduct given by spouses as leading to inferences affecting the legitimacy of issue. Where such evidence is not used for that purpose it has in the giving of it no evil conse- quences." Accordingly, the learned Lord, it appears to me, regards such evidence as admissible where it relates to issues arising in a

1(1924) A.C. 687. 2(1924) A.C., at pp. 728, 729. 61 CLR 394

nullity suit or issues of cruelty or condonation. The wife relies upon the proposition, as Lord Dunedin puts it, that "proof of connection in itself is condonation." Here the evidence of the spouses is the only evidence available to the direct fact in issue," namely, the fact of intercourse. Evidence of non-intercourse is used, in such a case, to exclude the inference of condonation. It has no "reference to the point of legitimacy of issue "-it is simply evidence available to the direct fact in issue."

It is urged, however, that the ratio decidendi of Russell v. Russell 1 is that neither husband nor wife is permitted to give evidence of non-intercourse after marriage when the result of such evidence is such as to justify a conclusion that a child born of the wife during wedlock is illegitimate, even though legitimacy is not actually in issue and though no conclusion as to legitimacy is required for the purposes of the case. It is said that the reasoning of the judgments of the learned Lords who constituted the majority in Russell V. Russell (1) shows that the rule is not limited in any manner by con- siderations of the purpose or object of the evidence. There are statements in the judgments which support this view. Lord Birken- head said: The rule as laid down is not limited to any special class of case. It is absolutely general in the comprehensiveness of its expression. It has no geographical qualification. It does not, for instance, lay down that where husband and wife are present in the same bed the same bedroom; the same house or the same town, the evidence must be repelled but that it may on the other hand be received if the husband has (for instance) been absent from the country for twelve months before the birth of the child. It says, upon the contrary, that such evidence shall not be given at all and the reason given is that it would tend, if given, to bastardize the issue and to invade the very special sanctity inherent in the conjugal relation and the reason is assigned which led first the delegates and then the ordinary courts to a conclusion SO widely expressed. It is a reason founded upon decency, morality, and policy'' 2. Lord Finlay said: "The application" (of the rule)

extends to all evidence of non-access 3. Lord Dunedin said

1(1924) A.C. 687. 2(1924) A.C., at p. 698. 3(1924) A.C., at p. 718. 61 CLR 395

It is against the interests of decency and public policy that the spouses should be allowed to give evidence of non-access, and thus de facto, even if not de jure, to bastardize their issue, when conception and birth alike fell within the time of wedlock" 1.

These and other portions of the judgments were relied upon in R. V. Seaton 2, where the rule was applied in a criminal case. In that case a man was accused of incest with his daughter. It was sought to prove that she was his illegitimate daughter. Her mother was the wife of another man. The mother was not allowed to give evidence to show that her daughter was not the child of her husband. The court held that Russell v. Russell 3 decided that a married woman could not give evidence to prove the non-access of her husband during marriage where the result would be to show that a child born to her would be illegitimate. In a case like R. v. Seaton (2), if there were no other evidence of non-access, the accused would escape conviction by reason of the application of the rule, but it is evident that the application of the rule would, in other circumstances, lead to the conviction of a man for incest though in fact he was innocent. If the prisoner were charged with incest with a daughter of his wife, then, though the girl was in fact an adulterine bastard got upon his wife by another man, the application of the rule would prevent either the accused or his wife (the mother of the girl) from giving evidence to establish non-access by the husband, with the result that (if there were no other evidence of non-access) he would necessarily be convicted of incest although the girl, the other party to the offence, was not in fact related to him in any degree. It may be noted, however, that in R. v. Seaton (2) it was necessary for the jury to come to a conclusion as to the legitimacy of the girl with whom, it was alleged, the offence had been committed. The question of legitimacy was directly involved. In G. v. G. 4 (a nullity case) the Full Court of New Zealand followed and applied R. v. Seaton (2).

But although there are statements in the majority judgments in Russell v. Russell (3) which describe the rule as absolute, and as exclud- ing the evidence irrespective of its purpose, there are many statements to the contrary effect. Lord Birkenhead refers to the legitimacy of

1(1924) A.C., at p. 728. 2(1933) N.Z.L.R. 548. 3(1924) A.C. 687. 4(1934) N.Z.L.R. 246. 61 CLR 396

OF A. the child as a vital issue" in the case (1). Where there is no such

issue, there does not appear to be any reason why the rule should be applied. Lord Finlay states that the rule applies to " all cases in which the evidence is given to prove the illegitimacy of the wife's child born during the marriage" (2). "The rule evidence by husband and wife of non-access in order to show that a child born of the wife after marriage is illegitimate" (3). "This evidence is given directly for the purpose of bastardizing the child" (4). Lord Finlay appears to me to adopt both points of view in a single sentence (5). He says "It is immaterial with what object such evidence is given, but, unless there is a child, and the evidence is to show that that child is not the child of the husband, the rule

comes into play at all." This statement might be rewritten, without any change in meaning, as follows: "It is immaterial with what object such evidence is given, but, unless there is a child, and the evidence is given with the object of showing that that child is not the child of the husband, the rule never comes into play at all."

I have difficulty in understanding how it can be that the object with which the evidence is given should be immaterial, and yet that the application of the rule should depend upon the fact that the evidence is given to show that the child is illegitimate. The passage which I have already quoted from Lord Dunedin's judgment (6) condemns the use of evidence of conjugal conduct " not as a thing in itself, but as leading to other inferences "_that is, inferences as to legitimacy of issue.

Thus an examination of the reasoning in these particular parts of the majority judgments shows that what the rule excludes is evidence of husband and wife as to non-access only when such evidence is used for the purpose of proving the illegitimacy of a child of the wife. But this reasoning does not exclude such evidence where it is not used for such a purpose. In Russell v. Russell (7) the petitioner's argument passed from non-intercourse through illegitimacy to adultery. In this case the petitioner's argument passes from non- intercourse direct to the negation of the condonation which proof of

(I) (1924) A.C., at p. 701.

(4) (1924) A.C., at p. 716. (2) (1924) A.C., at p. 712.

(5) (1924) A.C., at pp. 718, 719. (3) (1924) A.C., at p. 713.

(6) (1924) A.C., at p. 728. (7) (1924) A.C. 687.

61 CLR 397

intercourse would establish. The argument does not use any evidence for the purpose of establishing illegitimacy.

But, it may be argued, this method of approach looks at the case entirely from the point of view of the husband and ignores the point of view of the wife. It may be said that the wife relies upon the presumption of legitimacy to establish intercourse and consequent condonation, while the husband's evidence of non-access is intended to be used for the purpose of rebutting the presumption of legitimacy upon which the wife relies and therefore is excluded by the reasoning in Russell v. Russell 1 to which reference has just been made. The answer to this argument is to be found in the fact that the husband does not seek to use the evidence of non-access for the purpose of rebutting the presumption of legitimacy. He uses it merely to establish the absence of the fact of intercourse which would amount to condonation. That is the only purpose of the evidence and, according to the view of the decision in Russell v. Russell (1) which I have suggested, the absence of the prohibited purpose removes the only ground upon which the evidence would be inadmissible. Upon similar reasoning evidence of non-access by the husband was admitted in a nullity suit, where the result of admitting the evidence was that bastardization of a child born to the wife logically followed (Farnham V. Farnham 2 ). A decree of nullity was pronounced propter frigiditatem quoad hunc, the evidence of non-access by the husband being admitted to negative the fact of intercourse, and not to estab- lish the illegitimacy of the child, though it had that effect. In the present case also the evidence of non-access is a denial of the fact of intercourse.

It may be added that there is some conflict in the majority opinion in Russell v. Russell (1) as to the ultimate ground of the rule. In the judgment of Lord Birkenhead the rule is said to depend upon the 'broad ground of general policy affecting the children born during the marriage as well as the parties themselves." The Lord Chancellor also refers to the protection of the sanctity of married intercourse as the ground, or a ground, of the rule 3. Viscount Finlay that the rule is absolute and wisely SO because " it is not decent

1(1924) A.C. 687. 2(1937) P. 49. 3(1924) A.C., at p. 700. 61 CLR 398

OF A. that husband or wife should give evidence to bastardize the issue

of the wife during marriage, however decorous the evidence might be in itself 1. Lord Dunedin said: "It is against the interests of decency and public policy that the spouses should be allowed to give evidence of non-access and thus de facto, even if not de jure, to bastardize their issue, when conception and birth alike fell within the time of wedlock' 2.

If the ground of the rule is to be found in the protection of the interests of the child, then prima facie the rule should exclude only evidence of the husband and wife tending to bastardize the child.

A logical application of the principle would permit evidence in favour of legitimacy (that is, evidence of intercourse) but would reject evidence against legitimacy (that is, evidence of non-intercourse). If the real principle according to which (to use the phrase of Lord Birkenhead 3 ), the rule should "be applied in the full generality of its scope to all cases which it is wide enough to cover" is that evidence of husband or wife tending to bastardize issue should not be received, then in the present case the evidence of the wife as to intercourse would be admitted but the husband would not be permitted to contradict that evidence. It would appear, however, that the general principle is not a principle which merely excludes evidence tending to bastardize a child, although it is often stated in that form. The rule, where it is applicable, excludes evidence by both husband and wife as to either access or non-access. See the statement of the rule in Halsbury's Laws of England, 2nd ed., vol. 2, pp. 562, 563 'Any direct evidence of access or non-access may be given except that neither husband nor wife is permitted to give any evidence proving or tending to disprove the fact of sexual intercourse between them." (See also R. v. The Inhabitants of Sourton, per Littledale J. 4, and Patteson J. 5; Atchley v. Sprigg 6 (approved in the Aylesford Peerage Case 7; and see Hamp v. Robinson 8 and Wright v. Holdgate 9 for applications of the rule where evidence of access was excluded.) But evidence of access by one spouse to another cannot possibly

1(1924) A.C., at p. 706. 2(1924) A.C., at p. 728. E.R., at p. 1137. 3(1924) A.C., at p. 704. 4(1836) 5 Ad. &E., at pp. 188, 189; 5(1836) 5 Ad. &E., at p. 189 111 6(1864) 33 L.J. Ch. 345, at p. 347. 7(1885) 11 App. Cas. 1, at p. 9. 111 E.R., at pp. 1136, 1137. 8(1875) 16 L.T. 29. 9(1850) 3 Car. &K. 158 175 E.R. 503. 61 CLR 399

tend to bastardize a child. Thus the general principle upon which the decision in Russell v. Russell 1 is based must be a principle which excludes evidence of access as well as of non-access and therefore cannot be a principle that evidence tending to bastardize a child born during wedlock cannot be given by husband or wife. The ground of the rule, notwithstanding statements to the contrary by the learned law Lords, must be sought in some other principle than the protection of the child against bastardization by the evidence of the mother or her husband.

The other principle suggested as the ultimate ground of Russell V. Russell (1) is the principle that the privacy of the marriage chamber shall not be invaded. Such a principle would have an advantage in logic over the other suggested principle because it would lead to the true result, namely, that where the rule applies, it excludes evidence of both access and non-access. But it is impossible to accept the proposition that no inquiries into sexual intercourse between married persons are permitted in any court. The existence of the divorce court and the normal exercise of its jurisdiction contradict such a suggestion. In cases of nullity and of cruelty connected with sexual intercourse, such inquiries are regularly and necessarily made. Where intercourse is relied upon as condonation, the court hears direct evidence of the facts from both spouses. The existence of a child is a circumstance which cannot affect any obligation to safe- guard the sanctity and the privacy of the marital chamber.

Thus, in my opinion, it is not possible to discover any principle in the majority judgments in Russell v. Russell (1) which can be relied upon to extend the decision beyond cases in which it is sought to establish adultery by a married woman by proving that a child born of her is illegitimate. Russell v. Russell (1) must be accepted as declaring the law applicable in such a case. But there is, in my opinion, no coherent principle disclosed in the case which would justify the extension of a rule long thought to be in the last stages of decay (See Law Quarterly Review, vol. 26, p. 47) to cases which have not yet been declared by compulsive authority to be subject to a rule which, in my humble opinion, belongs to a past age.

1(1924) A.C. 687. 61 CLR 400

Ever since Russell v. Russell 1 was decided the courts have, except in the New Zealand cases of R. v. Seaton 2 and G. v. G. 3, been not unwilling to distinguish it wherever respect for the authority of the House of Lords has made it possible to do so. Thus it has been held that a husband or wife may give evidence of non-access where the wife is delivered of a still-born child (Hollana v. Holland 4; or where the wife is pregnant (Aarnes v. Aarnes 5 ): or even where there is a child born, if a maintenance order is in force contain- ing a provision that the wife is no longer bound to cohabit with the husband (Andrews v. Andrews 6 or if there is a subsisting decree for judicial separation (See Hetherington v. Hetherington 7 and Russell v. Russell 8 ); or a deed of separation (Mart v. Mart 9 Stafford v. Kidd 10 ) and careful distinctions have been drawn which result in admitting evidence of adultery by a wife who has given birth to a child in SO far as it proves adultery, but in excluding such evidence in SO far as it proves non-access (Warren v. Warren 11; Justice v. Justice 12; Dick v. Dick 13; Pearce v. Kitchin 14 Roast v. Roast 15 Ex parte Letherbarrow; Re McMurray 16). The general course of judicial decision since Russell v. Russell (1) does not tend to encourage any inclination to extend the significance or effect of that decision.

I summarize what I have said with respect to Russell v. Russell (1) as follows: The case is not a decision upon condonation four out of five of the learned law Lords express their opinions in such a way as to exclude condonation from the scope of the rule; it is not possible to discover any general principle in the majority judgments which goes beyond the rejection of evidence of non-access by a husband where such evidence is tendered to establish the adultery of his wife by proving that her child born during wedlock is illegiti- mate; and there is no reason why such a rule, which operates only to exclude the best evidence of relevant facts, should be extended in every, or any, direction.

1(1924) A.C. 687. 2(1933) N.Z.L.R. 548. 3(1934) N.Z.L.R. 246. 4(1925) P. 101. 5(1929) 2 D.L.R. 298. 6(1924) P. 255. 7(1887) 12 P.D. 112. 8(1924) A.C., at p. 717 W.N. (N.S.W.) 78. 9(1926) P. 24. 10(1937) 1 K.B. 395. 11(1925) P. 107. 12(1925) S.A.S.R. 278. 13(1927) Q.S.R. 365. 14(1931) 26 Tas. L.R. 38. 15(1938) P. 8. 16(1938) 38 S.R. (N.S.W.) 281 ; 55 61 CLR 401

It may further be observed that, if in this case the husband had been able to get the case heard before the child was born, then, even though his wife was proved to be pregnant, he would have been allowed to give evidence of non-intercourse, because the rule in question would not have applied. In that case there would appar- ently have been no objectionable violation of the sanctity of marital relations and the bastardization of the child which was going to be born within possibly a few days would not have been regarded as an element of importance.

For the reasons given I am of opinion that the learned Chief Justice was right in admitting the evidence of the husband denying that he had intercourse on the dates alleged by the wife and in admitting the evidence of the wife that he did have intercourse with her on those dates.

If then the evidence was properly admitted it becomes necessary to consider whether the appeal should succeed upon the basis of the evidence actually given.

The Matrimonial Causes Act of Tasmania 1860 (No. 1), sec. 20, so far as material, provides In case the court is satisfied on the evidence that the case of the petitioner has been proved and does not find that the petitioner

has condoned the adultery complained of

then the court shall pronounce a decree declaring such marriage to be dissolved." It will be observed that the statute does not require that the court should be satisfied that there has not been condonation. It is sufficient that the court does not find that there has been condonation. Sec. 19 of the Act requires that the court should affirmatively find that the petitioner has con- doned the adultery before the petition can be dismissed on the ground of condonation.

The wife can upon this issue call in her aid two important presump- tions. The first is the presumption of legitimacy. From this presumption it follows that the husband is presumed to have had intercourse with her. (The foregoing discussion relates, of course, only to the admissibility of the evidence of the husband to rebut this presumption, not to the applicability of the presumption.) This is a rebuttable presumption (Halsbury's Laws of England, 2nd ed., vol. 2, p. 561). It means that, in the absence of other

61 CLR 402

evidence, a child will be held to be legitimate; but if other admissible 1938.

evidence shows that there has been no intercourse between husband and wife the presumption is rebutted (Aylesford Peerage Case 1; Bosvile v. Attorney-General 2; Burnaby v. Baillie 3 ). But, further, there is a presumption of fact which the wife is entitled to rely upon in the present case, namely, the presumption that, where opportunities for sexual intercourse between the husband and wife existed, sexual intercourse took place. This presumption is more easily rebutted than the other presumption mentioned because all the circumstances such as the age, health and temperament of the parties, and their friendly or hostile relations at the relevant time, may be considered for the purpose of supporting or rebutting the inference of intercourse: See the discussion of these two presumptions in In the Estate of L. 4.

In the present case the wife gave evidence of the alleged acts of intercourse and of the relations of the parties at the time. The husband denied the acts of intercourse, and his father and another witness gave evidence showing the relations which in June 1937 existed between the husband and the wife. The evidence has been very fully and critically examined by counsel upon the hearing of the appeal. The attention of the court has been directed to improbabilities in the story of the husband. It is, however, very difficult indeed to prescribe any standard of probability in matters relating to sexual relations. The character and temperament, the frigidity and excitability of individuals vary SO greatly and indeed

SO indefinitely that it is unsafe to attempt to formulate any general rules for the purpose of estimating the credibility of evidence with respect to such matters.

In the present case it is quite plain that the wife was prepared to lie to her husband where her interests appeared to her to justify a policy of deceit. At the time when, it was alleged, the acts of inter- course took place, the position was that she had been away to New South Wales for a holiday with her husband's consent. She had given him an undertaking that she would not see the co-respondent and that she was going to stay with a woman friend at Yass. In

1(1885) 11 App. Cas. 1. 2(1887) 12 P.D. 177. 40 A.L.T., at pp. 157 et seq. 3(1889) 42 Ch. D. 282. 4(1919) V.L.R., at pp. 26 et seq. : 61 CLR 403

fact she went straight to the co-respondent's home in Sydney, where he was living with his mother, and stayed there for about twelve days. She committed adultery with him there. She admitted only one act of adultery, but the correspondence between her and the co-respondent which was put in evidence makes it very difficult indeed to believe that adultery did not take place on other occasions. She told her husband untruths about her holiday in New South Wales, and for a long time denied that she had been guilty of misconduct. Eventually she admitted adultery, and it was upon the days immedi- ately succeeding this admission that, as she alleges, the acts of intercourse took place, generally in a motor car, but upon one occasion upon the beach. At this time when, according to her account, these acts of intercourse took place, she was in correspondence with the co-respondent. His letters to her were of the warmest possible character. At that time they were proposing to get married as soon as she was divorced from her husband. The husband was in a condition of anxiety, indecision, and uncertainty. This fact is established by the evidence of all the witnesses. The parties were engaged in discussions about their future, and he ultimately decided to seek a divorce upon the ground of adultery. He knew at the time that his wife was proposing to join the co-respondent and to marry him. She agreed to leave Tasmania while he obtained a divorce, upon the understanding that the divorce should be put through quietly. It is not impossible that in these circumstances intercourse took place, but it is prima facie unlikely.

After the wife left Tasmania at the end of June she wrote letters to her husband from Melbourne. On 8th (or possibly 15th) July she wrote a letter saying that the proposal to marry the co-respondent was off, that her menstrual period had been delayed and that she was very anxious about it, and that as her husband had put her

in an awkward position " he could hardly go on with the divorce." In effect she told him that she believed or feared that she was pregnant. The husband did not reply to these letters and indeed refused to speak to her when she rang him up on the telephone. He did not take the letters to his solicitor until some time afterwards. In my opinion no satisfactory inference can be drawn from this conduct of the husband. It is admitted that the letters of the wife

61 CLR 404

cannot be used as evidence of the facts which they purport to state or suggest-she cannot make evidence for herself by merely writing letters-but it is argued that the failure of the husband to reply, or at least to give them at once to his solicitor, amounts to an admission that what she wrote or suggested was true. The husband admits a difficulty in describing his state of mind, and if it is true that he did not have intercourse with her in June, it is easy to understand his uncertainty and mental upset. He said in evidence " When I got the letter I thought she was in the family way and I was not going to have anything to do with her." He said that she had promised him that she would go straight until she married the co-respondent and he (the husband) considered that he " had done enough." This appears to me to be a reasonably intelligible position and it is not irrelevant to recall the practical observation of Lord Carson in Russell v. Russell 1 that where a woman finds herself pregnant by another man than her husband, the knowledge of the conception of the child may be "the very reason why the woman invented a false story of the husband's intercourse." There is nothing in these letters inconsistent with the wife having had intercourse with some man other than her husband, though there is no evidence that she did have any such intercourse.

In this case the learned trial judge had the benefit of seeing and hearing the witnesses and of judging their character and tempera- ment, as well as the benefit of the criticisms of the evidence which have been presented (and effectively presented) to this court. As

I have already stated, it appears to me to be very difficult indeed to define standards of credibility in relation to matrimonial and sexual relationships, and I am not prepared to say that the learned judge was wrong in coming to the conclusion, which he expressed in emphatic language, that the husband was telling the truth and that the wife was lying with respect to the alleged sexual intercourse in June.

I am of opinion that the appeal should be dismissed. The co-re- spondent did not appear upon the hearing and has not appeared upon the appeal. No order should be made against him with respect to the costs of the appeal. I can see no reason why in this case the

1(1924) A.C., at p. 753. 61 CLR 405

husband should be ordered to pay the adulterous wife's costs of her unsuccessful appeal. In accordance with the request of the parties an order is made returning the case to the Tasmanian registry.

The members of the court are equally divided in opinion. The result is that, under sec. 23 (2) (a) of the Judiciary Act, the decision appealed from is affirmed.

The question of the meaning and the scope of the rule in Russell v. Russell 1 has been a matter for controversy ever since that decision was given. In South Australia it was thought wise to legislate upon the subject. Sec. 40 of the Matrimonial Causes Act 1929-1936 of South Australia is as follows: 'In any pro- ceedings under this Act either party to a marriage may give evidence proving or tending to prove that the said parties did not have sexual relations with each other at any particular time notwith- standing that such evidence would show or tend to show that any child born to the wife during marriage was illegitimate." Possibly the Parliament of Tasmania might be prepared to consider the desirability of making a definite statutory provision which would clearly either permit or exclude evidence of the character mentioned,

SO that there should be no room for controversy in the future.

STARKE J. Russell Piggott petitioned the Supreme Court of Tasmania for the dissolution of his marriage with his wife, Dorothy, on the ground of her adultery on divers occasions with Gordon Whitebrook in Sydney, New South Wales.

The wife, Dorothy, filed an answer to the petition in which she admitted that on 23rd April 1937 she committed adultery with Whitebrook but otherwise denied the allegation of adultery. She also alleged that the petitioner condoned the said act of adultery and gave particulars of acts of sexual intercourse with her husband on nine days on and between 11th and 22nd June 1937. She also alleged that the petitioner had been guilty of wilful neglect and misconduct which had conduced to the adultery.

The petitioner in his reply joined issue upon the allegation of adultery except in SO far as it was admitted and he denied the con- donation and also the wilful misconduct and cruelty alleged. The husband and wife were both domiciled in Tasmania.

1(1924) A.C. 687. 61 CLR 406

The petition was heard by the Chief Justice, and he found that the wife had committed adultery as alleged by her husband, that there was no such wilful neglect and misconduct on the petitioner's part as conduced to the adultery and upon the issue of condonation he also found for the petitioner. The wife was not, on her own admis- sion, a virtuous woman, and the Chief Justice had no doubt that she was a person capable of great deceit. Consequently a decree nisi for dissolution of the marriage was granted. Hence this appeal on her part to this court.

The matters argued upon this appeal were :-First, that the evidence of parents to bastardize issue born in wedlock by evidence of non-access is inadmissible. Bastardize the issue means giving evidence tending to establish the illegitimacy of the child (Russell V. Russell 1; In the Estate of L. 2 Wilson v. Wilson 3 ) Second, that the conclusion of the Chief Justice was contrary to the evidence and the weight of evidence. Strangely enough the plea of condonation in this case does not suggest that the legitimacy of any child born in wedlock is involved. It gives particulars of specific acts of sexual intercourse that are relied upon but makes no allega- tion of the illegitimacy of any child. In Russell v. Russell (1) it was alleged that the wife " had committed adultery with a man unknown

in consequence of which adultery' " the wife gave birth to a male child of which the said man unknown, and not your petitioner, is the father": See the statement of Viscount Finlay 4. The legitimacy of a child born in wedlock was thus, in Russell v. Russell (1), directly challenged.

The last-mentioned case is not technically, I suppose, binding upon this court, but the House of Lords propounds in it a rule of English law or rather of evidence which this court should accept unless some other authority, such as the Privy Council, more directly binding upon this court, departs from the rule or it is altered by some statutory authority. The rule in Russell v. Russell (1) has not,

I believe, convinced the judiciary or the Bar in view of the powerful and destructive criticism of Lord Sumner. Moreover there is no historical or legal basis for the rule if the criticism of Professor

1(1924) A.C. 687. 2(1919) V.L.R., at p. 24 ; 40 3(1926) V.L.R. 17; 47 A.L.T. 78. 4(1924) A.C., at p. 704. A.L.T., at p. 156. 61 CLR 407

Wigmore be accepted (Wigmore on Evidence, 2nd ed. (1923), vol. IV., pars. 2063 et seq., and the supplement (1934) to the 2nd ed., par. 2063). Again, the content of the rule itself is not very certain. The Supreme Court of New Zealand in R. v. Seaton 1 held that the rule is not confined to legitimacy cases but is a general rule of evidence applicable in all proceedings and in all courts. But it is clear that the prohibited evidence "concerns" the specific fact of non-access, i.e., testimony to any other fact constituting illegitimacy, or illegitimacy in general, is admissible (Wigmore on Evidence, 2nd ed. (1923), vol. IV., at p. 387). The parents may testify that there was no marriage ceremony or that the child was born before marriage or that one of the parents was already married to a third person (Wigmore on Evidence, 2nd ed. (1923), vol. IV., p. 388, par. 2064). R. v. Seaton 2 was a criminal case. The prisoners were jointly charged with incest. It was fundamental to the charge that the prisoners were father and daughter. The marriage of the male prisoner and his wife took place in December 1905. The female prisoner was born in 1914 during wedlock. At the trial the mother of the female prisoner deposed to facts from which it might be inferred that no sexual intercourse could have taken place between her and her husband by reason of the absence of the husband. The evidence was held inadmissible on the authority of Russell v. Russell 3. The decision was favourable to the prisoners, but the court faced, I think, the logical position that the rule might also operate prejudicially to a prisoner charged with incest in which the relation of father and daughter was fundamental to the charge. A person might be charged and put in danger of conviction of incest because of the rule that parents cannot bastardize a child born in wedlock by evidence on the part of both or either of them of non-access. If SO it is a strange rule of English law. But the Supreme Court of New Zealand had again to consider the rule in a nullity suit (G. v. G. 4 ). It was decided that the rule applied to nullity suits. On a wife's petition for nullity on the ground of impotence quoad hanc it was disclosed that a child had been born during wedlock but the rule of Russell V. Russell (3) was applied and the evidence of a spouse tending to

1(1933) N.Z.L.R., at p. 563. 2(1933) N.Z.L.R. 548. 3(1924) A.C. 687. 4(1934) N.Z.L.R. 246. 61 CLR 408

bastardize the child was rejected. But the English decisions, Farn- ham v. Farnham 1 and Burgess v. Burgess 2, are hard to reconcile with this decision. Again, the rule has no application unless a child be born alive during wedlock (Holland v. Holland 3: Fosdike V. Fosdike 4; Roberts v. Roberts 5; Aarnes v. Aarnes 6)-Cf. Wilson V. Wilson 7. Again, if the spouses are living apart under a decree or order or deed of separation, the legal presumption of access and of the legitimacy of a child born while the spouses are SO living apart is not applicable (Hetherington v. Hetherington 8 Andrews V. Andrews 9; Mart v. Mart 10; Stafford v. Kidd 11)-Cf. Bowen

V, Norman 12; In re Bromage 13; Wilson v. Wilson (7).

Finally, all that the rule excludes is evidence by a husband or wife of non-intercourse after marriage that tends to bastardize a child of the marriage. Other evidence of the spouses may be admis- sible, such, for instance, as their conduct or admissions (Warren V. Warren 14 In re Hamer's Estate Public Trustee v. Attorney- General 15; Roast v. Roast 16; Ex parte Letherbarrow; Re McMurray 17; In the Estate of L. 18; Wilson v. Wilson (7) Justice v. Justice 19 ). These cases certainly indicate the uncertain content of the rule in Russell v. Russell 20 and its irregular appli- cation in matrimonial and other matters. None of them, I think, directly decides the present case, and one must go back to the decision in Russell v. Russell (20) itself. Lord Carson thus puts the case :-

My Lords, let us take another case where the issue is condonation by resuming cohabitation. Supposing a woman avers that her husband resumed cohabitation after knowledge of her adultery and thereby condoned it, will he be allowed to give evidence that no such cohabitation took place if no child has been born, and will he be precluded from giving such evidence if in the meantime a child has been born ?-the knowledge of the conception of which may

1(1937) P. 49. 2(1937) P. 60. 3(1925) P. 101. 4(1925) 132 L.T. 672. 5(1927) 2 D.L.R. 1082 ; (1928) 1 6(1929) 2 D.L.R. 298. 7(1926) V.L.R. 17; 47 A.L.T. 78. W.N. (N.S.W.) 78. 8(1887) 12 P.D. 112. 9(1924) P. 255. 10(1926) P. 24. 11(1937) 1 K.B. 395. 12(1938) 1 K.B. 689. 13(1935) Ch. 605. 14(1925) P. 107. 15(1936) 53 T.L.R. 275. 16(1938) P. 8. 17(1938) 38 S.R. (N.S.W.) 281; 55 18(1919) V.L.R. 17 40 A.L.T. 153. 19(1925) S.A.S.R. 278. 20(1924) A.C. 687. 61 CLR 409

have been the very reason why the woman invented a false story of the husband's intercourse" 1. The noble and learned Lord Dunedin at all events faced this very position. "Let me see," he says, "how the case of adultery stands. Adultery is a fact. In the case we have here to do with there is no direct proof of the fact, but the fact is logically and properly inferred from two other facts-namely birth (which includes conception which again infers fecundation) and non-access of the husband. Now the two facts give rise, not to one, but to two logical and proper inferences they lead as I have already said, to adultery, but they also lead to illegitimacy of the child that is born. Is there any real difference, then, between this case and the cases of settlement, as to legitimacy being the true issue ? In both cases it is the solution of the underlying question, whether the legal father is the real father that determines the issue. In the one case the result is to declare a certain status, in the other it is to affirm a certain fact. I confess that, SO far as I am concerned,

I see no real difference between the two cases is said that the testimony of the spouses has been admitted in many other cases - in nullity, condonation, cruelty, and lastly, in adultery, in the divorce court of recent years. Now as regards nullity, cruelty and condonation I do not feel the slightest difficulty the whole point of Lord Mansfield's dictum rests on the concluding words:

and to make the issue spurious, in other words, it is when conjugal conduct is used, not as a thing in itself, but as leading to other inferences that the harm comes in. No proof of conduct or want of conduct which shows nullity, no proof of cruelty, such as communicat- ing venereal disease, no proof of connection such as in itself is con- donation, has the remotest reference to the point of legitimacy at issue. The evidences of the spouses in these cases is the only evidence available of the direct fact in issue, and has in the giving of it no evil consequences 2.

These passages deny and, as I understand them, are intended to deny the dilemma which perplexed Lord Carson. The conjugal conduct here is directed to the issue of condonation, the forgiveness of a matrimonial offence, and no proof of connection which is in

1(1924) A.C., at p. 753. 2(1924) A.C., at pp. 724, 725, 728, 729. 61 CLR 410

A. fact condonation has the remotest reference to the point of legitimacy.

It is quite immaterial to the issue in this case whether a child born during marriage is or is not legitimate or illegitimate it is a fact extraneous to the issue and a regrettable but irrelevant result which accidentally happens from the decision of an issue which is in its nature quite independent thereof (Farnham v. Farnham 1. Conse- quently the Chief Justice was right in admitting the evidence which has been objected to.

The other ground of appeal that the judgment is against evidence and the weight of evidence may be dismissed shortly. Counsel for the wife put the obvious question: How was the child conceived if her husband were not its father ? The medical evidence makes it fairly certain that the child was not the result of her adulterous intercourse with Whitebrook in Sydney. But she was not a virtuous woman and in addition was a deceitful woman. The Chief Justice who saw and heard her and had knowledge of the places where she deposed her husband had intercourse with her had much better opportunity of forming a right conclusion upon the matter than judges sitting in the remote atmosphere of appellate courts and with printed transcripts.

In my opinion the appeal should be dismissed, and I see no reason why the unsuccessful wife should not pay the costs of the appeal.

DIXON J. The petitioner's knowledge that his wife had committed adultery with the co-respondent was complete on 11th June 1937. She left Hobart on 23rd June 1937. The issue upon which the decree nisi now under appeal depends is whether within the twelve days covered by those dates the petitioner condoned his wife's adultery by having sexual intercourse with her. She bore a child on 1st March 1938, the paternity of which she imputes to her husband. It appears to be quite clear that it is not the co-respondent's child. He resides in Sydney. The adultery with him was committed during a visit by the respondent to New South Wales, whence she arrived back in Hobart on 4th May 1937. She certainly did not leave Tasmania between that date and her departure from Hobart. The co-respondent was in Sydney throughout the period. Probably,

1(1937) P., at pp. 58, 59. 61 CLR 411

24th April 1937 was the date when she had last seen the co-respon- dent. The child was born about a fortnight earlier than had been anticipated, and medical opinion agreed that it presented character- istics of an infant born two or three weeks before its full time. One medical witness put it down as a 255-days child. From 11th June until 1st March is 262 days, and from 23rd June is 250 days. During the period between 5th May and 23rd June 1938, the petitioner and respondent were not dwelling under the same roof. The conduct of the latter during her visit to Sydney had caused her husband, who had not accompanied her, and his father, under whose advice the son acted, to suspect her relations with the co-respondent. On her return, after an interrogation in which she admitted some familiarity but denied any guilty connection with the co-respondent, her husband refused to reopen their home to her or to rejoin her, and, after solicitors' letters had been exchanged on 7th and 10th May, husband and wife do not appear to have met until the end of May or perhaps 10th June. From 10th June until 23rd June 1937 the parties engaged upon a course of negotiations and discussions in relation to their matrimonial arrangements. A proposal for a divorce and a subsequent alliance between the co-respondent and the respon- dent seems to have been regarded with general favour. During the period of these communings opportunities occurred for sexual inter- course between husband and wife of which, according to the wife, they availed themselves. The possibility of the child having been conceived after 23rd June seems to have been small. At the hearing of the suit no third man was pointed to or suggested as a probable or possible father of the child, but this is a consideration to which, perhaps, not much weight should be attached.

The respondent, who went from Hobart to live in Melbourne, began to suspect that she was pregnant at the end of the first week in July. She at once wrote to her husband acquainting him of her fears and saying that, unless he could suggest something, he could hardly go on with the divorce as he had placed ber in an awkward position.

At the hearing before Crisp C.J., the husband's denial that he had sexual intercourse with her was accepted and the wife's evidence was disbelieved.

61 CLR 412

In support of the appeal three reasons were advanced on behalf of the wife for setting aside the decree nisi based on this conclusion. First, it was said that, as a child had been born in wedlock, evidence by the husband that he had not had access to his wife at or about the time when the child was conceived ought not to have been admitted. Secondly, it was contended that, even if it was admissible, the bare denial of the husband was not in the circumstances of the present case enough to sustain the very heavy burden which the presumption of legitimacy imposed upon him. Thirdly, the place or want of place which the learned judge gave to that presumption in his consideration of the issue of condonation is said to amount to a legal error upon the burden of proof in face of which his conclusion cannot stand. For his Honour said "It is claimed that the birth of the child creates a presumption of legitimacy but I am not trying any such issue. I am considering whether or not the petitioner condoned admitted adultery."

For reasons which will appear, I think that the question whether the husband's evidence of non-access ought to have been admitted depends upon the question how far in an issue of condonation the presumption of legitimacy applies. Both for this reason and because of the opinion I have formed as to the proper conclusion upon the facts of the case, I shall deal first with the burden of proof, the measure of persuasion required, and the sufficiency of the proofs to discharge the burden, notwithstanding that at first sight it might seem both more logical and more convenient to decide first whether the husband's evidence should have been received.

When, as in the present case, a child is born in wedlock, there is a presumption in favour of its legitimacy. ' As legitimacy involves sexual intercourse between husband and wife, there is therefore a presumption when a child is conceived and born during wedlock that such intercourse took place at a time when, according to the laws of nature, the husband could be the father of the child" (In the Estate of L. 1, per Cussen J.). This presumption applies although the question at issue is not the status of the child but the identity of the man by whose sexual connection with the mother it was begotten (Gaskill v. Gaskill 2 ). In that case, it is true, the

1(1919) V.L.R., at p. 26 40 A.L.T., at p. 157. 2(1921) P. 425. 61 CLR 413

object of proving the birth of the child was to show that the mother had had sexual intercourse with some man, SO that, upon proof that her husband was not that man, it would necessarily follow that she had committed adultery. But the decision proceeded upon the ground that the birth of a child in wedlock raised a presumption of legitimacy necessarily involving intercourse between husband and wife, a legal presumption which cast a special burden of disproof upon the husband. After referring to the nature and strength of the presumption in legitimacy cases, Lord Birkenhead L.C., who was sitting in divorce, said It is true that the observations were made in reference to a legitimacy suit, but I cannot conceive that in the present case any different principle can apply; otherwise it might happen that the mother would be condemned for adultery on evidence which would not disentitle the child to be declared to be the legitimate issue of her husband" 1. If the presumption SO operates where the reason why the husband proposes to disprove any relevant congress between himself and his wife is because an inference of adultery would necessarily follow, it must apply also when he desires to disprove sexual relations between them simply because such relations spell condona- tion. The presumption is a presumption of law. It throws over the burden of proof. Further, it sets a special standard of persuasion. More is required to rebut the presumption than proof of the contrary to the reasonable satisfaction of the tribunal of fact. A higher degree of persuasion must be reached. The presumption of legitimacy must be clearly distinguished from the presumptive conclusion or prima- facie inference which the law authorizes when it appears that oppor- tunities existed for sexual intercourse between husband and wife. From the occurrence of such opportunity the inference may be drawn that sexual intercourse occurred. But it is not a definite presumption of law concluding the matter unless and until proof to the contrary is adduced. A discussion of the part that may be played by this prima-facie inference in assisting or supporting the presumption of legitimacy will be found in the judgment of Cussen J. in In the Estate of L. 2. He speaks of it as " a presumption merely of fact or of ordinary experience," and again as "a prima-facie conclusion or

The medical evidence proves that the child to which the appellant gave birth on 1st March 1938 was premature and was probably conceived about 255 days before birth. The presumption of law that the child was conceived in lawful wedlock operates, and, unless it is repelled by sufficient evidence, there is in the present case presumptive proof that sexual intercourse took place between the appellant and her husband about 255 days before 1st March 1938, that is, after she returned to Tasmania from her adulterous excursion to New South Wales. It should be observed that the circumstances of time and place exclude the possibility of the co-respondent being the father of the child. Indeed, that was not disputed by the husband, who rested on the denial that he was not the father of the child. If he had relied on the birth of the child as evidence of the appellant's adultery, it would have been unmistakably contrary to the decision in Russell v. Russell (1) for him to seek to complete the chain of proof by giving evidence that he did not have sexual inter- course with his wife during the period in which the child could have been conceived. Although the husband's evidence of non-access is tendered on the issue of condonation, it denies the legitimacy of the

1(1924) A.C. 687. 2(1811) 1 Sm. &St. 153 ; 57 E.R. 7 E.R., at p. 385. 3(1837) 5 Cl. &Fin., at p. 215; 4(1921) P., at p. 433. 61 CLR 423

child no less than if it were tendered in proof of a charge of adultery based on the birth of the child.

There is a high and explicit authority for the appellant's submission that it was contrary to law to admit evidence denying sexual inter- course with his wife, which the husband gave in the present case. Viscount Finlay said in Russell v. Russell 1: If it should, in the course of the consideration of such a defence" (i.e. condonation),

become necessary to ascertain the paternity of a child, non-access could no more be proved by the husband or wife than it could in other proceedings." This statement is, in my opinion, an instance of the application of the law which was laid down by the majority in that case. It is useful to refer in some detail to their reasons in order to see the scope and operation of the doctrine which they expounded. The Earl of Birkenhead stated that the question which was to be decided was whether or not by the law of England evidence of non-access may, in proceedings for divorce, be tendered by a spouse and received by a court with the object or possible result of bastardizing a child of the marriage" " 2. The Lord Chancellor gave this short answer to that question: "I have formed the clear opinion that such evidence is not receivable; that it ought not to have been allowed to go to the jury; and that therefore, unless there was other evidence proper to go to them, the verdict cannot stand (2). In the reasons which were given for this opinion the Lord Chancellor, after observing 3 that it was conceded every- where that the rule laid down in Goodright's Case 4 applies to legitimacy proceedings, said that the task of the court was to deter- mine " whether evidence inadmissible in such proceedings is admis- sible in divorce." The Lord Chancellor explicitly stated that

'the rule as laid down is not limited to any special class of case. It is absolutely general in the comprehensiveness of its expression (3). Again he said When we are told that a rule is founded upon public policy, decency, and morality, it would seem natural to propose it in all cases to which it applies verbally, provided that we are still able to bring ourselves within the public considerations which were the expressed basis of the rule. If, for instance, in an

1(1924) A.C., at p. 719. 2(1924) A.C., at p. 697. 3(1924) A.C., at p. 698. 4(1777) 2 Cowp. 591 ; 98 E.R. 1257. 61 CLR 424

A. issue where the child himself is a party-a legitimacy proceeding

in rem in the true sense-it is against public policy to admit the evidence of a parent to prove the bastardy of that child, why should an entirely different policy permit such evidence in the case where a vital issue is still the legitimacy of the child, even though it be raised for a different purpose, and perhaps with secondary emphasis?" 1. One further passage makes it abundantly clear that the doctrine which the Lord Chancellor was enunciating has a general application in divorce: It is said that the view recommended above will revolutionize the practice of the divorce court. And my attention was called to the fact that in an undefended divorce case which I myself tried at nisi prius I had admitted evidence of non-access, bastardizing the issue, and given by a soldier petitioner. The matter was not argued SO that my attention was not directed to the point. With argument, I am sure that I should have reached my present conclusion. In such cases the non-access can almost always be established aliunde. If it cannot be SO proved the practice of the divorce court must accommodate itself to the authority of the rule" 2.

Viscount Finlay gave reasons for judgment which leave no room for argument that any rule of less general application was being explained than that expounded by the Lord Chancellor. Viscount Finlay said that the substantial question was whether the evidence of the respondent as to non-access to his wife was rightfully admitted and made the following statement with regard to the presumption of legitimacy :- There is a strong presumption that the child of a married woman was begotten by her husband. This, however, is not a presumption juris et de jure it may be rebutted by evidence The fact that the wife had immoral relations with other men is not of itself sufficient to displace the presumption of legitimacy non- access by the husband at the time when the child must have been begotten must (unless there be incapacity) further be proved. Proof of non-access cannot be given for this purpose either by the husband or by the wife neither of them can be asked any question tending to prove such non-access; it must be established entirely by the evidence of other witnesses" 3. Referring to this rule,

1(1924) A.C., at pp. 700, 701. 2(1924) A.G., at p. 702. 3(1924) A.C., at pp. 705, 706. 61 CLR 425

Viscount Finlay observed 'It applies in every case in which, for any purpose, it becomes necessary to determine the question whether a child born of the wife during the marriage is the child of the husband 1. The rule, as had been observed by the Lord Chan- cellor in the course of his judgment, is based on Lord Mansfield's observations during the argument in Goodright v. Moss 2. And, speaking of those observations, Viscount Finlay said it was clear that Lord Mansfield's observation, "bastardizing the issue," is not confined to proceedings for a formal declaration as to legitimacy or the reverse. Viscount Finlay said: "The expression" (bastardiz- ing the issue) "merely denotes giving evidence to show that the child is illegitimate, and this, in the case of a child after marriage, no parent can be admitted to give" " 3. Referring specifically to the case before them, Viscount Finlay continued: The whole object of the evidence in the present case is to bastardize the child SO as to prove the adultery. The respondent's argument on this point rests on the misapprehension of the meaning of the term bastardize ¹ employed by Lord Mansfield in Goodright v. Moss (2), to which I have adverted in commenting on that case. The term covers all cases in which the evidence is given to prove the illegitimacy of the wife's child born during the marriage" " 4. Viscount Finlay returned to the same argument and again said very clearly that it was quite unsound: It was suggested on behalf of the respondent that the rule of evidence as to non-access not being proved by husband or wife does not apply to a suit for dissolution of the marriage. There is no authority for this proposition, and it is contrary to principle. Divorce is legal, but the party who wishes to obtain divorce is not dispensed from observing the rules of evidence. In the present case, the charge of adultery rests solely upon the birth of the child, which is said to be bastardized by the husband's proof of non-access. It falls within the very terms of the rulings of Lord Mansfield and Lord Ellenborough. This evidence is given directly for the purpose of bastardizing the child, and is the only evidence of the adultery" 5. This view is again confirmed by the following statement: It appears that in the divorce court a practice has

1(1924) A.C., at p. 706. 2(1777) 2 Cowp. 591 ; 98 E.R. 3(1924) A.C., at p. 708. 4(1924) A.C., at pp. 711, 712. 5(1924) A.C., at p. 716. 61 CLR 426

grown up of admitting such evidence by husband and wife, to the extent of allowing proof by them as to absence at the material time. The question now arises whether this practice is right. I have given my reasons for thinking that it is erroneous and that the rule excluding such evidence is not confined to legitimacy cases,' as the Master of Rolls thought, whatever that expression may denote, and that its application extends to all evidence of non-access" 1. There is one other passage in the judgment of Viscount Finlay which should be noticed. It contains the clear statement, which has already been quoted, to the effect that, if, in the course of the consideration of a defence of condonation, it should become necessary to ascertain the paternity of a child, neither the husband nor the wife could prove non-access :- We have been referred to the practice of admitting the evidence of the husband and wife in nullity suits, and in suits for cruelty as between husband and wife. Such suits have no relevance to the present question. In them there is no question of paternity, or of bastardizing issue. It is immaterial with what object such evidence is given, but, unless there is a child, and the evidence is to show that that child is not the child of the husband, the rule never comes into play at all. In condonation cases the question of bastardizing issue will, in the nature of things, very rarely, if ever, emerge. If it should, in the course of the consideration of such a defence, become necessary to ascertain the paternity of a child, non-access could no more be proved by the husband or wife, than it could in other proceedings " " 2.

A difficulty was raised by the contention that Lord Dunedin, a member of the majority in Russell v. Russell 3, used words which meant that a case depending on the issue of condonation was not within the operation of the rule laid down by the decision. His words, relied on for this contention, are " But then it is said that the testimony of the spouses has been admitted in many other cases --in nullity, condonation, cruelty, and, lastly, in adultery, in the divorce court of recent years. Now as regards nullity, cruelty and condonation I do not feel the slightest difficulty; the whole point of Lord Mansfield's dictum rests on the concluding words: and to

1(1924) A.C., at p. 718. 2(1924) A.C., at pp. 718, 719. 3(1924) A.C. 687. 61 CLR 427

make the issue spurious, in other words, it is when conjugal conduct is used, not as a thing in itself, but as leading to other inferences that the harm comes in. No proof of conduct or want of conduct which shows nullity, no proof of cruelty, such as communicating venereal disease, no proof of connection such as in itself is condona- tion, has the remotest reference to the point of legitimacy of issue. The evidences of the spouses in these cases is the only evidence available to the direct fact in issue, and has in the giving of it no evil consequences 1. It seems to me that Lord Dunedin's observations in respect to condonation, with which alone we are here concerned, were intended to refer to a case in which no child was born subsequently to the intercourse relied on as proving con- donation. I think the words quoted are intended to explain that the rule is confined to cases where the evidence relates not merely to "conjugal conduct

as a thing in itself," but leads to other inferences -or, specifically, in the limiting words of Lord Mansfield's dictum, where the evidence would "make the issue spurious." Moreover, the meaning which the respondent contends that the words just quoted have is inconsistent with Lord Dunedin's reason for saying that evidence of non-access cannot be given by husband or wife for the purpose of proving a charge of adultery based on the birth of a child. Lord Dunedin said Now let me see how the case of adultery stands. Adultery is a fact. In the case we have here to do with there is no direct proof of the fact, but the fact is logically and properly inferred from two other facts- namely, birth (which includes conception which again infers fecunda- tion) and non-access of the husband. Now the two facts give rise, not to one, but to two logical and proper inferences they lead, as

I have already said, to adultery, but they also lead to illegitimacy of the child that is born. Is there any real difference, then, between this case and the cases of settlement, as to legitimacy being the true issue In both cases it is the solution of the underlying question, whether the legal father is the real father, that determines the issue. In the one case the result is to declare a certain status, in the other it is to affirm a certain fact. I confess that, SO far as I am concerned,

I see no real difference between the two cases, SO that, according to

1(1924) A.C., at pp. 728, 729. 61 CLR 428

A. my thinking, the dictum of Lord Mansfield applies in terms 1.

Another observation clearly states the rule which Lord Dunedin intended to lay down "I am therefore of opinion that the words of Lord Mansfield are directly applicable to this case, and that it is against the interests of decency and public policy that the spouses should be allowed to give evidence of non-access, and thus de facto, even if not de jure, to bastardize their issue, when conception and birth alike fell within the time of wedlock" 2.

In Russell v. Russell 3 the evidence of non-access tendered for the purpose of proving adultery was held to be inadmissible because it would bastardize the child. The husband's evidence in the present case, although tendered on a different issue, would also bastardize the child. In either case, if the evidence of non-access were admitted, the mischief which Lord Mansfield's rule aimed at avoiding, namely, the bastardizing of a child born to the wife, would be done by the evidence. Moreover, the proof of condonation by the evidence of non-access in this case directly involves the proof of an act of adultery by the appellant SO that, when this is borne in mind, there is no real ground for discrimination between this case and Russell v. Russell (3) for the purposes of applying the rule. In my opinion, the rule in Russell v. Russell (3) operates to exclude the evidence of non-access tendered on the issue of condonation, for the reason that the evidence would bastardize the child born to the appellant on 1st March 1938.

The question now arises whether the residue of the evidence, after the omission of that excluded by the rule in Russell v. Russell (3), is sufficient to rebut the presumption that sexual intercourse resulting in the conception of the child took place between husband and wife. In Gaskill v. Gaskill 4 Viscount Birkenhead, adopting the language of Lord Lyndhurst, in Morris v. Davies 5 in relation to the case of Head v. Head 6, said that in order to rebut that presumption, which is one of law, the court must be satisfied that sexual inter- course did not take place " not upon a mere balance of probabilities, but upon evidence which must be such as to exclude all doubt,

1(1924) A.C., at pp. 724, 725. 2(1924) A.C., at pp. 727, 728. 3(1924) A.C. 687. 4(1921) P., at p. 433. 5(1837) 5 CI. &Fin., at p. 215: 7 6(1823) 1 Sim. &St. 150 ; 57 E.R. 61 CLR 429

61 C.L.R.]

OF AUSTRALIA. that is, of course, all reasonable doubt, in the minds of the court or jury, to whom that question is submitted." Viscount Birkenhead clearly disapproved of a mother being condemned for adultery on evidence which would not disentitle the child to be declared the legitimate issue of her husband. In the present case the strength of the husband's reply to the plea of condonation at the trial rested on his explicit denial of sexual intercourse. In the absence of this evidence, I am not satisfied that there is evidence upon which the court should hold that the presumption of law is rebutted. The residue of evidence is, in my opinion, insufficient to satisfy the standard of proof applicable to the case.

There is, therefore, presumptive proof, which has not been rebutted, that the husband had sexual intercourse with the appellant at a time in or about 255 days before the birth of the child on 1st March 1938. This time would be in or about the third week in June 1937. But, before the defence of condonation can succeed, it must be shown that, at the time the act or acts relied on as constituting the condonation were done, the husband knew of, or believed in, the wife's adultery. The trial judge found that, at the time of the respondent's return to Tasmania on 4th May 1937, "he" (the husband)

was satisfied that she had been unfaithful to him, and he refused to allow her to return to his home." The judge added He, how- ever, lacked proof of her misconduct, and, after her return, he saw her frequently and pressed her to tell him the truth. At last, on 11th June, she broke down and confessed. The petitioner was greatly distressed, and I believe him when he said that he did not know what to do." On this occasion the husband-as is clear from his own evidence---led the appellant to the point where she "broke down' by suggestions that she had been watched in Sydney. Her first admission of any improper conduct with the co-respondent was made in answer to the two questions of her husband "What about the sofa at Gordon's residence ? " and, "Do you admit something happened on the sofa ? " After giving evidence of the respondent's confession of adultery, the husband continued "I asked her if she would be prepared to admit her adultery to a third party. She didn't like the idea at first."

61 CLR 430

It is clear that if any act of sexual intercourse relied on as con- stituting condonation happened on or after 11th June, the husband had that degree of knowledge or belief which is a necessary element in condonation for he had the certain knowledge based on her confession. It is possible, however, that the sexual intercourse presumed to have taken place occurred before that date and no sexual intercourse took place on or after it, because I am not satisfied that the date of conception has been fixed within limits accurate enough to place it with certainty after 11th June.

It is necessary, therefore, to consider whether the husband had the necessary degree of knowledge or belief before 11th June. I am satisfied from the findings of the trial judge and the evidence of the husband quoted above that he was quite convinced, whatever was the source of his conviction, that his wife had been living in adultery during her stay in Sydney. His conversation before her confession and his request to her after it to make the admission before a third party indicate that he had a firm belief in her guilt, probably some actual information of the details of her adultery, and sought to obtain admissions of which evidence could be given in court. When attempts have been made by the courts to define condonation, frequently the word 'knowledge " simply has been used to describe the state of mind required of the condoning party. In Bernstein V. Bernstein 1, the words "full knowledge " were used. Lopes J. said: "Condonation

means the complete forgiveness and blotting out of a conjugal offence, followed by cohabitation, the whole being done with full knowledge of all the circumstances of the particular offence forgiven." He added " The husband, in my opinion, need not be aware of all the acts of adultery committed by the wife when he forgives her any particular act of adultery. Condonation means a full and absolute forgiveness, with knowledge of all that is forgiven' " 2. In the same case A. L. Smith L.J. used similar words: "In my judgment, the law as to condonation was accurately and clearly stated by Sir Cresswell Cresswell in Keats V. Keats 3, where he described it as ' a blotting out of the offence imputed, SO as to restore the offending party to the position he or

1(1893) P. 292. 2(1893) P., at p. 303. 3(1858) I Sw. &Tr., at p. 346 ; 164 61 CLR 431

OF AUSTRALIA she occupied before the offence was committed,' and again in Peacock V. Peacock 1, where it is described as the forgiveness of the conjugal offence with the full knowledge of all the circumstances attending it " 2. But where the precise nature of the knowledge has been more in issue, more explicit statements relating to the necessary state of mind of the condoning party have been made. In Keats V. Keats 3 Sir Cresswell Cresswell charged the jury in the following words " Again, it has been held that the person condoning, in order to condone, must know of the offence, otherwise he cannot be supposed to have condoned it. That perhaps is not strictly applicable in all cases, because a man may condone whether he knows of the offence or not, in this way: he may say, I have heard stories about my wife. A. and B. have told me that she has committed adultery.

I can hardly believe it. I am in doubt about it; but whether guilty or not, I will take her back; she shall be restored to my bed.' That would be condonation without actual knowledge and I think if after this he took her to his bed again, he could not afterwards, on acquiring more certain knowledge, revive the charge, as to which he had himself said, ' I care not whether it is true or false; I do no know whether it is true or not but be it one or be it the other,

I would equally take her back to my bed.'" The learned Judge Ordinary was upheld by the Full Court. The Lord Chancellor (Lord Chelmsford) said: "I see no reason at all to differ with the Judge Ordinary in the way in which he has stated the law of con- donation to the jury " 4. In Ellis v. Ellis 5 it was held that

in order to establish condonation, it is not enough to prove that the husband took his wife back after certain facts had come to his knowledge, after certain intelligence had been communicated to him tending to prove her adultery; it is necessary to prove that the husband took his wife back with the intention of forgiving her, believing her to be guilty." The Judge Ordinary found that the husband had not condoned his wife's adultery because he did not believe the facts at the time they were communicated to him by

1(1858) 1 Sw. &Tr. 183 164 E.R. 2(1893) P., at p. 312. 3(1858) 1 Sw. &Tr., at p. 346 : 164 164 E.R. 1475, at p. 1476. E.R., at p. 759. 4(1859) 1 Sw. &Tr., at p. 357; 164 5(1865) 4 Sw. &Tr. 154, at p. 157 61 CLR 432

a certain person, being convinced that that person was actuated

by spite in making the charges against his wife. In Cramp v. Cramp PIGGOTT

1 McCardie J., referring to the case of Bernstein v. Bernstein 2 PIGGOTT.

said "This requirement of knowledge however must always be considered in connection with the words of Sir Cresswell Cresswell in Keats v. Keats 3 where he pointed out that a husband may be in doubt as to his wife's guilt, but may say: Whether guilty or not I will take her back and she shall be restored to my bed.'" McCardie J. refers to the knowledge necessary in the words, "It is necessary that the spouse who condones should be substantially aware of the matrimonial sin committed " 4.

In my opinion, the condition of belief or knowledge laid down by these decisions to be a necessary element in condonation was satisfied by the amount of belief or knowledge of the appellant's adultery which he had during the time when, by presumption of law, he was proved to have had sexual intercourse with the appellant. It follows that her defence of condonation has been established for a husband who has sexual relations with his wife, after knowledge of her adultery, must be conclusively presumed to have condoned her offence (Cramp v. Cramp 5 ). The law recognizes that it would be unjust to a wife and immoral in a husband for him to enjoy such marital rights and seek, at the same time, to withhold his forgiveness.

For these reasons, I think that the appeal should be allowed and that the petition should be dismissed.

Appeal dismissed. No order as to costs of appeal.

Cause transferred to Tasmanian registry. Solicitors for the appellant, Crisp &Wright, Hobart, by J. W. Mound &Kelynack.

Solicitors for the respondent, Page, Hodgman, Seager &Doyle, Hobart, by Pigott, Stinson, Macgregor &Palmer.

1(1920) P., at p. 164. 2(1893) P. 292. 3(1858) I Sw. &Tr. 334 164 E.R. 754. 4(1920) P., at p. 163. 5(1920) P. 158