PGA v The Queen

M G Hinton QC, Solicitor-General for the State of South Australia with K G Lesses for the respondent and intervening on behalf of the Attorney-General for the State of South Australia (instructed by Director of Public Prosecutions (SA))

S J Gageler SC, Solicitor-General of the Commonwealth with G A Hill intervening on behalf of the Attorney-General of the Commonwealth (instructed by Australian Government Solicitor)

J D McKenna SC with G J D del Villar intervening on behalf of the Attorney-General of the State of Queensland (instructed by Crown Law (Qld))

Notice: This copy of the Court's Reasons for Judgment is subject to formal revision prior to publication in the Commonwealth Law Reports.

Criminal law - Rape - Husband's immunity from prosecution for rape of wife - Presumption of consent to intercourse by wife in marriage - Appellant charged in 2010 with two counts of rape contrary to s 48 of Criminal Law Consolidation Act 1935 (SA) - Alleged rapes committed in 1963 against then spouse - Legislative amendments enabled institution of proceedings despite lapse of time - Elements of offence of rape in 1963 supplied by common law - Whether in 1963 common law of Australia presumed consent by wife in marriage.

Precedent - Judicial method - Development of common law - Whether presumption of consent by wife in marriage was part of common law of Australia - Whether statement of common law in R v L (1991) 174 CLR 379 applied to events alleged to have occurred in 1963.

Words and phrases - "common law", "marital exemption", "marital immunity", "presumption of consent", "rape", "retrospective application".

Criminal Law Consolidation Act 1935 (SA), s 48. Matrimonial Causes Act 1857 (UK) (20 & 21 Vict c 85).

FRENCH CJ, GUMMOW, HAYNE, CRENNAN AND KIEFEL JJ. The appellant and his wife, the complainant, were lawfully married in South Australia on 1 September 1962. At the relevant times in 1963 they remained lawfully married and were cohabiting in South Australia as husband and wife at the house of her parents; there were in force no legal orders or undertakings of any kind which affected their matrimonial relationship.

On 5 July 2010, by information of the Director of Public Prosecutions of South Australia, the appellant was charged for trial in the District Court of South Australia with two counts of carnal knowledge, with four counts of assault occasioning actual bodily harm and, what is immediately relevant for this appeal, with two counts of rape (counts 3 and 5) contrary to s 48 of the Criminal Law Consolidation Act 1935 (SA) ("the CLC Act"). The particulars of count 3 were that between 22 March 1963 and 25 March 1963, at Largs Bay in South Australia, the appellant had vaginal sexual intercourse with his wife without her consent. The particulars of count 5 were that on or about 14 April 1963, also at Largs Bay, the appellant had vaginal sexual intercourse with his wife without her consent.

The issue before the Court is whether the appellant is correct in his contention that, as a matter of the common law, upon their marriage in 1962 his wife had given her consent to sexual intercourse and thereafter could not retract her consent, at least while they remained lawfully married, with the result that he could not be guilty of raping her as charged by counts 3 and 5.

The proposition of law upon which the appellant relies has its source in a statement in extra‑judicial writings of Sir Matthew Hale, Chief Justice of the Court of King's Bench (1671‑1676), which were first published in 1736 as The History of the Pleas of the Crown. The statement by Hale is more fully set out later in these reasons, but is encapsulated in the bald proposition that a husband cannot be guilty of a rape he commits upon his wife. It was repeated in East's work A Treatise of the Pleas of the Crown, published in 1803; by Chitty in his A Practical Treatise on the Criminal Law, published in 1816; and by Russell in A Treatise on Crimes and Misdemeanors, the first edition of which was published in 1819. In each case the proposition was further repeated in later 19th century editions. What, however, was lacking in all these standard texts was any statement and analysis of reasoning which might have supported the statement by Hale and its continued acceptance.

Given this state of affairs, it is perhaps not surprising that the Canadian Criminal Code of 1892 (s 266) and the Criminal Code of Queensland of 1899 (s 347), in defining the crime of rape, included the phrase "not his wife". The provisions in the Queensland Code, and those of Western Australia and Tasmania, were to be amended in 1989, 1985 and 1987 respectively. The attempted abstraction and statement of doctrine in provisions of a code by means of propositions which do not represent generalised deductions from particular instances in the case law occasions difficulty when the common law later is shown to be to different effect. Justice Holmes, in his essay "Codes, and the Arrangement of the Law", wrote:

"New cases will arise which will elude the most carefully constructed formula. The common law, proceeding, as we have pointed out, by a series of successive approximations - by a continual reconciliation of cases - is prepared for this, and simply modifies the form of its rule. But what will the court do with a code? If the code is truly law, the court is confined to a verbal construction of the rule as expressed, and must decide the case wrong. If the court, on the other hand, is at liberty to decide ex ratione legis, - that is, if it may take into account that the code is only intended to declare the judicial rule, and has done so defectively, and may then go on and supply the defect, - the code is not law, but a mere text‑book recommended by the government as containing all at present known on the subject." Indeed, in 1888, among the 13 judges sitting in the Court for Crown Cases Reserved, on the case stated in R v Clarence with respect to charges of "unlawfully and maliciously inflicting grievous bodily harm" and "assault occasioning actual bodily harm", contrary to s 20 and s 47 respectively of the Offences against the Person Act 1861 (UK) ("the 1861 UK Act"), differing views had been expressed as to whether the consent of the wife to intercourse with her husband had been vitiated by his failure to disclose to her that he was suffering from a contagious venereal disease.

Thereafter, in the annotation to s 48 of the 1861 UK Act which appeared in Halsbury's Statutes of England, published in 1929, it was said:

"It is said that a husband cannot be guilty of rape upon his wife as a principal in the first degree". (emphasis added) The 28th edition of Archbold's Pleading, Evidence & Practice in Criminal Cases, published in 1931, four years before the enactment of the CLC Act, cited Hale for the proposition expressed as:

"It is a general proposition that a husband cannot be guilty of a rape upon his wife ... but it would seem that the proposition does not necessarily extend to every possible case". In the intervening period there appears to have been no reported case in England in which a husband had been prosecuted for the rape of his wife during their cohabitation.

"Any person convicted of rape shall be guilty of felony, and liable to be imprisoned for life, and may be whipped." It is accepted that the elements of the offence of rape identified in s 48 were supplied by the common law.

Section 4 of the CLC Act had wholly repealed The Criminal Law Consolidation Act 1876 (SA). As amended by s 13 and the Schedule to the Criminal Law Amendment Act 1925 (SA), s 60 of the 1876 statute had read:

"Whosoever shall be convicted of the crime of rape shall be guilty of felony, and, being convicted thereof, shall be liable to be imprisoned for life, with hard labor, and may be whipped." The scheme of the legislation in South Australia, in its various forms, was to classify the offence of rape as a felony and to specify the range of punishments upon conviction. This followed the pattern in s 48 of the 1861 UK Act. The legislative emphasis upon the classification of the crime and the punishments which might be inflicted, leaving the elements of the crime itself to the common law, reflected past fluctuations in the statute law. Shortly after the enactment of the 1861 UK Act, there appeared in the 5th edition (1877) of Russell's work, A Treatise on Crimes and Misdemeanors, the following:

"This offence formerly was, for many years, justly visited with capital punishment; but it does not appear to have been regarded as equally heinous at all periods of our Constitution. Anciently, indeed, it appears to have been treated as a felony, and, consequently, punishable with death; but this was afterwards thought too hard; and, in its stead, another severe but not capital punishment was inflicted by William the Conqueror, namely, castration and loss of eyes, which continued till after Bracton wrote, in the reign of Henry III. The punishment for rape was still further mitigated, in the reign of Edward I, by the statute of Westm 1, c 13, which reduced the offence to a trespass, and subjected the party to two years' imprisonment, and a fine at the King's will. This lenity, however, is said to have been productive of terrible consequences; and it was, therefore, found necessary, in about ten years afterwards, and in the same reign, again to make the offence of forcible rape a felony, by the statute of Westm 2, c 34. The punishment was still further enhanced by the 18 Eliz c 7, s 1." The lapse of time

Something should be said respecting the legal significance of the length of time between the alleged conduct in 1963 and the institution of proceedings in 2010. As the CLC Act stood in 1963, it included s 76a. The effect of s 76a was that in respect of offences, including an offence against s 48, no information was to be laid more than three years after the commission of the offence. Section 76a was repealed by the Criminal Law Consolidation Act Amendment Act 1985 (SA). However, in R v Pinder it was held that the repeal of s 76a did not authorise the laying of an information which would deprive a person of immunity already acquired before the repeal of s 76a. The response of the legislature was to reverse the effect of this decision by the enactment of s 72A of the CLC Act by the Criminal Law Consolidation (Abolition of Time Limit for Prosecution of Certain Sexual Offences) Amendment Act 2003 (SA). The result was that a person, such as the appellant, who had acquired immunity by reason of the operation of the repealed s 76a had lost that immunity and could now be prosecuted.

Changes have been made to the elements of the offence of rape, beginning with the Criminal Law Consolidation Act Amendment Act 1976 (SA), but it has not been submitted that these changes to the elements of the offence apply retrospectively.

On 6 July 2010 Herriman DCJ gave reasons for dismissing an application by the appellant for a permanent stay of proceedings. His Honour's reasons included the following passage:

"The complainant's evidence is that in 1960 and 1961, when she was 15 or 16, the accused was in a relationship with her and she says that at that time they were living in her parents' house, albeit that he slept in a separate room. They were ultimately married in September 1962, when she was 17, but she says that before that age she had sexual intercourse with him on two occasions. Those two occasions represent counts 1 and 2 on the information. The parties then lived as husband and wife in her parents' house until mid‑1963, when they went to their own premises. They separated in 1969. The complainant says that on two occasions, in March and April 1963, which she relates to times immediately before and soon after the birth of their first child, the accused had forcible sexual intercourse with her against her will. She says that she did not, at any time during the marriage, complain of carnal knowledge or, indeed, of that forced sexual intercourse. The time for laying of any such charges was then within three years of the act, so that the time for laying a complaint with respect to the carnal knowledge counts expired in about 1964 and, with respect to rape, in about 1966. Those time limits were not abolished until the year 2003. More importantly, there was, and, indeed, there remains, a real question as to whether in 1963 an offence of rape in marriage, as it is commonly called, was then part of the common law of this State." His Honour went on to stay the trial pending the statement for the Full Court of the Supreme Court of South Australia of a case under s 350(2)(b) of the CLC Act. This dealt with the argument of the appellant that at the time of the alleged offences in 1963, he could not, as a matter of law, have committed the crime of rape upon his wife.

What was said in 1991 by four of the five members of this Court in R v L has been treated by the parties in the present litigation at least as having the result that by 1991 it was no longer the common law in Australia that by marriage a wife gave irrevocable consent to sexual intercourse with her husband. Herriman DCJ saw the outstanding issue for determination as being "was the offence of rape by one lawful spouse of another ... an offence known to the law of South Australia as at 1963?". A question to this effect was stated for consideration by the Full Court. The Court (Doyle CJ and White J; Gray J dissenting) ordered that the question be answered as follows:

"The defendant is liable at law to be found guilty of the offences of rape charged in count 3 and count 5 of the Information, notwithstanding that at the time of the alleged offence he was married to the alleged victim and was cohabiting with her, the marriage giving rise to no presumption of consent on her part to intercourse with her husband, and giving rise to no irrebuttable presumption to that effect." Gray J was of the contrary opinion and would have answered the question in the negative and applied the presumption of irrevocable consent.

By special leave the appellant appeals to this Court seeking an order setting aside the answer given by Doyle CJ and White J. By Notice of Contention the respondent submits that, regardless of what follows from the decision in R v L, the answer by Doyle CJ and White J, the majority in the Full Court, is to be supported on the basis that: (a) "the supposed marital exemption to the offence of rape ... was never part of the common law of Australia"; or (b) "if it ever was part of the common law of Australia, it ceased to be so as at the date of the commission of the offences in this matter".

For the reasons which follow, if the "marital exemption" ever was part of the common law of Australia, it had ceased to be so by the time of the enactment in 1935 of s 48 of the CLC Act and thus before the date of the commission of the alleged offences charged as count 3 and count 5. It follows that the appeal must be dismissed. That conclusion does not involve any retrospective variation or modification by this Court of a settled rule of the common law. At the time of the commission of the alleged offence the common law rule for which the appellant contends did not exist.

The references above to "the common law" and "the common law of Australia" require further analysis before consideration of the immediate issue concerning the crime of rape upon which this appeal turns.

In his contribution under the heading "common law" in The New Oxford Companion to Law, Professor A W B Simpson distinguishes five senses in which that term is used. The primary sense is that body of non‑statutory law which was common throughout the realm and so applicable to all, rather than local or personal in its application. An example of such local or personal laws is the customary mining laws which had applied in various localities in England. The second sense of the term is institutional, to identify the body of law administered in England by the three royal courts of justice, the King's Bench, Common Pleas and Exchequer, until the third quarter of the 19th century. The third sense is a corollary of the second, the expression "the common law" differentiating the law administered by those courts from the principles of equity administered in the Court of Chancery (and, one should add, from the law applied in the ecclesiastical courts until 1857 and the law applied in courts of admiralty).

In that regard, Sir George Jessel MR emphasised in In re Hallett's Estate that, while the rules of the common law were "supposed to have been established from time immemorial", those of equity had been invented, altered, improved, and refined by the Chancellors from time to time, and he instanced "the separate use of [ie trust for] a married woman". With the development since the second half of the 19th century of appellate structures governing all species of primary decisions, judicial reasoning has tended not to invoke time immemorial and rather to follow the course which had been taken by the Chancellors in expounding legal principle.

The fourth and fifth senses of "common law" identified by Professor Simpson are as follows:

"The term 'common law' came, in a fourth sense, to have the connotation of law based on cases, or law evolved through adjudication in particular cases, as opposed to law derived from the analysis and exposition of authoritative texts. Indeed sometimes 'common law' is more or less synonymous with the expression 'case law'. Since the common law was developed by the judges, interacting with barristers engaged in litigation, the expression 'common law' came, in a related fifth sense, to mean law made by judges." This draws attention to a difficulty in the appellant's reliance in this case upon a principle of the common law based upon a statement in a text published in 1736, many years after the death of the author, without citation of prior authority and lacking subsequent exposition in cases where it has been repeated.

In that regard, observations by six members of the Court in the Native Title Act Case are significant. Their Honours noted that the term "common law" might be understood not only as a body of law created and defined by the courts in the past, but also as a body of law the content of which, having been declared by the courts at a particular time, might be developed thereafter and be declared to be different.

Writing at the time of the establishment of this Court, and when he was Professor of Law at the University of Adelaide, Sir John Salmond said:

"The statement that a precedent gains in authority with age must be read subject to an important qualification. Up to a certain point a human being grows in strength as he grows in age; but this is true only within narrow limits. So with the authority of judicial decisions. A moderate lapse of time will give added vigour to a precedent, but after a still longer time the opposite effect may be produced, not indeed directly, but indirectly through the accidental conflict of the ancient and perhaps partially forgotten principle with later decisions. Without having been expressly overruled or intentionally departed from, it may become in course of time no longer really consistent with the course of judicial decision. In this way the tooth of time will eat away an ancient precedent, and gradually deprive it of all authority and validity. The law becomes animated by a different spirit and assumes a different course, and the older decisions become obsolete and inoperative." The term "the common law of Australia"

Finally, in his treatment of "common law", Professor Simpson refers to the expansion of British imperial power and the creation of "a common law world". The common law was received in the Province of South Australia with effect 19 February 1836, but despite the differing dates of the reception of the common law in the Australian colonies, the common law was not disintegrated into six separate bodies of law; further, what was received included the method of the common law, which in Australia involved judicial determination of particular parts of the English common law which were inapplicable to local conditions.

The "common law" which was received did not include the jurisdiction with respect to matrimonial causes (including suits for declarations of nullity of marriage, judicial separation (a mensa et thoro) and restitution of conjugal rights) which in England was exercised by the ecclesiastical courts. This exclusion appears to have been a deliberate decision by the Imperial authorities. Further, unlike the situation in England, in the Australian colonies there was to be no established religion. The Anglican church was expressly enjoined from exercising any authority or jurisdiction in matrimonial causes.

The result was that the jurisdiction with respect to matrimonial causes, as well as divorce, which has been exercised by the colonial and State courts always has been derived from local statute law, not received "common law".

Further, in Skelton v Collins, Windeyer J said of the reception in the Australian colonies of the doctrines and principles of the common law:

"To suppose that this was a body of rules waiting always to be declared and applied may be for some people satisfying as an abstract theory. But it is simply not true in fact. It overlooks the creative element in the work of courts. It would mean for example, that the principle of Donoghue v Stevenson, decided in the House of Lords in 1932 by a majority of three to two, became law in Sydney Cove on 26th January 1788 or was in 1828 made part of the law of New South Wales by 9 Geo IV c 83, s 25. In a system based, as ours is, on case law and precedent there is both an inductive and a deductive element in judicial reasoning, especially in a court of final appeal for a particular realm or territory." Inductive and deductive reasoning

This creative element of both inductive and deductive reasoning in the work of the courts in Australia includes the taking of such steps as those identified by Sir Owen Dixon in his address "Concerning Judicial Method". In his words, these are: (i) extending "the application of accepted principles to new cases"; (ii) reasoning "from the more fundamental of settled legal principles to new conclusions"; and (iii) deciding "that a category is not closed against unforseen instances which in reason might be subsumed thereunder".

To these steps may be added one which is determinative of the present appeal. It is that where the reason or "foundation" of a rule of the common law depends upon another rule which, by reason of statutory intervention or a shift in the case law, is no longer maintained, the first rule has become no more than a legal fiction and is not to be maintained.

An example is provided by a division of opinion in Brown v Holloway and Edwards v Porter respectively between this Court and the House of Lords, as to the consequences of the Married Women's Property Act 1882 (UK) ("the 1882 UK Act") and its Queensland counterpart. Of those cases, it was said in Thompson v Australian Capital Television Pty Ltd:

"The issue [in Edwards v Porter] concerned the effect of the provision in [the 1882 UK Act] that married women were to be capable of suing or being sued as if each were a feme sole, the immediate issue being whether a husband remained liable at common law with his wife for a tort committed by her during joint coverture. In this Court it had previously been decided by Griffith CJ, O'Connor and Isaacs JJ that the liability of the husband was gone. At common law the wife had been liable for her own torts but there was no way in which that liability could be enforced save by an action against her in which her spouse was joined as a party. The joinder of the husband was necessary only because the liability of the wife could not be made effective without his joinder as a party. The legislation removed that procedural disability and therefore the reason which had rendered the husband a necessary party. In Edwards v Porter, without consideration of the reasoning of this Court in Brown v Holloway, their Lordships divided 3:2 in favour of a decision that, notwithstanding the legislation, the husband remained liable to suit with his wife for her torts. One of the minority, Viscount Cave said: 'The whole reason and justification for joining a husband in an action against his wife for her post‑nuptial tort has therefore disappeared; and it would seem to follow, upon the principle "cessante ratione cessat lex," that he is no longer a necessary or proper party to such an action.'" It is with this reasoning in mind that there is to be understood the earlier statement by Dawson J in R v L that:

"whatever may have been the position in the past, the institution of marriage in its present form provides no foundation for a presumption which has the effect of denying that consent to intercourse in marriage can, expressly or impliedly, be withdrawn. There being no longer any foundation for the presumption, it becomes nothing more than a fiction which forms no part of the common law." That statement points the way to the resolution of this appeal.

The point should first be made that, the issue of irrevocable consent by a wife apart, the common law with respect to the crime of rape did not remain static.

Sir Edward Coke in The First Part of the Institutes of the Laws of England early in the 17th century wrote:

"'Rape.' Raptus is, when a man hath carnall knowledge of a woman by force and against her will." In 1957 in their joint reasons in Papadimitropoulos v The Queen, Dixon CJ, McTiernan, Webb, Kitto and Taylor JJ referred to Australian decisions given in 1915, 1919 and 1947 when stating:

"The modern history of the crime of rape shows a tendency to extend the application of the constituent elements of the offence. The 'violenter et felonice rapuit' of the old Latin indictment is now satisfied although there be no use of force: R v Bourke. The 'contra voluntatem suam' requires only a negative absence of consent; (as to the need of the man's being aware of the absence of consent, see R v Lambert). The 'violenter et felonice carnaliter cognovit' is established if there has been some degree of penetration although slight, and no more force has been used than is required to effect it: R v Bourke; R v Burles." Their Honours added:

"To return to the central point; rape is carnal knowledge of a woman without her consent: carnal knowledge is the physical fact of penetration; it is the consent to that which is in question; such a consent demands a perception as to what is about to take place, as to the identity of the man and the character of what he is doing. But once the consent is comprehending and actual the inducing causes cannot destroy its reality and leave the man guilty of rape." The reference in Papadimitropoulos to "[t]he modern history of the crime of rape" may be seen as foreshadowing two points with respect to the development of the common law made by Dixon CJ shortly thereafter. In Commissioner for Railways (NSW) v Scott Dixon CJ spoke of the gradual growth of the legal system by proceeding by reasoning from accepted notions about remedies and rights to the evolution of rules "to govern new or changed situations to which an ever developing social order gives rise"; he went on to observe that "[t]he resources of the law for superseding or avoiding the obsolescent have for the most part proved sufficient". It is upon that sufficiency that the respondent relies in this appeal.

What now follows in these reasons emphasises that some care is required when visiting what Professor Glanville Williams described as "the museum of the English criminal law".

The relevant passage in The History of the Pleas of the Crown appears in Ch 58, headed "Concerning felonies by act of parliament, and first concerning rape". The importance of statutory intervention in this respect may be seen from the passage from Russell's treatise set out earlier in these reasons.

Hale referred to the statement by Bracton that it was a good exception to an appeal (ie formal accusation) of rape that the parties were living in amicable concubinage, adding

"and the reason was, because that unlawful cohabitation carried a presumption in law, that it was not against her will". Hale went on to say:

"But this is no exception at this day[. I]t may be an evidence of an assent, but it is not necessary that it should be so, for the woman may forsake that unlawful course of life." (emphasis added) This is followed by the critical statement:

"But the husband cannot be guilty of a rape committed by himself upon his lawful wife, for by their mutual matrimonial consent and contract the wife hath given up herself in this kind unto her husband, which she cannot retract." (emphasis added) Several points may be made immediately. First, it is apparent from Hale's treatment of Bracton's view in the 13th century of concubinage that he did not regard what had been said in past times as necessarily expressing the common law "at this day" four centuries later.

Secondly, Hale gave, as the reason for the proposition that a husband cannot be guilty of a rape upon his wife, the nature in law of the matrimonial relationship. But, in that regard, it was well settled that marriage was constituted by the present consent of the parties expressed under such circumstances as the law required, but without the requirement for consummation to complete the marriage. Further, as explained later in these reasons, the ecclesiastical courts did not enforce any duty of sexual intercourse between husband and wife.

Thirdly, Hale did not explain the character in law of the proposition respecting rape in marriage, whether it stated an element of the offence, a defence, or an immunity. Nor did Hale refer to any prior cases which might be said to illustrate and support the proposition. From the immediately preceding treatment by Hale of Bracton it is apparent that the proposition is more than a bar to the reception of evidence by the wife or a statement of her absolute testimonial incompetence in this respect. This is further apparent from what immediately follows in Hale's text. This is a treatment of what had been decided at the trial of Lord Audley before the House of Lords in 1631 as follows:

"A the husband of B intends to prostitute her to a rape by C against her will, and C accordingly doth ravish her, A being present, and assisting to this rape: in this case these points were resolved, 1. That this was a rape in C notwithstanding the husband assisted in it, for tho in marriage she hath given up her body to her husband, she is not to be by him prostituted to another. 2. That the husband being present, aiding and assisting, is also guilty as a principal in rape, and therefore, altho the wife cannot have an appeal of rape against her husband, yet he is indictable for it at the king's suit as a principal. 3. That in this case the wife may be a witness against her husband, and accordingly she was admitted, and A and C were both executed." It should be added that in the 19th century, it was held in the Supreme Judicial Court of Massachusetts that there should be no arrest of judgment on the ground that the indictment had not alleged that the complainant was not the wife of any of those charged with raping her. The relevant passage from Hale had been cited, but Bigelow J responded:

"Such an averment has never been deemed essential in indictments for rape, either in this country or in England. The precedents contain no such allegation. See authorities before cited. A husband may be guilty at common law as principal in the second degree of a rape on his wife by assisting another man to commit a rape upon her; Lord Audley's case, 3 Howell's State Trials, 401; and under our statutes he would be liable to be punished in the same manner as the principal felon. Rev Sts c 133, s1. An indictment charging him as principal would therefore be valid. Of course, it would always be competent for a party indicted to show, in defence of a charge of rape alleged to be actually committed by himself, that the woman on whom it was charged to have been committed was his wife. But it is not necessary to negative the fact in the indictment." Thus it will be seen that whatever its character in law, Hale's proposition was not framed in absolute terms, given his treatment of Lord Audley's Case. But what is important for the present appeal is further consideration of the reason given by Hale, which was based in an understanding of the law of matrimonial status in the second half of the 17th century when he wrote.

In the period in which Hale wrote, and until the significant legislative changes in the course of the 19th century, each of the three jurisdictions in England represented by the courts of common law, the courts of equity and the ecclesiastical courts, had distinct roles in matters affecting matrimonial status. The law applied in the common law courts had absorbed much canon law learning and it defined basic concepts such as legitimacy, procedural rights at law between spouses, and the duties and responsibilities of husbands, including their rights and duties in respect of the contracts and torts of their wives. Marriage had important consequences in property law, for establishing and securing inheritance of legal estates in land. In such contexts a court of common law would determine whether there had been a marriage. The common law also provided forms of action such as breach of promise to marry, criminal conversation by adulterers and seduction of daughters.

As already observed by reference to the statement of Sir George Jessel MR in In re Hallett's Estate, equity intervened in a notable fashion by means of the trust to reserve separate property for a wife after her marriage. In his lecture entitled "Of Husband and Wife", Chancellor Kent, after referring to the incompetency at common law of a married woman to deal with her property as a feme sole, went on to contrast the position in equity and described the procedural consequences as follows:

"The wife being enabled in equity to act upon property in the hands of her trustees, she is treated in that court as having interests and obligations distinct from those of her husband. She may institute a suit, by her next friend, against him, and she may obtain an order to defend separately suits against her; and when compelled to sue her husband in equity, the court may order him to make her a reasonable allowance in money to carry on the suit." The provision in the 1882 UK Act and in the corresponding colonial married women's property legislation that a married woman was capable of acquiring, holding and disposing of any real or personal property as her separate property, as if she were a feme sole, "without the intervention of any trustee", represented a triumph in statutory form of the principles of equity. However, it was not until 1862, with the decision of Lord Westbury LC in Hunt v Hunt, that the Court of Chancery enforced a negative covenant in a deed of separation not to sue in the ecclesiastical courts (or after 1857 in the Divorce Court) for restitution of conjugal rights.

Ecclesiastical courts in England had limited powers to order separation of spouses but could not order the dissolution of marriage. This required a statute. Hale wrote in a period in which Parliamentary intervention was beginning. In 1669 a private Act was granted to Lord de Roos, and in 1692 to the Duke of Norfolk; only five such divorces were granted before 1714, but between 1800 and 1850 there were 90. (Divorce by private Act of the legislature was to be attempted in 1853 in New South Wales, but the Instructions issued to colonial governors required that any Bill dealing with divorce be reserved for the Queen's pleasure and the Royal Assent was only given to the Bill after some delay.)

However, it should be noted that in Scotland since the 16th century, provision had been made for judicial grant of divorce on grounds of adultery of either spouse or malicious desertion for at least four years. Given the significant settlement of Scots immigrants in the Australian colonies, this element of their inheritance should not be overlooked in understanding the development of Australian institutions.

In 1891, the English Court of Appeal held that habeas corpus would issue to free a wife confined by her husband in his house in order to enforce restitution of conjugal rights.

"The ecclesiastical courts made decrees for the restitution of conjugal rights but the decree commanded a general resumption of cohabitation and did not purport to compel a spouse to do or abstain from doing particular acts in performance of a connubial obligation. The legal significance of connubial obligations was to be found in the making of decrees based on breaches of those obligations. Breaches were established only by proof of conduct that was a gross infringement of a connubial right or by proof of a continuous failure to perform a connubial obligation in satisfaction of the corresponding connubial right of the other spouse. The courts exercising jurisdiction in matrimonial causes recognized the mutual rights of husband and wife relating to sexual intercourse and, in granting or withholding their decrees, ascertained whether either party had wilfully and persistently refused to accord the right of sexual intercourse to the other party. From the days of the ecclesiastical courts, however, it was accepted that no mandatory order to compel sexual intercourse would be made." In 1933, when describing the nature and incidents of a decree for restitution of conjugal rights under the jurisdiction conferred by Pt III (ss 6‑11) of the Matrimonial Causes Act 1899 (NSW), Dixon J observed in Bartlett v Bartlett that, so long as this remedy was retained, it must be treated as a process imposing an obligation, the performance or non‑performance of which is ascertainable, and he added:

"On the one hand, it is clear that the obligation requires cohabitation, a physical dwelling together. On the other hand, it is clear that it does not require the resumption of sexual intercourse. It cannot, in fact, and in principle ought not to be understood as attempting to, control motives, feelings, emotions, sentiment or states of mind. Its operation must be limited to overt acts and conduct. ... Perhaps, all that can be said is that the decree of restitution requires the spouse against whom it is directed again to dwell with the other spouse in outward acceptance of the relationship, to act as if they were husband and wife maintaining a matrimonial home and to commence no course of conduct intended to cause a separation." Evatt J set out a passage from the reasons of Salmond J in Fielding v Fielding in which, with reference to the jurisdiction conferred by s 7 of the Divorce and Matrimonial Causes Act 1908 (NZ) for the issue of decrees for restitution of conjugal rights, Salmond J had said:

"The Ecclesiastical Courts [in England] never professed or attempted by means of decrees for restitution of conjugal rights, and imprisonment for disobedience to such decrees, to enforce any duty of sexual intercourse between husband and wife. The basis of such a decree was the wrongful refusal of matrimonial cohabitation. The duty enforced was merely the duty of husband and wife to live together under the same roof in the normal relationship of husband and wife, but without reference to the question of intercourse." The divorce legislation

The passage of the Matrimonial Causes Act 1857 (UK) ("the 1857 UK Act") later was described by Dicey as "a triumph of individualistic liberalism and of common justice". But it was the culmination of many years of agitation. Of the delay, Professor Cornish writes:

"It is less easy to explain why, given the long availability of judicial divorce in Scotland and its spread to other Protestant countries, the step did not come earlier. Jeremy Bentham, for instance, had been an advocate of fully consensual divorce, but subject to time delays for reflection and a bar on the re‑marriage of a guilty party." (footnote omitted) The 1857 UK Act terminated the jurisdiction of the ecclesiastical courts in matrimonial matters (s 2) and vested that jurisdiction in the new Court for Divorce and Matrimonial Causes (s 6), but the Court was to act on the principles and rules which had been applied by the ecclesiastical courts (s 22). A decree dissolving marriage might be pronounced on a petition by the husband alleging adultery by the wife, and on a wife's petition, alleging adultery coupled with desertion for at least two years and without reasonable excuse, or alleging adultery with aggravated circumstances including "such Cruelty as without Adultery would have entitled her to a Divorce à Mensâ et Thoro" (ss 27 and 31).

In 1858 the Secretary of State for the Colonies conveyed to all colonial governors and legislatures the wish of the Imperial Government that steps be taken to introduce, "as nearly as the circumstances of the Colony will admit", the provisions of the 1857 UK Act.

The colonies acted accordingly, but at different paces: Matrimonial Causes Act 1858 (SA), Matrimonial Causes Act 1860 (Tas), Matrimonial Causes Act 1861 (Vic), Matrimonial Causes Act 1863 (WA), Matrimonial Causes Act 1865 (Q), Matrimonial Causes Act 1873 (NSW). This legislation did not need to abolish in the colonies the non‑existent jurisdiction of ecclesiastical courts. Rather, it conferred jurisdiction in matrimonial causes on the Supreme Courts. The differential treatment in the 1857 UK Act between the grounds of divorce available to husbands and wives was carried into the initial colonial legislation. But there followed attempts by New South Wales and Victoria to assimilate and expand the grounds for divorce; the Governor's Instructions required these Bills to be reserved for the Royal Assent on advice of the Imperial Government and, initially, in circumstances of considerable controversy in the colonies, the Royal Assent was refused.

Pressure for reform of legislation respecting divorce was, however, maintained, particularly in the more populous colonies of New South Wales and Victoria, and eventually succeeded. In Victoria The Divorce Act 1889 provided extended grounds for divorce. Advocates of the women's movement in New South Wales were able to press for further liberalisation of the laws, despite the opposition of the churches. The Divorce Amendment and Extension Act 1892 (NSW) was expressed in terms similar to those of the Victorian Act. The extended grounds gave colonial women greater access to divorce than their contemporaries in the United Kingdom.

What was the immediate significance of these 19th century legislative measures for the continued vitality of the reasoning upon which Hale in the 17th century had based his proposition respecting "rape in marriage"?

In answering that question it is convenient first to repeat what was said by the Supreme Court of New Jersey in State v Smith as follows:

"We believe that Hale's statements concerning the common law of spousal rape derived from the nature of marriage at a particular time in history. Hale stated the rule in terms of an implied matrimonial consent to intercourse which the wife could not retract. This reasoning may have been persuasive during Hale's time, when marriages were effectively permanent, ending only by death or an act of Parliament. Since the matrimonial vow itself was not retractable, Hale may have believed that neither was the implied consent to conjugal rights. Consequently, he stated the rule in absolute terms, as if it were applicable without exception to all marriage relationships. In the years since Hale's formulation of the rule, attitudes towards the permanency of marriage have changed and divorce has become far easier to obtain. The rule, formulated under vastly different conditions, need not prevail when those conditions have changed." To that may be added the statement in that case:

"If a wife can exercise a legal right to separate from her husband and eventually terminate the marriage 'contract', may she not also revoke a 'term' of that contract, namely, consent to intercourse?" In similar vein is the statement made from the New South Wales Supreme Court bench by Sir William Windeyer in 1886, in which he regretted that while the State regarded marriage as a civil contract and in this case the contract had been destroyed by the husband "having done his best to degrade you", by reason of the then limited grounds of divorce then available to her in New South Wales, she had no redress.

Insofar as Hale's proposition respecting the nature of the matrimonial contract was derived from an understanding of the principles applied by the ecclesiastical courts, the following may be said. First, as Lord Brougham observed in R v Millis:

"[Marriage] was always deemed to be a contract executed without any part performance; so that the maxim was undisputed, and it was peremptory, 'Consensus, non concubitus, facit nuptias vel matrimonium.'" Secondly, with respect to the exercise of their jurisdiction in suits for restitution of conjugal rights, the ecclesiastical courts did not accept that the exercise of the mutual rights of spouses was to be an occasion of abuse and degradation. The following further remarks of Brennan J in R v L are in point:

"To acknowledge a connubial obligation not to refuse sexual intercourse wilfully and persistently is to acknowledge that the giving of consent to acts of sexual intercourse is necessary to perform the obligation. It would have been inconsistent with such an obligation to hold that, on marriage, a wife's general consent to acts of sexual intercourse has been given once and for all. If no further consent was required on the part of a wife, how could there be a wilful and persistent refusal of sexual intercourse by her? The ecclesiastical courts never embraced the notion of a general consent to sexual intercourse given once and for all on marriage by either spouse." Thirdly, and in any event, in the Australian colonies jurisdiction with respect to matrimonial causes was not part of the general inheritance of the Supreme Courts. They received such jurisdiction only by local statute in the second half of the 19th century. That legislation, as interpreted in the period before the enactment of the CLC Act in 1935, did not require, for compliance with a decree for restitution of conjugal rights, more than matrimonial cohabitation; in particular the duty of matrimonial intercourse was one of imperfect legal obligation because it could not be compelled by curial decree.

Finally, although Hale did not expressly rely upon it, his proposition respecting irrevocable consent could not have retained support from any common law concept that the wife had no legal personality distinct from that of her husband. This was never wholly accepted by the Court of Chancery, given the development there of the trust. The references earlier in these reasons to the significance of the married women's property legislation indicate that, by statute, the attitudes of the equity jurisdiction were given effect in the latter part of the 19th century to a significant degree throughout the legal system in England and the Australian colonies.

To that may be added the significance of the conferral by the Commonwealth Franchise Act 1902 (Cth) of the universal adult franchise. It has been said that the gaining of suffrage for women in South Australia in 1894 was critical to the national suffrage movement. At the turn of the 20th century, suffragists in England were looking to what had been achieved in Australia. An English suffragist, Dame Millicent Garrett Fawcett, writing in 1911 when women in England had not yet been granted suffrage, observed that:

"In the Commonwealth of Australia almost the first Act of the first Parliament was the enfranchisement of women. The national feeling of Australia had been stimulated and the sense of national responsibility deepened by the events which led to the Federation of the Independent States of the Australian Continent." By 1930 Isaacs J was able to say that:

"women are admitted to the capacity of commercial and professional life in most of its branches, that they are received on equal terms with men as voters and legislators, that they act judicially, can hold property, may sue and be sued alone". By the time of the enactment in 1935 of the CLC Act, if not earlier (a matter which it is unnecessary to decide here), in Australia local statute law had removed any basis for continued acceptance of Hale's proposition as part of the English common law received in the Australian colonies. Thus, at all times relevant to this appeal, and contrary to Hale's proposition, at common law a husband could be guilty of a rape committed by him upon his lawful wife. Lawful marriage to a complainant provided neither a defence to, nor an immunity from, a prosecution for rape.

To reach that conclusion it is unnecessary to rely in general terms upon judicial perceptions today of changes in social circumstances and attitudes which had occurred in this country by 1935, even if it were an appropriate exercise of legal technique to do so. The conclusion follows from the changes made by the statute law, as then interpreted by the courts, including this Court, before the enactment of the CLC Act.

HEYDON J. The events giving rise to this appeal allegedly took place in 1963. At that time it was universally thought in Australia that a husband could not be convicted of having sexual intercourse with his wife without her consent save where a court order operated or where there were other exceptional circumstances. This immunity from conviction was thought to exist because Sir Matthew Hale, who died in 1676, had asserted its existence in The History of the Pleas of the Crown, published in 1736. The reason he assigned was that on marriage wives irrevocably consented to sexual intercourse with their husbands. Below the immunity will be called "the immunity" or "Hale's proposition".

By what warrant did the State of South Australia seek in 2010 to prosecute the appellant for allegedly having sexual intercourse with his wife without her consent more than 47 years earlier? A sufficient answer to that question would be: "It had none, for the reasons that Bell J powerfully states." However, in deference to the arguments put by South Australia, a fuller answer should be given.

One matter must be put aside, though the appellant may wish to rely on it at a later stage in these proceedings. This appeal is not directly concerned with any oppressiveness that results from the delay in prosecution. But that tardiness does support the appellant's submission that in 1963 there was no crime of rape for which he could be charged. One primary explanation which South Australia gave to the District Court for its delay was that the immunity created considerable doubt as to whether the appellant was liable for rape. Yet prosecutors have to demonstrate with clarity that the crimes they charge exist. South Australia tells the District Court that the appellant's liability was thought doubtful. It tells this Court that it is certain. South Australia's stance in the District Court is inconsistent with its dogmatic and absolute submissions in this Court. The first of its submissions in this Court was that the immunity never existed. The second, alternative, submission was that even if the immunity had existed at one time, it had ceased to exist at some indeterminate time before 1963.

South Australia put only those two submissions. It did not put a third submission - that even if the immunity existed and even if it had not ceased to exist up to now, it should be abolished now. That was a course which the English courts took in 1991. It is a course which would raise issues different in some respects from those discussed below.

South Australia's first submission: the detailed contentions

South Australia's first submission was that it had never been the law, in England or in Australia, that a husband was immune from prosecution for having sexual intercourse with his wife without her consent. The Commonwealth supported that submission. It was based on a number of contentions.

The first group of contentions centred on the following points. Hale's work was published 60 years after he died. The relevant part had not been revised before his death. Hale had not supported his statement with any reference to authority. Standing alone his proposition would not constitute the common law. At best it reflected "his view of a custom in 17th century England." As Blackstone asserted, "judicial decisions are the principal and most authoritative evidence, that can be given, of the existence of such a custom as shall form a part of the common law."

A second group of contentions concerned ecclesiastical law. In the ecclesiastical courts there was no support for Hale's proposition. In ecclesiastical law each spouse had a right to sexual intercourse, but it was only to be exercised reasonably and by consent. This undermined the foundation of Hale's proposition. It revealed him to be mistaken in thinking that the wife's consent was irrevocable. It caused his proposition to be affected by "frailty".

South Australia then turned to the history of Hale's proposition after he had enunciated it. It relied on Lord Lowry's very extreme statement that "Hale's doctrine had not been given the stamp of legislative, judicial, governmental and academic recognition."

So far as "academic recognition" was concerned, South Australia submitted that the only statement of support for the immunity in absolute terms was that of Hale, and that there was no support for it in Blackstone.

So far as "judicial … recognition" was concerned, South Australia submitted that Hale's proposition "was never authoritatively declared as part of the common law in Australia." It also submitted that no case had Hale's proposition as its ratio decidendi. There were only dicta and assumptions that the proposition existed. In some of the cases stating the dicta or resting on the assumptions Hale's proposition was cut down. There were also dicta to the contrary. Hale's proposition was further said to be inconsistent with some authorities.

South Australia did not advance detailed submissions about any lack of "legislative" and "governmental" recognition. Perhaps it is hard to say much in support of negative propositions. However, there is a lot to be said against those two.

Finally, South Australia submitted that the immunity was completely outdated and offensive to human dignity.

It is convenient to deal with South Australia's first submission under the following headings.

It is immaterial that Hale's work was published 60 years after his death, that the reference to the immunity appears in a part of it which Hale had not revised, and that he stated no elaborate reasons justifying the immunity. Hale's work is capable of being an accurate account of the law of his day despite these things. There is no reason to suppose that, had he revised the relevant part of his work, he would have considered it desirable to change it.

South Australia is not alone in complaining about Hale's failure to cite authorities. But it is anachronistic to do so. The modern approach to precedent was only struggling to be born in Hale's day. Hale himself said:

"the decision of courts of justice, though by virtue of the laws of this realm they do bind, as a law between the parties thereto, as to the particular case in question, till reversed by error or attaint; yet they do not make a law, properly so called; - for that only the king and parliament can do; yet they have a great weight and authority in expounding, declaring, and publishing what the law of this kingdom is; especially when such decisions hold a consonancy and congruity with resolutions and decisions of former times. And though such decisions are less than a law, yet they are a greater evidence thereof than the opinion of any private persons, AS SUCH, whatsoever". Hale's work often does not contain the dense citation of authorities characteristic of modern books. Indeed, many parts of it refer to only a few authorities. That is so of the passages in which he discusses the crime of rape. Hale did not cite direct authority for the immunity, or for his justification of the immunity. Whether there were in fact "authorities" of any kind to be cited on the present point is a matter which a 21st century court cannot easily deal with. It would need the assistance of close research into the question by modern legal historians with high expertise. Subject to that matter, Hale did point out that there was authority for other propositions that he asserted. Those propositions were not inconsistent with the immunity. To some extent they supported it. In view of Hale's high reputation for research into the criminal litigation of his day, it seems likely that in practice husbands were not prosecuted for raping their wives, so that there were no authorities to cite. Even nowadays, a proposition can be correct though no precedent supports it. Ethical and tactical considerations prevent counsel from arguing what they perceive to be the unarguable. There are some propositions which seem too clear to the profession to be contradicted by argument. Propositions of that kind are widely accepted as good law.

Subject to the research difficulties referred to in the previous paragraph, a lack of support from earlier authors such as Coke is, as Bell J explains, not significant and does not reveal Hale to be wrong.

Windeyer J once said: "an accepted rule of law is not to be overthrown by showing that history would not support it". None of the defects which supposedly existed in Hale's statement of the immunity prevented it from being an accepted rule of law.

South Australia's appeal to ecclesiastical law encounters two difficulties.

The first is that while the civil law of marriage was a matter for the ecclesiastical courts, the criminal law was a matter for the common law courts. Thinking in the ecclesiastical courts does not necessarily vitiate an account of the criminal law as administered in the common law courts.

The second difficulty is that ecclesiastical law in the 17th century is another field not to be entered without expert assistance. In R v L Brennan J wrote at some length about ecclesiastical law. But the sources to which he referred were largely modern. None were contemporary with or earlier than Hale.

Lord Lane CJ in R v R and Mason CJ, Deane and Toohey JJ in R v L pointed out that in Popkin v Popkin Sir William Scott (later Lord Stowell) stated: "The husband has a right to the person of his wife, but not if her health is endangered." Mason CJ, Deane and Toohey JJ commented that this showed that "even in the ecclesiastical courts, the obligation to consent to intercourse was not asserted in unqualified terms." If so, it also shows that Hale's proposition was not completely wrong. On the other hand, Brennan J did not think that Sir William Scott's statement showed that a husband had "a right to the person of his wife" without consent.

The submissions of the parties in this appeal did not take the matter further than Brennan J's researches took it. The parties did not cite any expert material throwing light on ecclesiastical law in or before Hale's time. For that reason, it is imprudent to examine it.

South Australia submits that there is no statement of support for Hale's proposition except Hale himself, and that Hale's proposition has received no "academic recognition". That submission is extremely ambitious. It is also utterly incorrect.

It is true that Hale's proposition is neither confirmed nor denied by Blackstone or Hawkins. Blackstone was writing at a considerable level of generality about much wider issues than those Hale wrote about. Whether or not it is right to describe Hawkins as "a somewhat second-rate institutional writer", it was not open to him to take up Hale's proposition in the first edition of his treatise. It appeared in 1716. Hale's work was not published until 1736.

South Australia echoes the Crown's complaint to the House of Lords in R v R that the first writer to refer to the immunity after Hale was East in 1803. That is, however, less than 70 years after History of the Pleas of the Crown was published in 1736. In truth, Hale has enjoyed a great reputation. Lord Denning called him "the great Chief Justice Sir Matthew Hale". Hale's proposition garnered massive support from professional writers after 1803, and, as academic lawyers emerged, from them too. Leading modern writers like Glanville Williams, Smith and Hogan and Cross and Jones acknowledged the correctness of Hale's proposition. Like others who have attacked courts that relied on Hale's proposition, South Australia failed to grapple with this uncomfortable point.

One source of law is "informed professional opinion". Where there is little authority on a question of law, the opinions of specialist writers, particularly their concurrent opinions, are very important in revealing, and indeed in establishing, the law. That is particularly true of books written by practitioners. But it has force in relation to well-respected academic writers as well. As Lord Reid said: "Communis error facit jus may seem a paradox but it is a fact." Owen J put the matter with trenchant simplicity in relation to the first edition of Archbold in 1822. It said: "A husband … cannot be guilty of a rape upon his wife." Owen J said:

"It seems to me that the consequences of that statement is this: if he was right, then practitioners would follow what he said. Equally, however, if he was wrong, practitioners would follow what he said." Hale's proposition in the courts

South Australia relied on Lord Lowry's statement that Hale's proposition had not been "given the stamp of … judicial … recognition." South Australia greatly exaggerated the extent to which the authorities cast doubt on the immunity before 1991, when the House of Lords decided it should be abolished, and four members of this Court said that it had ceased to represent the law.

An illustration is provided by South Australia's submission in relation to R v Clarence:

"The exemption was first the subject of judicial comment in R v Clarence. In R v Clarence, seven of the thirteen judges declined to comment on the issue; of the six judges who did, two of them reiterated and confirmed the marital rape proposition, three of them questioned or qualified it, and another briefly adverted to it without engaging in it. The comments of the judges in R v Clarence were obiter dicta, however, taken as a whole they indicate that even as at 1888 there existed no settled view." It is true that the few things said about Hale's proposition in R v Clarence were dicta. The submission is otherwise very misleading. It suggests that only A L Smith J and Pollock B favoured Hale's proposition. In fact the position is as follows. Stephen J supported Hale's proposition. The following seven judges concurred: A L Smith J, Mathew J, Grantham J, Manisty J, Huddleston B, Pollock B and Lord Coleridge CJ. Wills J, Hawkins J and Field J each stated or left open the possibility that in some circumstances a husband could be convicted of raping his wife. But at least the latter two judges plainly thought that Hale's proposition was correct in some circumstances. Day J concurred with Hawkins J. Charles J concurred with Field J.

South Australia correctly submitted that English trial judges assumed that Hale's proposition was correct, but qualified its operation in special circumstances. Examples of special circumstances included where there was a court non-cohabitation order, or a decree nisi of divorce had effectively terminated the marriage, or the husband had given an undertaking to the court not to molest the wife, or there was an injunction restraining the husband from molesting or having sexual intercourse with the wife, or there was an injunction and a deed of separation (even though the injunction had expired). The outer limit of these exceptions was unilateral withdrawal from cohabitation coupled with a clear indication that the wife's consent to sexual intercourse was revoked.

On one view, each of the courts that reached these decisions was attempting to achieve justice by tailoring the absolute nature of Hale's proposition to the circumstances before it. Even if the wife could be said to have given consent by marriage, in those cases it had been withdrawn as a matter of practical reality either because the wife had successfully invoked court process or because the spouses had reached a formal agreement negating consent. Another view is that these cases travel down a road "potholed with ever greater illogicalities"; produce "a gaggle of technical and anomalous distinctions" and "absurdity"; and lack any "relationship to the real world." But even if this latter view is correct, it is adverse to South Australia's position. The cases show the enduring toughness of Hale's proposition in legal thought. To destroy Hale's proposition might eliminate formal anomalies and technicalities. But it was a course which many judges found unattractive. Instead they turned their minds to devising narrow exceptions.

In two of the cases just referred to, the correctness of Hale's proposition, in the absence of special circumstances, was specifically acknowledged by quotation. In another, Hale's proposition was thought to be correct at common law though not necessarily satisfactory. And in R v Miller, where the prosecution failed to prove any special circumstance and could point to no more than the wife having petitioned for divorce, Hale's proposition was applied. R v Miller renders false South Australia's submission that "no binding precedent can be found where [Hale's] principle represented the ratio decidendi." Contrary to that submission, R v Miller was an "authoritative declaration of the common law on the matter." R v Miller also demonstrates that McGarvie J was wrong to say in R v McMinn: "There does not seem to have been any recent case in which it was considered whether [Hale's proposition] remains part of the common law." R v Miller was a binding precedent in England until 1991. A second case which applied Hale's proposition is R v J - (rape: marital exemption).

There are numerous cases, including Australian cases, in which courts have assumed Hale's proposition to be correct at common law. One is R v Kowalski, in which the English Court of Appeal described Hale's proposition as "clear, well-settled and ancient law". Another is R v Bellchambers, in which Neasey and Everett JJ said that Hale's proposition "still expresses the common law", even though they criticised it as "archaic, unjust and discriminatory". A third is Brennan J's statement in R v L:

"Irrespective of the validity of Hale's reason for declaring that a husband could not be guilty as a principal in the first degree of rape of his wife, it appears that a substantive rule of the common law was established by his declaration." That statement followed more than 10 pages denouncing Hale's reasoning. Thus these last two cases, too, reveal the enduring toughness of Hale's proposition in legal thought.

South Australia proffered three authorities that, in its submission, reveal that contrary to Hale's proposition, there was no irrebuttable presumption that on marriage the wife irrevocably consented to sexual intercourse with her husband. In R v Lister, it was held that while it was lawful for a husband to restrain his wife's liberty where she was making "an undue use" of it, either by "squandering away the husband's estate, or going into lewd company", he could not do so where he had entered into a deed of separation with his wife. In R v Jackson, the English Court of Appeal held that where a wife refused to live with her husband, he was not entitled to deprive her of liberty by kidnapping her and confining her to his house, even though he had obtained a decree for restitution of conjugal rights. In R v Reid, that Court held that matrimonial status conferred no immunity on a husband who kidnapped his wife.

These cases do not support South Australia's submission. They do not proceed on the basis that on marriage there was a presumption that the wife consented to having her liberty restrained and that she could rebut that presumption by withdrawing her consent. Indeed, in R v Reid the Court said of the doctrine in R v Miller that it was "impossible to stretch that doctrine to the extent of saying that on marriage a wife impliedly consents to being taken away by her husband using force or threats of force from the place where she is living." Accordingly, these cases are not inconsistent with Hale's proposition.

In 1991, the English Court of Appeal and the House of Lords overturned Hale's proposition. But it is notable that they did not accept the Crown's submission that "Hale's statement was never the law". The rulings of those Courts had retrospective consequences. But they did not hold that Hale's proposition had never been the law. They did not hold that the judgments which had decided, said or assumed that it was correct were wrong at the time they were handed down. Lord Lane CJ said in the Court of Appeal that Hale's proposition had been "accepted as an enduring principle of the common law." And the House of Lords altered the law because social conditions had changed quite recently. Hale's proposition was seen as reflecting the society of his day, and its rejection was seen as reflecting the different form which modern society had recently taken. On that reasoning, Hale's proposition was good law in South Australia in 1963 - a matter relevant to rejection of South Australia's second submission.

South Australia also relied on Lord Lowry's statement that Hale's proposition "had not been given the stamp of … governmental … recognition." That submission too must be rejected.

Hale's proposition has received indirect governmental recognition - by the Executive - in two ways.

One form of governmental recognition took place in a report of the Criminal Law and Penal Methods Reform Committee of South Australia published in 1976. That report acknowledged that Hale's proposition represented the common law. It recommended that the immunity be abolished where the event charged took place while the parties were living separately. There are several other reports before and after the South Australian report also resting on the view that Hale's proposition represented the common law.

Another form of governmental recognition has taken place. Not only in South Australia but in many other places, the authorities did not prosecute charges against husbands accused of raping their wives. This Court was not told of any prosecutions having been brought in England between Hale's time and 1949. In the second half of the 20th century, as exceptions developed to Hale's proposition, there were attempts to prosecute husbands, not for non-consensual sexual intercourse with their wives, but for crimes committed in connection with that conduct, such as assault or false imprisonment. So far as these crimes were distinct from sexual intercourse without consent, the prosecutions rested on sound legal thinking, though they were at peril of failing if there were difficulties in establishing that the husband's conduct had gone beyond sexual intercourse without consent. What the prosecution assumes about the law is not decisive as to what the law is. But it is some guide to the thinking of experienced criminal lawyers. That thinking can be highly persuasive as to what the law is.

Finally, South Australia relied on Lord Lowry's statement that Hale's proposition "had not been given the stamp of legislative … recognition." That submission must also be rejected.

Some forms of "legislative recognition" are of limited materiality. A statute expressly adopting Hale's proposition in South Australia would have superseded the common law. A statute expressly adopting it outside South Australia would have had slight relevance only to what the common law was in South Australia. Statutes rejecting it would have had little relevance to the position at common law unless they reflected a consistent legislative view of what the public interest demanded. But there is South Australian legislation recognising Hale's proposition in the sense that it did not interfere with it when there was an occasion to do so.

The South Australian Parliament did not adopt the recommendation of the Criminal Law and Penal Methods Reform Committee of South Australia to abolish the immunity when husband and wife were living separately. Instead two provisions relevant to the common law were enacted. First, s 73(3) of the Criminal Law Consolidation Act 1935 (SA) was introduced, removing any presumption that consent to sexual intercourse flowed from marriage. Secondly, s 73(5) was introduced: it prevented a spouse from being convicted not only of rape but also of indecent assault, attempted rape or indecent assault, and assault with intent to commit rape or indecent assault, unless the alleged offence was accompanied by various forms of aggravated conduct. In substance it overlapped with the common law position precluding convictions of husbands for the conduct of sexual intercourse with their wives without consent. Thus the South Australian Parliament assumed that the immunity existed at common law, and left it in existence. It appears to have gone further in creating an immunity without any common law counterpart for non-aggravated forms of the crimes other than rape to which s 73(5) referred. That state of affairs continued until 1992.

"Legislative recognition" in places other than South Australia assuming that the immunity existed at common law is relevant to the content of South Australian law - particularly Australian legislation, since there is a single common law in Australia. There are three points to be made. First, in the Code States (Queensland, Tasmania and Western Australia), the definition of rape excluded sexual intercourse by a husband with his wife. The relevant statutes assumed the correctness of Hale's proposition. Secondly, the numerous changes in State and Territory legislation in the 1970s and 1980s indicated an assumption by each legislature (and by each Executive, which had a large measure of control over what draft legislation was introduced) that Hale's proposition was sound at common law. If not, it would not have been necessary to abolish or qualify it. Thirdly, Canadian and New Zealand legislation assumed the correctness of Hale's proposition.

There are no doubt many criticisms to be made of Hale's proposition if it is to be applied in the circumstances of 2012. But these criticisms do not show that his proposition was necessarily anachronistic or offensive in 17th century circumstances. That would depend on historical analysis which the parties' submissions did not perform. The criticisms therefore do not demonstrate that Hale's proposition was wrong from the outset. They are, however, appropriate arguments to consider when deciding whether Hale's proposition ought to be abandoned. It is a question which legislatures answered affirmatively from the 1980s on. It is not a question which South Australia places before this Court. Whether they are appropriate arguments to consider when deciding whether Hale's proposition dropped out of the law at some point before 1963 is a question which relates to South Australia's second submission.

It was inherent in South Australia's first submission that all the writers, all the judges, all the government officials, all the law reformers, all the public servants advising Ministers, all the prosecution authorities and all the legislatures who wrote or acted on the assumption that Hale's proposition was the law were wrong. South Australia explicitly adopted a statement to this effect in the Full Court of the Supreme Court of South Australia. I flatly disagree.

On the assumption that Hale's proposition was correct for some time after he stated it, South Australia put various contentions denying its applicability in 1963.

South Australia contended that before 1963 the law had changed so as to nullify Hale's proposition, even though no case had stated this before 1991. This is an unusual invocation of the judicial process. South Australia's contention is different from a contention that this Court should now declare Hale's proposition to be wrong, and do so with effect retrospective to 1963. This latter course presents in an overt form the considerable difficulties which cluster around the making of retrospective changes to the criminal law. South Australia did not choose to tackle these difficulties head on. It did not suggest that this Court should now change the law. Rather, it submitted that the dicta of four Justices in R v L recognised that Hale's proposition had ceased to be the law at some time before 1991. South Australia submitted that this time was a time before 1963. In fact, nothing in R v L suggests that the demise of Hale's proposition took place at any specific time before 1991, such as 1963. South Australia's thesis that R v L bound the Full Court of the Supreme Court of South Australia to reach this conclusion must be rejected.

South Australia relied on the following arguments as indications that Hale's proposition had ceased to be the law before 1963.

South Australia submitted that in R v Jackson the denial of the husband's right to use physical coercion on his wife suggested that the immunity had disappeared from the law. But it does not follow that a husband lost the immunity where he had not employed physical coercion against his wife.

South Australia also submitted that Hale's proposition was based on the doctrine of coverture - that the legal status of a wife is assimilated with that of her husband. And it submitted that by the turn of the 20th century the law had come to acknowledge the rights of married women as independent of their husbands' rights. In effect, it submitted that so many inroads had been made on the doctrine of coverture that it could no longer support Hale's proposition. This submission has the drawback that Hale did not base his proposition on the doctrine of coverture.

However, the developments on which South Australia relied could be used, and to a degree were used, to support an argument that by the mid-20th century the rights and privileges of married women in Australia were inconsistent with any theory that on marriage wives gave their husbands irrevocable consent to sexual intercourse. South Australia advanced detailed submissions on the capacity of wives, gained by statute, to own property, to sue and be sued, to vote, to have custody of children, and to compel payment of and be compelled to pay child maintenance. It also relied on the termination by statute of discrimination between husbands and wives in relation to the grounds of divorce.

In that way South Australia advances an argument for permitting the appellant to be prosecuted now for conduct which allegedly occurred in 1963. Whether it should be accepted depends on four matters. One is whether in fact the changes in the rights and privileges of wives by 1963 were, to use the words of South Australia's written submissions, "entirely inconsistent with the principle that a wife gave irrevocable consent to sexual intercourse with her husband upon marriage." A second concerns the need for certainty in the criminal law. A third concerns whether South Australia's argument could, in a practical sense, work a retrospective change in criminal law. A fourth is whether the task being undertaken is appropriate for the courts as distinct from the legislature. It is convenient to deal with these points in order.

Is there inconsistency between the modern rights and privileges of wives and the immunity of husbands for rape?

Bell J gives convincing reasons for answering this question in the negative. Some ideas which tend to render Hale's proposition anachronistic can be discerned in 19th and early 20th century legislation. But the crucial triggers that would push Hale's proposition into disfavour arose in the 1970s. Before that decade there had been some questioning by lawyers of the stated justification for Hale's proposition. In that decade questioning began to grow about whether the proposition should be abolished by the legislature. The questioning grew as public concern about the law of rape in general and marital rape in particular began to rise. Law reform agencies began to examine numerous problems in detail. Legislative changes of different kinds were introduced. One trigger was the controversial decision in R v Morgan that a mistaken but honest belief that the victim had consented to intercourse was a defence to a rape charge, whether or not that belief was reasonable. R v Morgan was decided on 30 April 1975, seven months before the then Attorney-General for the State of South Australia requested the Criminal Law and Penal Methods Reform Committee of South Australia to report on the law relating to rape and other sexual offences. It was the very first topic the Committee dealt with. Another trigger concerned whether warnings about the desirability of finding corroboration for the evidence of those complaining that they had been raped rested on unsatisfactory stereotyping. Another trigger was discontent about the rules relating to the cross‑examination of complainants about their sexual histories. There were many other issues about which debates began to widen and intensify in those years. The immunity was only one of them.

Further, the reasons underlying the legislation which has altered the status of wives over the last 150 years are not necessarily inconsistent with the immunity. To describe Hale's proposition as creating a presumption which no longer has any foundation, and as a fiction not forming part of the common law, overlooks the fact that a common law rule can rest on a fiction, particularly when the rule in question develops a new and non-fictitious basis. As Lord Sumner observed: "an established rule does not become questionable merely because different conjectural justifications of it have been offered, or because none is forthcoming that is not fanciful." A fortiori, an established rule does not become questionable merely because a justification which appealed to the minds of lawyers more than 300 years ago has ceased to have appeal now. In Australia, the controversy has been resolved. The resolution lies in abolition of the immunity. Abolition came by degrees. It also came from legislatures. In England, on the other hand, the first decision to abolish the immunity was made by a judge - Simon Brown J, in 1990. Rougier J, sitting alone, at once refused to follow it. Then the immunity was abolished by the Court of Appeal and the House of Lords in R v R in 1991. The House of Lords decision has been supported on the ground that Hale's proposition was one which "nobody defended on the merits." That, however, is not true.

It is true to say, though, that R v R was based, as Lord Lowry later said, "on a very widely accepted modern view of marital rape". But the fact that an idea is very widely accepted does not mean that an inference from it automatically becomes a rule of law. The fact that a rule of law is disliked does not mean that it has ceased to be the law. The fact that a rule of law favourable to the accused is disliked does not mean that the courts rather than the legislature should abolish it. Indeed, after the English courts abolished the immunity, Parliament did as well, once the matter had been considered by the Law Commission. And the fact that very many people have disliked a rule of law favourable to the accused for a long time does not mean that it has ceased to be the law at some time in the past. The Permanent Court of International Justice said, in a somewhat different context, in Consistency of Certain Danzig Legislative Decrees with the Constitution of the Free City, "[s]ound popular feeling is a very elusive standard." And mere popular feeling, however widespread, is a very unsafe standard to apply in relation to claims that common law rules have fallen into desuetude.

Professor Robertson has contrasted R v R with C (A Minor) v Director of Public Prosecutions, in which the House of Lords declined to alter the common law rule that there is a rebuttable presumption that a child aged between 10 and 14 is doli incapax. He correctly noted that the speeches in R v R contained "almost no argument", only "a bald statement". He argued that R v R rested on "the assumption, though it is an untested one, that there is wide consensus in the general public on the question of marital rape." He also argued that Lord Lowry's attempt to reconcile a change in the criminal law in R v R with a decision not to change it in C (A Minor) v Director of Public Prosecutions was "specious". He said:

"The abolition of the rule on rape, though occasioned by a rape where the man and wife were separated, would in fact apply inside an ongoing marriage. It is sociologically extremely unlikely that this view would command anything like as much support amongst the mass public as would a rule that allowed the conviction of thirteen-year-old auto-thieves. The fact that there had been several cases where judges had attempted to convict husbands for rape is on par with the attempt by the Divisional Court to change the doli incapax rule, where there was extensive quotation from judges who had wanted to but were dutiful to precedent. What is true is that liberal elite opinion was uniform in the rape context, and largely missing in the criminal capacity case. Asked how to square the two results, one Law Lord who had been a member of the bench in C but not in R v R threw his hands in the air and admitted he could not imagine how they squared it. Another though, who had heard C, indicated that his decision was prompted by a desire to force the Government's hand and make it legislate." (emphasis in original) The task of assessing public opinion, or even the full range of legal opinion, whether now or in the past, is not an enterprise that is easy for courts to carry out. "The court does not, and cannot, carry out investigations or enquiries with a view to ascertaining whether particular common law rules … command popular assent." The question is not whether the view which the House of Lords stated in R v R about the general public's opinion is correct. The point is that if the general public's opinion is a relevant criterion, it is a criterion for the legislature to consider, not the courts.

Those who seek to foster the rule of law prize certainty. Ordinarily, certainty in the common law is assisted by the doctrine of precedent. Normally, a common law rule is supported by authorities. If an intermediate or ultimate appellate court decides to change the rule, it overrules the authorities and its decision creates a new binding authority. South Australia's submission is not that Hale's proposition be rejected, so that this Court's decision would be a new binding authority with retrospective effect. Instead South Australia submits that at some time which is not clearly specified, Hale's proposition ceased to be the law. At some time in the past that which had a solid existence is said to have dissolved into nothingness.

In State Government Insurance Commission v Trigwell, Gibbs J said:

"Although the rules of the common law develop as conditions change, a settled rule is not abrogated because the conditions in which it was formulated no longer exist. It is now fashionable to criticize the rule in Searle v Wallbank[] as anachronistic, inconsistent with principle and unsuitable to modern conditions, but it is by no means obvious that it would be a reasonable and just course simply to abolish the rule. The question whether the rule should be altered, and if so how, is clearly one for the legislatures concerned rather than for the courts." A fortiori, it does not follow from anachronism that a rule simply dissolves without any court ruling at the time it dissolved, leaving its dissolution to be detected by a court many years or decades later.

To the extent that they may be changed retrospectively, uncertainty is inherent in common law rules. But the standard technique is to make the change in a particular case. It is announced as having happened at the time of that case. Even though it operates retrospectively, that retrospective operation tends to affect only quite recent conduct. That was so in R v R and other cases following it: the conduct charged took place only a short time before the law changed. At least in non-criminal fields, if the change is the result of altering "the law's direction of travel by a few degrees" as distinct from setting "it off in a different direction", no great harm may follow. Assuming it is permissible for the courts to change the substantive criminal law, R v R is an example of the standard technique. It relied on quite recent changes in the status of married women. It did not purport to announce that a change had taken place many years ago, by reason of changes in status even earlier. South Australia's urging of the latter course engenders much more uncertainty. It invites Bentham's reproach:

"Nebuchadnezzar put men to death for not finding a meaning for his dreams: but the dreams were at least dreamt first, and duly notified. English judges put men to death very coolly for not having been able to interpret their dreams, and that before they were so much as dreamt." Retrospective change in the criminal law and the appropriate institutions to effectuate it

South Australia's arguments involve a retrospective change in the criminal law. Indeed, they willingly embrace it. They involve the proposition that conduct no-one saw as attracting criminal liability in 1963 in fact attracted that liability because an historical investigation in 2010-2012 is said to reveal that changes in legal and social conditions at some unspecified time before 1963 caused the conduct to become criminal. And this proposition involves a very serious crime. Rape in 1963 was punishable by life imprisonment and whipping.

The law's aversion to the judicial creation of crimes. In those circumstances, though it may be trite to do so, it is desirable to recall the law's aversion to the judicial creation or extension of crimes. In the early 17th century Bacon put the central difficulty in a retroactive criminal law thus in Aphorism 8 of his treatise De Augmentis:

"Certainty is so essential to law, that law cannot even be just without it. 'For if the trumpet give an uncertain sound, who shall prepare himself to the battle?' So if the law give an uncertain sound, who shall prepare to obey it? It ought therefore to warn before it strikes." And Aphorism 39 read in part:

"Let there be no authority to shed blood; nor let sentence be pronounced in any court upon capital cases, except according to a known and certain law. … Nor should a man be deprived of his life, who did not first know that he was risking it." Hobbes stated in 1651: "harm inflicted for a fact done before there was a law that forbade it, is not punishment, but an act of hostility: for before the law, there is no transgression of the law". Hence, said Hobbes in 1681:

"A Law is the Command of him, or them that have the Soveraign Power, given to those that be his or their Subjects, declaring Publickly, and plainly what every of them may do, and what they must forbear to do." Similarly, Glanville Williams said:

"The citizen must be able to ascertain beforehand how he stands with regard to the criminal law; otherwise to punish him for breach of that law is purposeless cruelty. Punishment in all its forms is a loss of rights or advantages consequent on a breach of law. When it loses this quality it degenerates into an arbitrary act of violence that can produce nothing but bad social effects." Stephen J stated in R v Price: "the great leading rule of criminal law is that nothing is a crime unless it is plainly forbidden by law." Hence Hayek saw it as crucial to the rule of law that "the coercive power of the state … be used only in cases defined in advance by the law and in such a way that it can be foreseen how it will be used." Finally, as Harris said, retroactivity in criminal law is "pointless … because of the brutal absurdity of today commanding someone to do something yesterday."

South Australia submitted, however, that the change it favoured "does not create a new offence, it merely removes a protection, arguably, formerly held by husbands." It submitted that there was no inhibition against judicial legislation which fell short of creating a new offence. South Australia relied on the following statement by the English Court of Appeal in R v R about the judicial abolition of the immunity:

"The remaining and no less difficult question is whether … this is an area where the court should step aside to leave the matter to the Parliamentary process. This is not the creation of a new offence, it is the removal of a common law fiction which has become anachronistic and offensive and we consider that it is our duty having reached that conclusion to act upon it." With respect to both South Australia and the English Court of Appeal, this is captious. The substantive effect of South Australia's argument is to expose persons to a risk of criminal prosecution for conduct which was not believed to be criminal at the time it was carried out. That is true even though South Australia sees this reasoning as doing nothing more pernicious than removing an anachronistic and offensive fiction. Sir John Smith said, correctly, "it is not clear that there is a difference in principle" between the judicial creation of a new offence in Shaw v Director of Public Prosecutions and the judicial abolition of the immunity.

Zecevic v Director of Public Prosecutions (Vict). South Australia also relied on the following statement of Deane J (dissenting) in Zecevic v Director of Public Prosecutions (Vict):

"There may be circumstances in which an ultimate appellate court is justified in overruling a previous decision of its own with the consequence that what had previously been accepted as a defence to a charge of murder is no longer, and never was, such a defence". This was a somewhat selective quotation. There are three reasons why it does not support South Australia's case.

First, Deane J gave as an illustration the case of R v Shivpuri. In that case the House of Lords departed from earlier authority in order to state the true construction of a statute. That is a very different matter from changing the common law. There are more difficulties in courts continuing to apply an erroneous construction of a statute than continuing to apply the received common law. The courts are masters of the common law, but servants of statutes. Further, the case in question was a special one. The change in construction did not cut down any "liberty" the accused had enjoyed. On the earlier and rejected construction, the accused was free to attempt to commit a crime if circumstances unknown to him made it impossible to do so. On the later and favoured construction, an attempt to commit the crime in those circumstances was criminal. The freedom arising under the earlier construction was a strange type of freedom. It was not a freedom which persons in the accused's position could be said to have been able to rely on: the relevant circumstances were unknown to them. Thus, as Juratowitch says:

"The absence of sensible reliance or liberty considerations in the case meant that the prohibition on criminal retroactivity was, without diminishing the strength of that prohibition in general, eminently susceptible to being justifiably overcome in Shivpuri." Secondly, Deane J gave very detailed reasons for not applying his tentative observation to the case before him in Zecevic v Director of Public Prosecutions (Vict). They do not support South Australia's argument.

Thirdly, Deane J made it plain that the undesirability of retroactive changes in criminal law adverse to the accused applies as much to abolishing defences as it does to creating new offences. Thus he said:

"It is simply wrong that an accused may be adjudged not guilty or guilty of murder according to the chance of whether his trial is completed before or after this Court has abolished a defence which, under the law which the Court itself had definitively settled at the time the offence was committed, reduced the offence from murder to manslaughter." He called this "a macabre lottery". The macabre character of the lottery is heightened in this case. Those who have caused the prosecution to be brought have allowed 47 years to pass before charging the appellant.

Another problem with South Australia's argument is that it invites retrospective judicial legislation which collides with legal structures created by parliamentary legislation. Thus Brennan J, the only Justice in R v L not to state that Hale's proposition had ceased to exist, succinctly and correctly said:

"a mere judicial repeal of the [exception] would extend the liability for conviction of the crime of rape to cases which would be excluded from liability for conviction by s 73(5) of the Criminal Law Consolidation Act." Brennan J's point was that s 73(5) permitted convictions for rape in the aggravated circumstances stated in the sub-section, but otherwise preserved the common law "exception" from liability. To "repeal" the common law "exception" would expose husbands to a greater risk of prosecution for acts carried out before s 73(5) was enacted in 1976 than after it. Section 73(5) did not apply retrospectively. The greater exposure of husbands to risk of prosecution would depart significantly from the legislative scheme. It would mean that in 1976 Parliament narrowed the scope of a husband's liability for sexual offences committed against his wife, rather than expanded it. And yet the seeming function of the legislation was to expand liability, not narrow it.

Professional attitudes in 1963-1965. The retrospectivity involved in South Australia's arguments is highlighted by considering the following question. What would actually have happened if, instead of the appellant being charged with rape in 2010, he had been charged immediately after the second of the alleged offences had occurred in April 1963?

Lord Reid said: "the law is what the judge says it is." Mr Justice Holmes, as is well known, remarked that the bad man "does want to know what the Massachusetts or English courts are likely to do in fact. I am much of his mind. The prophecies of what the courts will do in fact, and nothing more pretentious, are what I mean by the law." No doubt it is often unrealistic to assume that people take account of the criminal law in deciding what conduct to engage in. It is probably particularly unrealistic in relation to violent sexual crimes. However, people should be able to know, by recourse to a competent lawyer, what the legal consequences of a proposed course of action are before embarking on it.

What would a bad man in South Australia have learned if he had asked for a prophecy as to what the South Australian courts, and this Court, would be likely to say in the years 1963 to 1965, for example, if he had been charged in April 1963 with raping his wife in March and April and he had challenged the validity of the indictment or appealed against a conviction?

There were at that time seven judges in the Supreme Court of South Australia. The Chief Justice was Sir Mellis Napier, in his 40th year of service on that Court. The senior puisne judge was Sir Herbert Mayo, in his 22nd year of service. Next in seniority came Sir Reginald Roderic St Clair Chamberlain: universally known as "Joe", he did not share the impulsiveness or the radicalism of his namesake. The other judges were Justices Millhouse, Travers, Hogarth and Bright. This Court comprised Chief Justice Dixon and Justices McTiernan, Kitto, Taylor, Menzies, Windeyer and Owen. It is hard to see where a majority of two was to be found in the Supreme Court of South Australia in favour of the view that Hale's proposition never was, or had ceased to be, the law in South Australia. It is equally hard to see where a majority of three or four in favour of that view was to be found in this Court. Indeed, it is hard to find even one vote for that proposition.

This is not simply a crass piece of legal "realism". It does not rest on the personal idiosyncrasies of the individual judges. The probabilities were strongly against either majority because of the particular ideas of the time. They were universal ideas among the Australian judiciary. To overturn Hale's proposition, or to deny that it ever had been the law, or to hold that it had earlier dissolved into nothingness, was to widen the criminal law. It was a legal commonplace in the middle of the 20th century that it was wrong for judges "to declare new offences": that "should be the business of the legislature." So spoke Lord Goddard CJ, Sellers and Havers JJ in 1953, in R v Newland. They also stated that it was wrong for:

"the judges to declare new crimes and enable them to hold anything which they considered prejudicial to the community to be a misdemeanor. However beneficial that might have been in days when Parliament met seldom or at least only at long intervals it surely is now the province of the legislature and not of the judiciary to create new criminal offences." And in Director of Public Prosecutions v Withers, in 1974, Lord Simon of Glaisdale said that it was an "undoubted [principle] of law" that "it is not open to the courts nowadays either to create new offences or so to widen existing offences as to make punishable conduct of a type hitherto not subject to punishment". The same view would prevent any judicial widening of so extremely serious a crime as rape.

That view was well-entrenched among judges, practising lawyers and academic lawyers. Contemporary reaction to two surprising decisions of the House of Lords in 1960 and 1961 demonstrates how well-entrenched it was.

Director of Public Prosecutions v Smith. The first was a murder case, Director of Public Prosecutions v Smith. The trial judge (Donovan J) directed a jury that if the accused, in doing what he did, must as a reasonable man have contemplated that serious harm was likely to occur to the victim, he was guilty of murder - whether or not the accused actually had that contemplation. The Court of Criminal Appeal (Byrne, Sachs and Winn JJ) allowed Smith's appeal and substituted for the verdict of capital murder a verdict of manslaughter. On 28 July 1960, the House of Lords (Viscount Kilmuir LC, Lords Goddard, Tucker, Denning and Parker of Waddington) allowed a prosecution appeal and restored the conviction for capital murder. Viscount Kilmuir LC said that an accused person who was accountable for his actions and who carried out an unlawful and voluntary act was guilty of murder if the ordinary reasonable man would, in all the circumstances of the case, have contemplated grievous bodily harm as the natural and probable result of that act.

Shaw v Director of Public Prosecutions. The second case was decided on 4 May 1961. The House of Lords (Viscount Simonds, Lords Tucker, Morris of Borth-y-Gest and Hodson; Lord Reid dissenting) held in Shaw v Director of Public Prosecutions that the courts could create new crimes. Viscount Simonds said that the courts had:

"a residual power, where no statute has yet intervened to supersede the common law, to superintend those offences which are prejudicial to the public welfare. Such occasions will be rare, for Parliament has not been slow to legislate when attention has been sufficiently aroused. But gaps remain and will always remain since no one can foresee every way in which the wickedness of man may disrupt the order of society." Lord Reid, whose reputation, still high, was extremely high in the early 1960s, dissented. He quoted with approval the second passage from R v Newland set out above. He said: "the courts cannot now create a new offence".

Contemporary reactions to Shaw v Director of Public Prosecutions. Shaw v Director of Public Prosecutions attracted heavy criticism. Probably for that reason, the principle it enunciated has since been narrowed. The modern English view corresponds with Lord Reid's. Thus Lord Bingham of Cornhill said: "there now exists no power in the courts to create new criminal offences".

An example of the contemporary reaction to Shaw v Director of Public Prosecutions is what P J Fitzgerald, a prominent Anglo-Irish criminal lawyer, wrote in 1962:

"Few cases in recent years have been quite so disturbing as this. The resuscitation of the judicial power to create crimes runs counter to two cardinal principles of free and democratic government." Fitzgerald put the first as follows:

"[T]he idea of the rule of law … is based on the demand that the citizen should be ruled by laws and not by the whims of men. In the sphere of criminal law this idea has become crystallized as … a principle according to which only breaches of existing criminal law should be punishable. The justification of this principle, which has been adopted as an actual rule in some legal systems, though not in the English legal system, is that the citizen should be able to know beforehand what conduct is permitted and what forbidden; for only in this way can he order his affairs with certainty and avoid coming into conflict with the law. It is this demand for certainty with regard to the provisions of the criminal law that militates against retrospective criminal legislation. When Parliament creates a new crime, it almost invariably legislates for the future only. This, however, is just what the courts cannot do. Our legal system is such that a court can only decide a point of law which arises in some actual case before the court, and consequently the court's decision always relates back to the facts of this case, facts which of course precede the decision. If, therefore, a court manufactures a new crime, it thereby determines after the event that the defendant's conduct, which at the time of commission was not prohibited by law, is a criminal offence. To countenance this type of retrospective criminal legislation means that certainty and consequently freedom are at an end. Bentham long ago pointed out that when the judges make law like this, they are treating the citizen as a man treats his dog, hitting him every time he does something to which the master takes exception. Animals and young children can only be trained in this way. Sane and adult members of a free society, however, are entitled to demand first to be told what conduct is forbidden so that they may choose whether or not to keep within the law." Fitzgerald put the second objection to "the creation of new offences by the courts" thus:

"Even suppose that a court could decide that the kind of act which the defendant had done would in future, though not in the instant case, constitute a crime, there is still the objection that this type of proceeding is not consonant with democratic government. If Parliament creates a new crime, the citizens whose liberty is thereby restricted have the consolation that this was done by their elected representatives whom they chose to perform this sort of activity, and whom in due course they may re-elect or reject. The judges, on the other hand, are appointed by the Crown, virtually irremovable and in practice accountable to no one. That such a body should have the power to decree that certain acts shall constitute crimes is totally incompatible with the notion of democracy." In similar vein, Lord Reid said that judicial legislation should be avoided when "public opinion is sharply divided on any question". The development of the criminal law raises questions which often sharply divide public opinion.

Lord Reid employed arguments similar to those of Fitzgerald in Shaw v Director of Public Prosecutions. And twice in R v Newland, the Court of Criminal Appeal referred to Sir James Fitzjames Stephen's statement of related arguments 70 years earlier:

"it is hardly probable that any attempt would be made to exercise [a power of declaring new offences] at the present day; and any such attempt would be received with great opposition, and would place the bench in an invidious position. … In times when legislation was scanty, [that power was] necessary. That the law in its earlier stages should be developed by judicial decisions from a few vague generalities was natural and inevitable. But a new state of things has come into existence. On the one hand, the courts have done their work; they have developed the law. On the other hand, parliament is regular in its sittings and active in its labours; and if the protection of society requires the enactment of additional penal laws, parliament will soon supply them. If parliament is not disposed to provide punishments for acts which are upon any ground objectionable or dangerous, the presumption is that they belong to that class of misconduct which it is not desirable to punish. Besides, there is every reason to believe that the criminal law is, and for a considerable time has been, sufficiently developed to provide all the protection for the public peace and for the property and persons of individuals which they are likely to require under almost any circumstances which can be imagined; and this is an additional reason why its further development ought to be left in the hands of parliament." Lord Diplock used similar reasoning in Knuller (Publishing, Printing and Promotions) Ltd v Director of Public Prosecutions to advocate a retreat from Shaw v Director of Public Prosecutions.

The views of a scholar who received his legal education in Adelaide shortly before the appellant allegedly committed the conduct charged are representative of how lawyers thought at that time and in that place:

"the administration, or working-out, of the criminal law's prohibitions is permeated by rules and principles of procedural fairness ('due process of law') and substantive fairness (desert, proportionality), which very substantially modify the pursuit of the goal of eliminating or diminishing the undesired forms of conduct: such principles as nulla poena sine lege (and rather precise leges, at that), and the principles which outlaw retroactive proscription of conduct (at the known cost of letting some dubious characters slip through the net), and restrain the process of investigation, interrogation, and trial (even at the expense of that terror which a Lenin knows is necessary for attaining definite social goals)." (emphasis in original) Ideas of this kind, though perhaps less congenial to the mentalities of recent decades, were very familiar to Australian judges in the early 1960s. They universally assented to those ideas.

Contemporary reactions to Director of Public Prosecutions v Smith. Director of Public Prosecutions v Smith, too, attracted immediate criticism. In Australia it was rightly seen as an extension of the law of murder. Shortly before his death, Sir Wilfred Fullagar, then a Justice of this Court, entered Sir Owen Dixon's chambers and observed: "Well, Dixon, they're hanging men for manslaughter in England now."

The doctrine stated in Director of Public Prosecutions v Smith was soon abolished by statute in England. Glanville Williams called it "the most criticised judgment ever to be delivered by an English court." Lord Reid called it a "disaster". Dixon CJ, in his 35th year on the High Court and nearing the end of his eighth decade, levelled the most damaging criticism of all at it in Parker v The Queen. Judgment was delivered on 24 May 1963. That was at or shortly before the time the present appellant could have had the question of his immunity from prosecution for allegedly raping his wife in March and April 1963 considered by the courts, had the complainant, the police, the prosecuting authorities, and the courts moved expeditiously. Dixon CJ delivered a dissenting judgment. But it concluded with a passage with which all other members of the Court (Taylor, Menzies, Windeyer and Owen JJ) agreed. In reading that passage, it must be remembered that up to 1963 it had been the High Court's practice to follow decisions of the House of Lords. It had also been the Court's practice to pay great respect to the decisions of the English Court of Appeal, and decisions of English High Court judges. That was so even though no appeal lay from any Australian court to those Courts.

"In Stapleton v The Queen we said: 'The introduction of the maxim or statement that a man is presumed to intend the reasonable consequences of his act is seldom helpful and always dangerous'. That was some years before the decision in Director of Public Prosecutions v Smith, which seems only too unfortunately to confirm the observation. I say too unfortunately for I think it forces a critical situation in our (Dominion) relation to the judicial authority as precedents of decisions in England. Hitherto I have thought that we ought to follow decisions of the House of Lords, at the expense of our own opinions and cases decided here, but having carefully studied Smith's Case I think that we cannot adhere to that view or policy. There are propositions laid down in the judgment which I believe to be misconceived and wrong. They are fundamental and they are propositions which I could never bring myself to accept." Dixon CJ then said that Smith's case "should not be used as authority in Australia at all."

Those were terrible words. They were brooding, sombre and unusually passionate. In them the aged Chief Justice revealed that at the end of his career he had plumbed the depths of an intolerable nightmare. His reaction shows the Court being provoked by a retrospective judicial expansion of criminal liability in England into a determination to preserve crucial common law principles in Australia, not applaud or foster their destruction. This Court had changed its own rules of stare decisis in order to preserve Australian law. Those rules are fundamental to the judicial method. The change was very substantial. Though the High Court continued to be bound by Privy Council decisions, on most points of law there was much more authority from the House of Lords and the English Court of Appeal than the Privy Council. For those reasons Parker v The Queen astonished the Australian legal profession. But its repudiation of the thinking underlying Director of Public Prosecutions v Smith accorded with the ideas of the Australian legal profession.

What would the courts have done in 1963-1965? Had the appellant been charged with rape in April 1963, the immediate background to any claim by him of immunity from prosecution would have included the following elements. There was a continuing furore in which Lord Reid's dissent in Shaw v Director of Public Prosecutions was receiving overwhelming favour. There had been an explicit repudiation of English authority for the first time in Australian history in part because of its retrospective expansion of criminal liability. There was universal acceptance in the Australian judiciary of conceptions of the kind stated by Stephen, Lord Reid and Fitzgerald. They were conceptions ultimately rooted in the common understanding of the rule of law. Recourse to the principal English works on criminal law which were available in 1963 or soon to be published would have revealed that Hale's proposition as reflected in recent authorities was stated as the law. The same was true of Australian works

available in 1963 or shortly thereafter. This Court was not taken to any works stating that Hale's proposition was not the law. The leading English and Australian academic lawyers specialising in criminal law - Professor Glanville Williams, Sir Rupert Cross, Sir John Smith, Professor Hogan, Professor Howard, Professor Brett, Professor Waller and Professor Morris - were agreed that the immunity existed. No Australian case denied Hale's proposition. A handful of English cases had qualified it, but only to a small degree.

Against that background, four questions arise. What prospect was there that the South Australian courts or the High Court would accede to an attempt by South Australia to effect a judicial extension, retrospectively, of criminal liability? What prospect was there that they would rule that the immunity had never existed? What prospect was there that they would accede to a submission that though the immunity had existed for a long time, it had disappeared some decades earlier? What prospect was there that they would accede to a submission that though the immunity had existed up to 1963, it should be abolished (necessarily with retrospective effect)? To each of those four questions the answer must be: "None". That answer is supported by the fact that once they came to consider the problem, neither the Australian nor the English courts wavered, until 1991, from the view asserted and assumed until then that Hale's proposition was substantially correct. It is necessary, with respect, emphatically to reject the statement that "in 1963, a respectable challenge to Sir Matthew Hale's opinion could have been mounted." To believe that is to believe that history can be rewritten in complete defiance of all contemporary evidence. It contradicts the reasoning of the House of Lords in R v R.

Foreseeability. South Australia did not rely on an argument which appealed to the European Court of Human Rights. But it is convenient to mention it. That Court held that the United Kingdom was not in contravention of Art 7 of the European Convention on Human Rights by reason of the decision in R v R because the abolition of the immunity was reasonably foreseeable. This is a highly questionable justification for retrospective judicial change in the criminal law. But even if it is an arguable justification, it cannot apply here. It may be one thing to hold that it was reasonably foreseeable in 1990 that the immunity might be abolished in 1991. But in 1963 it was not reasonably foreseeable that if the matter came to court there would be an immediate abolition of the immunity by judicial means.

The significance of R v L. South Australia submitted that statements in R v L supported its second submission. But it accepted that they were unnecessary to the decision in that case, and hence were dicta only. They were dicta about an aspect of the common law - a presumed incapacity to withdraw consent - which had been abolished by statute in every Australian jurisdiction. Further, they were dicta which said only that Hale's proposition was not in 1991 part of the common law of Australia. They said nothing in terms about what the position was in 1963. For the Court in this appeal the question is whether, as a matter of ratio decidendi, not obiter dicta, South Australia's second submission should be recognised as correct. An annihilatingly powerful reason for not recognising it is that it criminalises conduct which, if it took place, was lawful at the time it took place.

Conclusion on South Australia's second submission. A decision by the legislature of South Australia after 1963 to enact a law retrospectively providing that the immunity was abolished with effect from a date before 1963 would have been subject to criticism from many quarters. That would have been significant, not because the critics would have been numerous, but because their criticisms would have been most trenchant. South Australia's submission that the same result is to be achieved by a judicial decision to that effect is open to even greater criticism. The position of the judiciary in this respect is not superior to that of the legislature.

For those reasons South Australia's second submission must be rejected.

The appellant contended that a common law rule should be created by this Court to the effect that when there is a judicial change to the common law it only operates prospectively, not retrospectively. This would involve overruling prior authority. That contention would only become a live issue if the common law as stated by Hale were held to have changed in the past (as South Australia submitted) or were to be changed now (as South Australia did not submit). It is not correct to arrive at either holding. Hence the need to consider the contention does not arise.

The appeal should be allowed. For the answer to the question of law given by the majority in the Full Court of the Supreme Court of South Australia there should be substituted the answer: "No".

BELL J. In 2010, an Information was filed in the District Court of South Australia, charging the appellant with offences including two counts of rape. The complainant in each count was his then wife, GP, with whom he was living at the time. The offences of rape are alleged to have occurred in March and April 1963. It cannot be sensibly suggested that the appellant would have been prosecuted for those offences, had the allegations come to the attention of the authorities in 1963. This is because at that time it was understood that the crime of rape could not be committed by a husband against his wife with whom he was living ("the immunity"). A husband was amenable under the criminal law for any other offence of violence committed against his wife.

The imposition of criminal liability on a person for an act or omission to which criminal liability did not attach at the date the act was done or omitted to be done is contrary to fundamental principle. It is said that the prosecution of the appellant today for his conduct in 1963 does not offend that principle because the immunity has never formed part of the common law of Australia or, if it did, it had ceased to do so sometime before 1963. The first of these alternatives rests on demonstrating either the absence of an authoritative source for the immunity or that in R v L this Court declared the common law in terms that denied its existence. For the reasons that follow, neither of those propositions should be accepted. Nor should this Court now hold that, on some date before 1963, a settled rule of the common law affecting liability for a serious criminal offence ceased to exist.

The appellant was due to stand trial in the District Court of South Australia (Herriman DCJ) on 5 July 2010. On 29 June 2010, he applied to quash the counts in the Information charging him with rape contrary to s 48 of the Criminal Law Consolidation Act 1935 (SA) ("the CLC Act"). Herriman DCJ stated a case reserving a question of law for the determination of the Full Court of the Supreme Court of South Australia. His Honour set out the following facts: each count charged an act of non-consensual penile-vaginal sexual intercourse with GP; GP and the appellant were married and cohabiting as husband and wife at the date of each alleged offence; and no legal orders or undertakings of any kind affected the marital relationship on those dates. The question of law that his Honour reserved is:

"Was the offence of rape by one lawful spouse of another, in the circumstances as outlined above, an offence known to the law of South Australia as at 1963?" The Full Court, by majority (Doyle CJ and White J, Gray J dissenting), answered the question in this way:

"The defendant is liable at law to be found guilty of the offences of rape charged in count 3 and count 5 of the information, notwithstanding that at the time of the alleged offence he was married to the alleged victim and was cohabiting with her, the marriage giving rise to no presumption of consent on her part to intercourse with her husband, and giving rise to no irrebuttable presumption to that effect." By special leave, the appellant appeals to this Court against the answer given by the majority in the Full Court.

Before turning to the Full Court's reasons, some reference should be made to the history of the law governing liability for rape in South Australia and to the decision in R v L.

In 1963, the punishment for the offence of rape was provided by s 48 of the CLC Act:

"Any person convicted of rape shall be guilty of felony, and liable to be imprisoned for life, and may be whipped." The elements of the offence of rape were supplied by the common law.

The understanding that a husband could not be guilty as a principal in the first degree of the rape of his wife is traced to the statement of Sir Matthew Hale in The History of the Pleas of the Crown:

"But the husband cannot be guilty of a rape committed by himself upon his lawful wife, for by their mutual matrimonial consent and contract the wife hath given up herself in this kind unto her husband, which she cannot retract." There does not appear to have been a single case in which a husband had been prosecuted for the rape of his wife with whom he was living in any common law jurisdiction at the date of the conduct with which the appellant is charged. By that date, as will appear, the justification for the immunity may have come to rest more upon the notion that the criminal law ought not to intrude into the marital bedroom, than upon the fiction of the wife's irrevocable consent. By the 1970s, the idea that there could be any justification for conferring immunity on a husband for the rape of his wife was the subject of critical academic attention and pressure for reform of the law.

South Australia was the first of the Australian jurisdictions to respond to the call for reform of the law of rape. In December 1975, the Attorney-General appointed a Committee of persons distinguished for their knowledge of the criminal law to report on the law relating to sexual offences. The Committee was chaired by Justice Roma Mitchell of the Supreme Court of South Australia. The Committee submitted its Report to the Attorney-General in March 1976. The Report contained a summary of the law stating that a husband could not be guilty as a principal in the first degree of the rape of his wife. The Committee noted judicial development of the law in England allowing an exception to the immunity in the case of a wife who had obtained an order for separation relieving her from the obligation to cohabit with her husband. It recommended that the immunity should be confined such that a husband should be liable to conviction for the rape of his wife whenever the act constituting the rape was committed while the two were living apart and not under the same roof.

Following receipt of the Committee's Report, the South Australian Parliament amended the CLC Act by introducing s 48(1), which stated the elements of the offence of rape, and s 73, which relevantly provided:

"(3) No person shall, by reason only of the fact that he is married to some other person, be presumed to have consented to sexual intercourse with that other person. … (5) Notwithstanding the foregoing provisions of this section, a person shall not be convicted of rape or indecent assault upon his spouse, or an attempt to commit, or assault with intent to commit, rape or indecent assault upon his spouse (except as an accessory) unless the alleged offence consisted of, was preceded or accompanied by, or was associated with - (a) assault occasioning actual bodily harm, or threat of such an assault, upon the spouse; (b) an act of gross indecency, or threat of such an act, against the spouse; (c) an act calculated seriously and substantially to humiliate the spouse, or threat of such an act; or (d) threat of the commission of a criminal act against any person." In the period following the South Australian reforms, the parliaments of each of the States and Territories enacted legislation with the evident intention of modifying or abolishing the immunity. This process of reform was completed by December 1991, when R v L was decided. In the Code States, this was achieved by removing the words "not his wife" from the definition of the offence. In the Northern Territory, it was achieved by enacting the Criminal Code (NT) in terms that did not limit rape to an offence outside marriage. In the Australian Capital Territory and New South Wales, it was done by enacting that the fact of marriage was no bar to conviction for the offence. In Victoria and South Australia, any presumption of spousal consent to sexual intercourse on marriage was, in terms, abolished. South Australia was alone in providing a limited immunity for husbands in the case of non-aggravated offences. Further amendments introduced into the CLC Act in 1992 removed this partial immunity.

In R v L, the validity of s 73(3) of the CLC Act was challenged on the ground of inconsistency with Commonwealth law. The claimed inconsistency was with s 114(2) of the Family Law Act 1975 (Cth), which conferred power on the Family Court of Australia to make an order relieving a party to a marriage from any obligation to perform marital services or to render conjugal rights. The Court held that there was no direct or indirect inconsistency between the State and Commonwealth laws. Resolution of the issue presented in R v L did not require consideration of proof of the offence of rape under the common law. Among the submissions advanced on L's behalf was that s 114(2) of the Commonwealth statute preserved the common law inability of a wife to withhold her consent to sexual intercourse with her husband. In their joint reasons, Mason CJ, Deane and Toohey JJ said that, "if it was ever the common law that by marriage a wife gave irrevocable consent to sexual intercourse by her husband, it is no longer the common law". This statement was prominent in the respondent's submissions before the Full Court and on this appeal.

Doyle CJ, writing for the majority in the Full Court, answered the reserved question on the footing that it was likely that Hale's statement of the immunity would have been accepted as a correct statement of the common law of Australia in 1963. Nonetheless, his Honour said the Full Court should apply the considered statement of this Court that any presumption of irrevocable consent to sexual intercourse no longer formed part of the common law. His Honour encapsulated the operation of the declaratory theory of the common law in the following statement:

"Mr P is charged with offences against the then s 48 of the [CLC Act]. In 1963 the elements of that offence were determined by the common law. Today, those elements are determined by the common law as stated by the majority in R v L." Gray J dissented. His Honour considered that the majority in R v L had not declared the common law with respect to liability for rape. He reviewed the history and concluded that the appellant could not have been convicted of the rape of GP in 1963. His Honour would have answered the reserved question in the negative.

Before returning to the decision in R v L, reference should be made to judicial development of the law relating to the immunity in Scotland and England.

In S v HM Advocate, an accused was indicted in the High Court of Justiciary in Scotland for the rape of his wife, with whom he was cohabiting. He challenged the count, contending that no crime known to the law of Scotland had been committed. The motion was dismissed and the dismissal upheld on appeal. Lord Justice-General Emslie, giving the judgment of the Court, noted that there was no authority holding that a cohabiting husband could be convicted of the rape of his wife. His Lordship considered the state of English law to be sufficiently summarised in Glanville Williams' Textbook of Criminal Law:

"A husband is legally incapable of perpetrating rape upon his wife unless the parties are judicially separated, or (probably) separated by consent, or unless the court has issued an injunction forbidding the husband to interfere with his wife, or the husband has given an undertaking to the court in order to avoid the issue of the injunction." Lord Emslie referred with approval to Glanville Williams' views on the justification for the immunity:

"The reason traditionally given for the general rule is the totally unconvincing one that the wife's consent is given on marriage, and she cannot revoke it. It would be an understatement to say that this authentic example of male chauvinism fails to accord with current opinion as to the rights of husbands." The immunity in the law of Scotland was traced to the unequivocal statement of it by Baron Hume, which, in turn, drew on Hale. The Court accepted that Hume's statement of the law may have been correct in the 18th and early 19th centuries. However, the application of the rule in the late 20th century depended on the reasons justifying it and it was said that irrevocable consent, "if it ever was a good reason, no longer applies today".

In 1985, in R v Roberts, the Criminal Division of the Court of Appeal of England and Wales said:

"In our judgment the law is now quite plain on this topic [marital rape]. The status of marriage involves that the woman has given her consent to her husband having intercourse with her during the subsistence of the marriage. She cannot unilaterally withdraw it."

The decision in Roberts followed Steele and allowed that a husband might be convicted of the rape of his wife in circumstances in which he and she had, by mutual agreement or court order, effectively put an end to the wife's fictional consent.

The enactment of s 1(1)(a) of the Sexual Offences (Amendment) Act 1976 (UK), which defined rape in terms incorporating the expression "unlawful sexual intercourse", led to conflicting decisions at the trial court level as to the ability to judicially develop further exceptions to the immunity. The perceived difficulty was occasioned by the recognition that the word "unlawful" in this context had always been understood to refer to sexual intercourse outside marriage.

The Court of Appeal addressed this controversy in R v R. The accused had been convicted of the attempted rape of his wife committed on an occasion in 1989 when the two were living separately. The wife had informed the accused of her intention to petition for divorce but had not commenced proceedings before the date of the offence. Lord Lane CJ, giving the judgment of the Court, said:

"It seems to us that where the common law rule no longer even remotely represents what is the true position of a wife in present day society, the duty of the court is to take steps to alter the rule if it can legitimately do so in the light of any relevant Parliamentary enactment." It was held that the word "unlawful" in the definition was surplusage. Lord Lane CJ stated:

"We take the view that the time has now arrived when the law should declare that a rapist remains a rapist subject to the criminal law, irrespective of his relationship with his victim."

The House of Lords affirmed the decision. Lord Keith of Kinkel (with whom the other members of the House agreed) observed:

"It may be taken that [Hale's dictum] was generally regarded as an accurate statement of the common law of England. The common law is, however, capable of evolving in the light of changing social, economic and cultural developments. Hale's proposition reflected the state of affairs in these respects at the time it was enunciated. Since then the status of women, and particularly of married women, has changed out of all recognition in various ways which are very familiar and upon which it is unnecessary to go into detail." The European Court of Human Rights dismissed an application in respect of the decision in R v R, holding that the accused's conviction did not violate Art 7(1) of the European Convention on Human Rights. The decision had continued a perceptible line of authority dismantling the immunity and the development of the law "had reached a stage where judicial recognition of the absence of immunity had become a reasonably foreseeable development of the law".

The courts in S v HM Advocate and R v R declared the common law of Scotland, England and Wales, taking into account changes in the conditions of society. In this respect, Lord Keith adopted the following statement from S v HM Advocate:

"By the second half of the 20th century, however, the status of women, and the status of a married woman, in our law have changed quite dramatically." The decisions in S v HM Advocate and R v R necessarily operated with retrospective effect. In each case, the conduct giving rise to the charge was alleged to have occurred not long before the date of the decision. That was important to the reasoning of the European Court of Human Rights in dismissing the application in R v R. One issue raised by this appeal that was not present in S v HM Advocate or R v R concerns the imposition of criminal liability in consequence of developing the law to take account of changed social conditions, for conduct that may have occurred before those changes took place.

The Attorneys-General for South Australia, Queensland and the Commonwealth intervened to address a constitutional issue raised by the appellant's third ground of appeal. This ground asserts that, if the common law was capable of further development following the 1976 amendments to the CLC Act, it should only be developed on a prospective basis. The submission was argued by reference to the decision of the House of Lords in In re Spectrum Plus Ltd (in liquidation). It was said in that case that the flexibility inherent in the English legal system permits the prospective overruling of a previous decision in a case in which it would otherwise produce gravely unfair and disruptive consequences for past transactions or events. However, it has been held that a constitutional limitation on the exercise of judicial power does not permit this Court flexibility of that kind.

In their joint reasons in R v L, Mason CJ, Deane and Toohey JJ discussed the content of conjugal rights in the law of marriage, rejecting the submission that the doctrine imposes a continuing obligation on the part of a spouse to consent to sexual intercourse as a legal consequence of marriage. Their Honours noted Lord Lane CJ's statement in R v R that "there can be little doubt that what [Hale] wrote was an accurate expression of the common law as it then stood". They went on to say:

"Without endeavouring to resolve the development of the common law in this regard, it is appropriate for this Court to reject the existence of such a rule as now part of the common law of Australia. … The notion is out of keeping also with recent changes in the criminal law of this country made by statute, which draw no distinction between a wife and other women in defining the offence of rape. It is unnecessary for the Court to do more than to say that, if it was ever the common law that by marriage a wife gave irrevocable consent to sexual intercourse by her husband, it is no longer the common law." (emphasis added; citations omitted) It was unnecessary for the Court to "resolve the development of the common law" because, as their Honours observed, the law had been changed by statute. There was no jurisdiction in Australia in which a presumption of spousal consent to sexual intercourse had any bearing on a person's liability for rape.

The answer to the question of law reserved by Herriman DCJ requires consideration of an issue that was not addressed by the joint reasons or the reasons of Dawson J in R v L, which is the liability under the common law of a cohabiting husband for the rape of his wife.

Did the immunity form part of the received common law?

The laws and statutes of England applicable to the Province of South Australia were received on 19 February 1836. The relevant history and principles are explained by Mason J in State Government Insurance Commission v Trigwell. It is not in question that, if Hale's statement of the immunity was a rule of the common law in 1836, it was part of the laws of England received in South Australia.

In light of the history leading to the enactment of s 73(3) and (5) of the CLC Act, there can be little doubt that the common law of Australia was understood as embodying a rule that a husband was not amenable to conviction for the rape of his wife. It is also evident that, by 1976, the justification for that immunity was not perceived to depend upon the concept of irrevocable consent to intercourse, since the Parliament of South Australia abolished that presumption while maintaining the immunity save for offences committed in circumstances of aggravation.

As will appear, the Parliament of South Australia was not alone in acting upon acceptance that a husband was immune under the common law for the rape of his wife. Nonetheless, it is said that, correctly understood, the common law has never conferred the immunity. This is because Hale did not cite any authority for it and there is no binding judicial decision confirming its existence. These criticisms will be addressed in turn.

An authoritative statement of the law of rape before Hale?

It would be foolish to attempt to state the elements of liability for the offence of rape in the period before Hale. Holdsworth gives an account of the development of the offence in general terms, observing that Bracton would have confined the offence to violent intercourse with a virgin. At the time of the writings attributed to Glanvill, rape was a plea of the Crown, which could be prosecuted by private appeal or on the presentation of the jury. It appears that, in the early period, most prosecutions were by private appeal and that an appeal could be compromised by the marriage of the victim to her assailant. There is evidence that, before the time of Hale, it was a good defence to an appeal of rape to say that the woman was one's concubine.

Holdsworth saw the essentials of the offence of rape as having been defined sometime after the Statute of Westminster II c 34, which made all rapes punishable as felonies. The statute was passed in 1285 in the reign of Edward I. The only authority cited in Holdsworth for the statement of the essentials of the offence is Hale. The explanation for this gap of some 400 years may lie in Professor Baker's account of the development of the criminal law. In the period up to the mid-16th century, the common law comprised the "common learning" found, not only in the yearbooks, but in the oral tradition of the Inns of Court. It was the latter that shaped the criminal law. Few criminal cases were decided in the courts at Westminster and only a small number of criminal cases "trickled into the year books". Much of the record of the criminal law is found in the notes made by readers and these, it would seem, contain little discussion of rape. Professor Baker says that the most visible result of the body of experience of the courts disposing of criminal cases is to be found in the treatises of Crompton, Dalton and Hale, all of whom drew heavily on rulings made at gaol deliveries. It was their selection, rather than the rulings at large, which he suggests influenced the future development of the law.

Hale's statement was of a negative condition of liability for rape. This circumstance tends to explain the absence of prosecutions of husbands for the offence. Consideration of whether Hale's statement of the immunity came to acquire the status of a rule of law (if it was not one in 1736) requires some account of the standing of the Pleas of the Crown among common lawyers.

Sir Matthew Hale held office as Chief Baron of the Exchequer and Chief Justice of the King's Bench successively in the years 1660 to 1676. He died in 1676, leaving instructions in his will prohibiting the publication of any work other than that which he had permitted to be published in his lifetime. At the time of his death, he was writing the Pleas of the Crown, which he had planned as a work in three volumes. Only the first volume was completed. Four years after his death, the House of Commons ordered that it be printed. However, it was not until 1736 that the first edition appeared under the editorship of Sollom Emlyn, barrister of Lincoln's Inn.

Sir William Blackstone acknowledged his debt to Hale and drew on the Pleas of the Crown in his account of felonies in the Commentaries.

Sir James Fitzjames Stephen accorded the composition of the Pleas of the Crown an important place in the evolution of the criminal law in the 17th century. It was, in his estimate, a work "of the highest authority", demonstrating both "a depth of thought and a comprehensiveness of design" that put it in "quite a different category" from Coke's Institutes. Important principles of criminal responsibility were hardly noticed before Hale. Stephen saw the definition of many crimes as settled in the period that separates Coke from Blackstone, and Hale and Foster as having contributed more than any other writers to that development.

Maitland said of Hale that "none had a wider or deeper knowledge of the materials; he was perhaps the last great English lawyer who habitually studied records; he studied them pen in hand and to good purpose". He was, in Maitland's estimate, "the most eminent lawyer and judge of his time". Holdsworth accounted Hale "the greatest historian of English law before Maitland". He considered the Pleas of the Crown to have been left in the best state of any of Hale's works that had not been published at the date of his death. Holdsworth, like Stephen, regarded the treatise highly:

"It was a branch of the law which could not then be adequately described without a very complete knowledge of the history of the law; and, partly because it contained very ancient ideas and rules, partly because it had been added to and in many details modified by a variety of statutes, it greatly needed systematic treatment. Coke and Crompton had summarized it, in a somewhat unsystematic form. Hale, because he was both a competent historian, a competent jurist, and a competent lawyer did the work which they endeavoured to do infinitely better. Ever since its first publication it has been a book of the highest authority." (citations omitted) Holdsworth saw Coke as standing midway between the medieval and the modern law, and Hale as "the first of our great modern common lawyers".

The analysis of the offence in the Pleas of the Crown

Hale described the offence of rape as "the carnal knowledge of any woman above the age of ten years against her will, and of a woman-child under the age of ten years with or against her will". Hale acknowledged Coke for this statement, but proceeded to a much more detailed analysis of proof of the offence. He discussed additional elements (any degree of penetration was sufficient and it was not necessary to prove emission of semen); accessorial liability for the offence; liability in the case of infants under 14 years; liability in the case of consenting females under 12 years; and consent obtained by threat of violence, among other matters. He also gave a deal of attention to the older law concerning appeals of rape, including the concubinage exception. The account of the husband's immunity follows discussion of the latter. The relevant passage is set out below:

"It appears by Bracton ubi supra, that in an appeal of rape it was a good exception, quod ante diem & annum contentas in appello habuit eam ut concubinam & amicam, & inde ponit se super patriam, and the reason was, because that unlawful cohabitation carried a presumption in law, that it was not against her will. But this is no exception at this day, it may be an evidence of an assent, but it is not necessary that it should be so, for the woman may forsake that unlawful course of life. But the husband cannot be guilty of a rape committed by himself upon his lawful wife, for by their mutual matrimonial consent and contract the wife hath given up herself in this kind unto her husband, which she cannot retract." Writing of the criticism that Hale had conjured up the immunity without authority, one commentator has observed that it might be thought incongruous that the law allowed an exception in the case of de facto relationships (for which there is clear evidence before Hale's time) but not de jure relationships. The writer suggests that Hale lacked authority, not for the existence of the immunity, but for confining it to marriage. It should be noticed that Hale said the concubinage exception was no longer good law because of the recognition that a woman may forsake her unlawful way of life. This was stated by way of contrast to sexual intercourse within marriage, which was seen as lawful.

Hale had a commanding knowledge of the work of the courts administering criminal justice. It may safely be taken that husbands were not prosecuted for rape of their wives in the period before the publication of his treatise. Given the subordinate status of married women under the law, this may not surprise. Among the few benefits that the law conferred on the married woman was to immunise her from prosecution for a crime committed by her in her husband's presence. The presumption of the law was that she was bound to obey her husband's command. This is not an idea that readily accommodates the prosecution of the husband for an act of non-consensual sexual intercourse with his wife.

Hale is the source for locating the immunity in contract. It is a rationale that is consistent with Blackstone's treatment of the relations between husband and wife at law. The latter's celebrated account of the nature and effect of coverture was prefaced by the statement: "[o]ur law considers marriage in no other light than as a civil contract". While "[t]he holiness of the matrimonial state" (emphasis in original) was a matter for the ecclesiastical courts, Blackstone emphasised that the temporal courts treated marriage like all other contracts, asking whether the parties were willing and able to contract. It is an analysis which has been seen as a civilised advance on the medieval concept of the husband's natural and God-given power over his wife. Professor Stretton suggests that, for Blackstone, the fundamental point was that married women consented to their modified legal status by their agreement to marriage:

"It was therefore the logic of contract that justified married women's particular treatment at law. However, it was a narrow concept of consent that ended abruptly at the church door, with no room for renegotiation during marriage and virtually no effective ability to escape the legal effects of marriage through separation or divorce." Blackstone's treatment of rape was largely taken from Hale. He did not refer to the immunity, but it is evident that Hale's statement of it was not controversial. Blackstone drew attention to those occasions on which Hale's account of the law departed from the views of other writers. In Blackstone's analysis of the offence of rape, there was one such occasion. He noted that Hale considered that carnal knowledge of a girl aged under 12 years was rape regardless of consent, but that the law had in general been held only to extend to the carnal knowledge of a girl aged under 10 years.

Hale's statement of the law may be analysed in either of two ways. First, that the offence comprises two elements: (i) carnal knowledge of a female (involving some degree of penetration); (ii) without her consent. On this analysis the immunity arises by the operation of an irrebuttable presumption of law. The alternative analysis is that the first element of the offence requires proof of the "unlawful" carnal knowledge of a female and that "unlawful" in this context means outside marriage. The latter view accords with the treatment of the offence by text-writers, including Coke and Hawkins writing before the publication of Hale's treatise. It is the analysis adopted by the Supreme Court of South Australia in those cases in which consideration has been given to the question. Bray CJ, discussing the elements of the offence in R v Brown, considered that they were as stated in the 37th edition of Archbold: "Rape consists in having unlawful sexual intercourse with a woman without her consent by force, fear or fraud". The word "unlawful" was thought by Bray CJ to exclude intercourse between spouses. Wells J appears to have been of the same view. King CJ in R v Sherrin (No 2) also considered proof of the unlawfulness of the act of intercourse to have undoubtedly been an element of the offence at common law.

The resolution of the reserved question does not turn on whether the rule of law traced to Hale requires proof of the unlawfulness of the intercourse as an element, or is an irrebuttable presumption of consent. The latter, while "disguised in the language of adjective rules", is in truth a substantive rule of law. A husband could not be convicted as principal in the first degree for the rape of his wife on either analysis.

At issue is the existence of the immunity, not whether the reason given for it is flawed or has, over time, ceased to provide a principled basis for it. A number of common law rules of liability for criminal offences have their origins in discredited ideas. The definition of the offence of murder stated by Coke, and thereafter accepted as an authoritative statement of the elements of the offence, required that the death of the deceased take place within a year and a day of the act causing death. The reason for the rule is suggested to be the limitations of medieval medical knowledge. If that is the reason, it must be said that the rule survived long after its justification ceased. The rule has since been abolished by statute. In the same category is the presumption that a boy under 14 years of age is physically incapable of sexual intercourse. This, too, is traced to the statement of the law in the Pleas of the Crown. The presumption is patently absurd. Nonetheless, it was accepted as a rule of law precluding the conviction of boys for rape until it was abolished by statute. It was sufficient for Lord Coleridge CJ in R v Waite to observe that the rule had been "clearly laid down by Lord Hale" and, on that authority, judges had "refused to receive evidence to shew that a particular prisoner was in fact capable of committing the offence".

Hale's statement of the immunity was taken as an authoritative statement of the law by all the leading text-writers.

The first judicial consideration of the immunity was in R v Clarence. A bench of 13 judges was constituted to consider the question of whether the transmission of gonorrhoea by husband to wife in an act of consensual sexual intercourse could amount to the malicious infliction of grievous bodily harm. Wills, Field and Hawkins JJ each left open that circumstances may exist in which a husband could be liable for the rape of his wife. Wills J doubted that "between married persons rape is impossible". Field J thought that there may be cases in which a husband could be convicted of a crime arising out of forcibly imposing sexual intercourse on his wife; he did not say whether for rape or some other offence. Hawkins J accepted that, by the marriage contract, a wife confers on her husband "an irrevocable privilege to have sexual intercourse with her during such time as the ordinary relations created by such contract subsist between them" and that a husband could not be convicted of a rape committed by him upon the person of his wife. However, a husband was not at liberty to endanger his wife's health and cause her grievous bodily harm by the exercise of "the marital privilege" at a time when he was suffering from venereal disease and when the natural consequence of sexual intercourse would be the communication of that disease to her. He explained the principles in this way:

"Rape consists in a man having sexual intercourse with a woman without her consent, and the marital privilege being equivalent to consent given once for all at the time of marriage, it follows that the mere act of sexual communion is lawful; but there is a wide difference between a simple act of communion which is lawful, and an act of communion combined with infectious contagion endangering health and causing harm, which is unlawful. … The wife submits to her husband's embraces because at the time of marriage she gave him an irrevocable right to her person. The intercourse which takes place between husband and wife after marriage is not by virtue of any special consent on her part, but is mere submission to an obligation imposed upon her by law. Consent is immaterial." (emphasis in original) Pollock B said:

"The husband's connection with his wife is not only lawful, but it is in accordance with the ordinary condition of married life. It is done in pursuance of the marital contract and of the status which was created by marriage, and the wife as to the connection itself is in a different position from any other woman, for she has no right or power to refuse her consent." Consideration of the immunity in Clarence appears to have been prompted by a submission based on a footnote in Stephen's Digest of the Criminal Law. The law in the first edition of the Digest was stated, relevantly, in this way:

"Rape is the act of having carnal knowledge of a woman without her conscious permission ... Provided that: - (1) A husband [it is said] cannot commit rape upon his wife by carnally knowing her himself, but he may do so if he aids another person to have carnal knowledge of her." The footnote relevantly said:

"Hale's reason is that the wife's consent at marriage is irrevocable. Surely, however, the consent is confined to the decent and proper use of marital rights. If a man used violence to his wife under circumstances in which decency or her own health or safety required or justified her in refusing her consent, I think he might be convicted of rape, notwithstanding Lord Hale's dictum. He gives no authority for it, but makes the remark only by way of introduction to the qualification contained in the latter part of clause (1), for which Lord Castlehaven's Case (3 St Tr 402) is an authority."

Stephen J gave the leading judgment in Clarence. He used the occasion to draw attention to the alteration of the footnote, removing the suggestion that a man might in certain circumstances be indicted for the rape of his wife, in the most recent edition of his text.

Stephen was a great master of the criminal law. An account of his draft criminal code and the subsequent Commission appointed to report upon it is contained in the joint reasons in Darkan v The Queen. To the extent that the Draft Code appended to the Report of the Commissioners differed from Stephen's original draft, the differences were noted in the Report. The provisions dealing with offences against the person were said to correspond (as did the provisions in Stephen's original draft) with the Offences against the Person Act 1861 (UK), "supplemented by a reduction to writing of the common law doctrines and definitions". The Offences against the Person Act 1861 (UK) prescribed the punishment for rape but left the definition of the offence to the common law. It is apparent that the definition of rape in the Draft Code was understood by its authors to be a statement of the common law. Relevantly, the offence was defined as "the act of a man having carnal knowledge without her consent of a female who is not his wife". The Criminal Code Indictable Offences Bill 1878 (UK), on which the Commissioners' draft was based, and which defined rape in the same terms, had been circulated to the Judges, Chairmen and Deputy Chairmen of Quarter Sessions, Recorders and "many members of the bar and other gentlemen having practical experience in the administration of the criminal law" in England and Ireland with the invitation to comment on it. The absence of any suggestion in the Commissioners' Report that the offence of rape was to be modified under the Code is eloquent of the acceptance by those engaged in the administration of the criminal law in England and Ireland at the time that the offence could not be committed by a husband against his wife.

Sir Samuel Griffith drew on the English Draft Code in preparing his draft criminal code for Queensland. In the latter, the offence of rape was defined, relevantly, as the "carnal knowledge of a woman, not his wife". The marginal notes reveal that Sir Samuel Griffith considered this definition to be a statement of the common law.

In Canada, before the enactment of the Criminal Code in 1892, the offence of rape, while punishable as a felony under legislation modelled on the Offences against the Person Act 1861 (UK), depended upon the common law for its elements of proof. It is apparent that the understanding in that jurisdiction was that the offence could not be committed by a husband against his wife. The Criminal Code defined rape as involving the "carnal knowledge of a woman who is not his wife". It does not appear that this was thought to involve any departure from the existing law.

The absence of a binding decision does not mean that a rule stated in authoritative texts and accepted and acted upon by the legal profession over many years may not acquire status as law. The point is made by Sir John Smith

in his commentary on R v C, by reference to Foakes v Beer. In the latter case, the House of Lords held itself bound to follow a rule stated by Coke to have been laid down in Pinnel's Case in 1602, although their Lordships disliked it and there was no decision in which it had been applied. As the Earl of Selborne LC put it:

"The doctrine itself, as laid down by Sir Edward Coke, may have been criticised, as questionable in principle, by some persons whose opinions are entitled to respect, but it has never been judicially overruled; on the contrary I think it has always, since the sixteenth century, been accepted as law. If so, I cannot think that your Lordships would do right, if you were now to reverse, as erroneous, a judgment of the Court of Appeal, proceeding upon a doctrine which has been accepted as part of the law of England for 280 years." Brennan J, the only Justice in R v L to consider proof of the offence of rape under the common law, considered the elements to have been fixed by Hale's statement of them. The evidence in favour of that conclusion is compelling.

It was submitted that legal and social changes to the status of married women had produced the result that the immunity had ceased to be a rule of law on a date before the subject events. There were differing views about when that change to the law occurred, a circumstance which tends to highlight a difficulty with accepting the underlying premise. The respondent and the Attorney-General for South Australia contended that the foundation for the immunity had "crumbled to dust" as at the "early to mid twentieth century". The Attorney-General of the Commonwealth contended that the relevant change in circumstances had occurred "by the end of the 19th century". Reference was made to the enactment of the Married Women's Property Acts; the amendment of matrimonial causes statutes removing the "double-standard" relating to adultery as a ground for dissolution of marriage; and, more generally, the extension of the franchise to women, as combining to produce a state of affairs that was inconsistent with the continued existence of the immunity. These submissions were maintained in the face of a good deal of evidence to the contrary.

The one case relied on to support the submissions was R v Jackson. In that case, Lord Halsbury LC rejected the proposition that the relation of husband and wife gave the husband "complete dominion over the wife's person". The holding that an order for restitution of conjugal rights did not confer on the husband a right to imprison his wife is a tenuous basis for concluding that the husband was now amenable to prosecution for having sexual intercourse with his wife without her consent.

In the first edition of Halsbury, published in 1909, almost 20 years after the decision in Jackson, the law was stated as being that "[a] man cannot be guilty as a principal in the first degree of a rape upon his wife, for the wife is unable to retract the consent to cohabitation which is a part of the contract of matrimony".

In Tasmania, the Married Women's Property Act was enacted in 1882. Women had been granted the franchise for both federal and State parliamentary elections by 1904. The Matrimonial Causes Act 1860 (Tas) was amended in 1919 to remove the double‑standard with respect to adultery. Nonetheless, when the Parliament enacted the Criminal Code for Tasmania in 1924, a quarter of a century after the enactment of the Queensland Criminal Code, the crime of

rape was defined in the same way as under the latter. The significant changes in the legal status of married women which had occurred by 1924 do not appear to have been viewed at the time as inconsistent with the immunity.

In the same year, the House of Lords delivered judgment in G v G. That was an appeal from the dismissal of an application for a decree of nullity of marriage brought by a husband on the ground of his wife's impotency. The appellant and his wife were married in 1913 and the evidence of their relations spanned the period from that date to 1921. The wife had evinced an hysterical reluctance to engage in sexual intercourse. The question for the court was whether this psychological obstacle to consummation amounted to incapacity, as distinct from the mere wilful refusal of conjugal rights. The court below had doubted that the husband's repeated attempts at intercourse had exhibited "a sufficient virility". It was in this context that Lord Dunedin observed:

"It is indeed permissible to wish that some gentle violence had been employed; if there had been it would either have resulted in success or would have precipitated a crisis so decided as to have made our task a comparatively easy one." His Lordship considered the husband's account "as to why he did not use a little more force than he did" to have been an acceptable explanation and the appeal was allowed. The speeches in G v G speak to another age. The decision in that case is closer to the date of the acts charged against the appellant than was the hearing of this appeal.

More than a decade after the events giving rise to this appeal, in 1975, Lawton LJ, giving the judgment of the English Court of Appeal in R v Cogan, proceeded upon acceptance that it was a legal impossibility for a man to rape his wife during cohabitation. The accused bore accessorial liability for the rape of his wife by another. In the following year, Geoffrey Lane LJ extended the exception to the immunity to allow the conviction of a husband for the rape of his wife where he had given an undertaking not to molest her. Of present significance is his Lordship's view that, "[a]s a general principle, there is no doubt that a husband cannot be guilty of rape upon his wife". The undertaking given in lieu of an injunction operated in that case to eliminate the wife's matrimonial consent to intercourse.

A convenient account of the law in England as it was understood in December 1983 is contained in the Report of the Criminal Law Revision Committee, which had been asked to review the law relating to, and penalties for, sexual offences:

"In defining rape the Sexual Offences (Amendment) Act 1976 uses the term 'unlawful sexual intercourse'. What is 'unlawful' is left to the common law. The general rule is that sexual intercourse is 'unlawful' if it occurs outside marriage. Sexual intercourse between husband and wife is not 'unlawful' except in a fairly narrow class of cases, which can be broadly described as cases where the parties have separated and their separation has been acknowledged by a court." The existence of the immunity was also accepted in decisions of Australian courts delivered after 1963. Reference has been made earlier in these reasons to decisions of the South Australian Supreme Court. In New South Wales, Victoria and Tasmania, the English line of authority allowing an exception to the immunity in the case of a wife living separately and under the protection of a court order was adopted. In R v McMinn, Starke ACJ observed:

"There can be no doubt that for centuries the law in England (and in Australia) has been that a man cannot rape his wife. That this principle of law is out of tune with modern thinking has been recognized in Victoria by the Crimes (Sexual Offences) Act 1980 and there are similar Acts in other States." In New Zealand, a statute enacted in 1961 provided that no man could be convicted of rape of his wife unless, at the time of the intercourse, there was in force a decree nisi of divorce or nullity and the parties had not resumed cohabitation, or there was in force a decree of judicial separation or a separation order. An amendment to the statute in 1981 maintained the immunity, save in cases where the husband and wife were living separately. This restricted immunity was not removed until 1986.

The Model Penal Code, first published by the American Law Institute in 1962, relevantly provided that "[a] male who has sexual intercourse with a female not his wife is guilty of rape". In the revised commentary, published in 1980, this "traditional limitation" of the offence was maintained.

The proposition that by the mid-20th century or earlier the immunity had fallen into desuetude as the result of changes in the conditions of society is without support. In this country, as in other common law countries, the continued existence of the immunity does not appear to have been seen as inconsistent with the recognition of the equal status of married women. There is the curious spectacle in this appeal of the respondent and the Attorney-General for South Australia contending that the maintenance of the immunity by the mid-20th century was inconsistent with the rights and privileges of married women, notwithstanding that as late as 1976 the Parliament of South Australia chose to preserve it.

By the mid-20th century, the notion that the immunity depended on the wife's irrevocable consent to intercourse may no longer have been seen as the justification for it. However, this is not to accept that the immunity had "crumbled to dust". The contemporary evidence suggests that the immunity was a recognised and accepted feature of the law of rape, albeit that the rationale supporting it may have changed.

In 1954, Norval Morris and A L Turner, both then senior lecturers in law at the University of Melbourne, writing of the law respecting marital rape, were critical of irrevocable consent as the justification for the immunity. They went on to discuss the "special position" of a married couple in law and in fact and to say:

"Intercourse then is a privilege at least and perhaps a right and a duty inherent in the matrimonial state, accepted as such by husband and wife. In the vast majority of cases the enjoyment of this privilege will simply represent the fulfilment of the natural desires of the parties and in these cases there will be no problem of refusal. There will however be some cases where, the adjustment of the parties not being so happy, the wife may consistently repel her husband's advances. If the wife is adamant in her refusal the husband must choose between letting his wife's will prevail, thus wrecking the marriage, and acting without her consent. It would be intolerable if he were to be conditioned in his course of action by the threat of criminal proceedings for rape." The leading Australian text on the criminal law published in 1965 praised the decision in R v Clarke, which allowed an exception to the immunity; however, the author went on to observe:

"[A] husband should not walk in the shadow of the law of rape in trying to regulate his sexual relationships with his wife. If a marriage runs into difficulty, the criminal law should not give to either party to the marriage the power to visit more misery upon the other than is unavoidable in the nature of things." The Mitchell Committee explained its reasons for proposing to confine the immunity in this way:

"The view that the consent to sexual intercourse given upon marriage cannot be revoked during the subsistence of the marriage is not in accord with modern thinking. In this community today it is anachronistic to suggest that a wife is bound to submit to intercourse with her husband whenever he wishes it irrespective of her own wishes. Nevertheless it is only in exceptional circumstances that the criminal law should invade the bedroom. To allow a prosecution for rape by a husband upon his wife with whom he is cohabiting might put a dangerous weapon into the hands of the vindictive wife and an additional strain upon the matrimonial relationship. The wife who is subjected to force in the husband's pursuit of sexual intercourse needs, in the first instance, the protection of the family law to enable her to leave her husband and live in peace apart from him, and not the protection of the criminal law. If she has already left him and is living apart from him and not under the same roof when he forces her to have sexual intercourse with him without her consent, then we can see no reason why he should not be liable to prosecution for rape." (emphasis added) The views expressed by the Mitchell Committee were in line with those expressed by the authors of the revised commentaries to the US Model Penal Code in 1980 and by the English Criminal Law Revision Committee in 1984.

In R v C, the English Court of Appeal set out the advice that an imagined solicitor might have given a husband who inquired as to the legality of marital rape in 1970. This was in the context of a submission respecting the foreseeability of further development of the law in light of decisions which had allowed exceptions to the immunity. The Commonwealth Attorney-General submitted that the hypothesised advice applied with equal force in this case. The determination of the issue raised by this appeal does not depend upon consideration of foreseeability of change to the law. Nonetheless, the opinions of the academic lawyers and the members of law reform committees set out above may suggest that the solicitor in R v C was a man in advance of his times.

There is a more fundamental difficulty with the submission that the Court should hold that a substantive rule of law affecting liability for a serious criminal offence has simply disappeared because of a perception that changed conditions of society no longer provided a justification for it. The powerful reasons against an ultimate court of appeal varying or modifying a settled rule or principle of the common law apply with particular force to a variation or modification which has the effect of extending criminal liability. It is for the parliament to determine that a rule of exemption from criminal liability is no longer suited to the needs of the community.

The respondent and the Attorney-General for South Australia submitted that it is the responsibility of this Court to modify the law to avoid the "unjust" operation of a rule of immunity respecting criminal liability. The submission is singular, given that there is no jurisdiction in Australia in which the common law governs a husband's liability for the rape of his wife. No occasion arises to modify the law to make it "an effective instrument of doing justice according to contemporary standards in contemporary conditions". The law of marital rape in each Australian jurisdiction has been brought into line with contemporary standards. Any statement of the common law respecting the liability of a husband for the rape of his wife with whom he was living could only apply to offences alleged to have been committed before the enactment of the statutory reforms.

The declaration of the law for which the respondent contends carries with it that the parliaments of the States and Territories legislated over the course of the last century upon a wrong understanding of the law. That understanding was reflected in the Code States in the way in which the offence of rape was defined. In those States, the position remains that a husband is not liable to be convicted for the rape of his wife before the date on which the words "not his wife" were removed from the Criminal Code. In the jurisdictions which preserved the common law, the declaration would make it possible to reach back beyond the date on which statutory reforms were effected and attach liability to conduct occurring not less than a quarter of a century ago. In South Australia, it would be possible to successfully prosecute a man for the rape of his wife in the years up to 1976. In the more recent past, the same man would enjoy an immunity for the same conduct. That is because the 1976 amendments enacted by the South Australian Parliament with the evident intention of limiting the immunity would now be seen to have conferred it. The fact that the parliaments of every Australian jurisdiction enacted legislation upon the understanding that the immunity was a rule of the common law provides some evidence that it was; and is a good reason for this Court not to now declare it to be otherwise.

The rule of law holds that a person may be punished for a breach of the law and for nothing else. It is abhorrent to impose criminal liability on a person for an act or omission which, at the time it was done or omitted to be done, did not subject the person to criminal punishment. Underlying the principle is the idea that the law should be known and accessible, so that those who are subject to it may conduct themselves with a view to avoiding criminal punishment if they choose. However, its application does not turn on consideration of whether a person might be expected to have acted differently had he or she known that the proposed conduct was prohibited. Deane J's dissenting reasons in Zecevic v Director of Public Prosecutions (Vict) explain why that is so:

"The vice of such a retrospective abolition of a defence to a charge of murder lies not in the prospect of injustice to some imaginary killer who has killed on the basis that his crime will be reduced from murder to manslaughter in the event that he was found to have been acting excessively in self-defence. It lies in the fundamental injustice of inequality under the law which is unavoidable when the administration of the criminal law is reduced to a macabre lottery by what the late Professor Stone described as flagrant violation of the 'well-established judicial policies of the criminal law in favorem libertatis, and against ex post facto punishment'." The departure from the statement of the elements of self‑defence in Viro v The Queen, sanctioned by the majority in Zecevic, was undertaken in circumstances in which it was considered unlikely to occasion injustice and in which it was acknowledged that the endeavour to state the "defence" by reference to the onus had proved unworkable. Nothing in the judgments in Zecevic affords support for the acceptance of the respondent's contention that this Court should restate the common law with the effect of extending criminal liability to a class of persons previously exempt from that liability.

The common law was demeaning to women in its provision of the immunity. It is no answer to that recognition to permit the conviction of the appellant for an act for which he was not liable to criminal punishment at the date of its commission.

For these reasons I would allow the appeal, set aside the answer to the question of law given by the majority in the Full Court and, in lieu thereof, answer that question "no".