|[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]|
United Kingdom House of Lords Decisions
You are here: BAILII >> Databases >> United Kingdom House of Lords Decisions >> A, R v.  UKHL 25 (17 May 2001)
Cite as:  1 AC 45,  2 WLR 1546,  AC 45,  UKHRR 825,  2 Cr App R 21,  HRLR 48, 11 BHRC 225, (2001) 165 JPN 750,  UKHL 25, (2001) 165 JP 609,  3 All ER 1,  2 Cr App Rep 21,  Cr App R 21
[New search] [Context] [Printable version] [Buy ICLR report:  1 AC 45] [Buy ICLR report:  2 WLR 1546] [Help]
Lord Slynn of Hadley Lord Steyn Lord Hope of Craighead Lord Clyde Lord Hutton
(ON APPEAL FROM THE COURT OF APPEAL (CRIMINAL DIVISION))
ON 17 MAY 2001
 UKHL 25
LORD SLYNN OF HADLEY
1. In recent years it has become plain that women who allege that they have been raped should not in court be harassed unfairly by questions about their previous sex experiences. To allow such harassment is very unjust to the woman; it is also bad for society in that women will be afraid to complain and as a result men who ought to be prosecuted will escape.
2. That such questioning about sex with another or other men than the accused should be disallowed without the leave of the court is well established. It was recognised in section 2 of the Sexual Offences (Amendment) Act 1976 which provided that without the leave of the judge there should be no evidence or cross examination by or on behalf of the defendant of a complainant's sexual experience with a person other than the accused. Leave was only to be given by the judge "if and only if he is satisfied that it would be unfair to that defendant to refuse to allow the evidence to be adduced or the question to be asked".
3. Such a course was necessary in order to avoid the assumption too often made in the past that a woman who has had sex with one man is more likely to consent to sex with other men and that the evidence of a promiscuous woman is less credible.
4. Evidence of previous sex with the accused also has its dangers. It may lead the jury to accept that consensual sex once means that any future sex was with the woman's consent. That is far from being necessarily true and the question must always be whether there was consent to sex with this accused on this occasion and in these circumstances.
5. But the accused is entitled to a fair trial and there is an obvious conflict between the interests of protecting the woman and of ensuring such fair trial. Such conflict is more acute since the Human Rights Act 1998 came into force. The question is whether one of these interests should prevail or whether there must be a balance so that fairness to each must be accommodated and if so whether it has been achieved in current legislation. That is essentially the question which arises in this case. I gratefully adopt the statement of the facts and the relevant statutory provisions set out in the text of the speech prepared by my noble and learned friend Lord Steyn.
6. The question certified by the Court of Appeal which gave leave to appeal to your Lordships' House is
7. Section 41 of the Youth Justice and Criminal Evidence Act 1999 prohibits the giving of evidence and cross examination about any sexual behaviour of the complainant except with leave of the court. Leave may be given where a) consent is an issue and where the sexual behaviour of the complainant is alleged to have taken place "at or about the same time as the event which is the subject matter of the charge against the accused" (section 41(3)(b)) and b) where the sexual behaviour of the complainant to which the question or evidence relates is alleged to have been "in any respect, so similar" to the sexual behaviour which is shown by evidence to have taken place as part of the event which is the subject matter of the charge or to any other sexual behaviour of the complainant which took place at or about the same time as that event "that the similarity cannot reasonably be explained as a coincidence" (section 41(3)(c)).
8. Such questions are not to be allowed if their purpose is to establish material to impugn the credibility of the complainant as a witness. Leave may also be given if the evidence of the complainant's sexual behaviour goes no further than to rebut prosecution evidence.
9. It is apparent that prima facie the restriction placed on the court's power to give leave seriously limits the opportunities for cross examination or the adducing of evidence on behalf of the accused. The limitation in section 41(3)(b) to conduct "at or about the same time" as the event charged would prima facie prohibit questions as to a continuous period of cohabitation or sexual activity, or as to individual events more than a very limited period before the event, the subject matter of the charge. The requirement that the sexual behaviour relied on must be so similar to the sexual activity which took place as part of the event charged or be so similar to any other sexual behaviour which took place "at or about the same time" as the event charged that the similarity cannot "reasonably be explained as a coincidence" is on the face of it very restrictive.
10. The need to protect women from harassment in the witness box is fundamental. It must not be lost sight of but I suspect that the man or woman in the street would find it strange that evidence that two young people who had lived together or regularly as part of a happy relationship had had sexual acts together, must be wholly excluded on the issue of consent unless it is immediately contemporaneous. The question whether such evidence should be believed and whether it is sufficient to establish consent or even belief in consent are different matters. The man and woman in the street might also find it strange that evidence may be given and cross examination allowed as to belief in consent but not to consent itself when the same evidence was being relied on. That distinction has been recognised in the cases but without in any way resiling from a strong insistence on the need to protect women from humiliating cross examination and prejudicial but valueless evidence, it seems to me clear that these restrictions in section 41 prima facie are capable of preventing an accused person from putting forward relevant evidence which may be evidence critical to his defence, whether it is as to consent or to belief that the woman consented. If thus construed section 41 does prevent the accused from having a fair trial then it must be declared to be incompatible with the Convention.
11. But the prima facie let alone the literal readings are not the end of the inquiry. Section 3 of the Human Rights Act 1998 requires that
12. I was initially tempted to think that the words "at or about the same time as the event" could be given a wide meaningcertainly a few hours perhaps a few days when a couple were continuously together. But that meaning could not reasonably be extended to cover a few weeks which are relied on in the present case and I consider in the event that even if read with Article 6 they must be given a narrow meaning which would not allow the evidence or cross examination in the present case or in other than cases where the acts relied on were really contemporaneous.
13. Section 41(3)(c) raises a different issue. Although if read literally or even perhaps purposively this provision is very restrictive, I think disproportionately restrictive, it is less precise than section 41 (3)(b). The section must be read and given effect in a way "which is compatible with the Convention rights" in so far as it is possible to do so. It seems to me that your Lordships cannot say that it is not possible to read section 41(3)(c) together with Article 6 of the Convention rights in a way which will result in a fair hearing. In my view section 41(3)(c) is to be read as permitting the admission of evidence or questioning which relates to a relevant issue in the case and which the trial judge considers is necessary to make the trial a fair one.
14. I do not consider that the provisions of section 41(5) admitting rebuttal evidence are sufficient in themselves to avoid unfairness. They are limited in their effect.
15. I agree with the statement in paragraph 46 of Lord Steyn's speech as to the effect of the decision today.
16. Despite the somewhat unusual procedural route which this case has taken, I think that the right course is to dismiss the appeal. The case should now be referred back to the trial judge for him to continue the case in the light of the present decision.
I. The Judge's preliminary rulings
17. In December 2000 the respondent (the defendant) was due to stand trial in the Crown Court on an indictment charging him with an offence of rape, the particulars being that on 14 June 2000 he raped the complainant. The defendant's defence is that sexual intercourse took place with the complainant's consent. It appears that he will alternatively rely on the defence that he believed that she consented.
18. The Crown's case is that the complainant first met the defendant together with a friend on or about 26 May 2000. The complainant and the defendant's friend formed a sexual relationship. The complainant visited the friend at the flat which he was then sharing with the defendant. At about 9 pm on 13 June 2000 the complainant and the friend had sexual intercourse at the flat when the defendant was not there. Later, when the defendant returned, the complainant, the friend and the defendant went for a picnic on the riverbank of the Thames. The friend and the defendant drank whisky and beer. When they got back to the flat the friend collapsed. An ambulance was called and the friend was taken to hospital. Later, in the early hours of 14 June 2000, the defendant and the complainant left the flat intending to walk to the hospital. The defendant led the way and chose a route which took them close to the river. As they walked along the towpath the defendant fell down. The complainant's account is that she tried to help him to his feet, whereupon he pulled her to the ground and had sexual intercourse with her. Later that day the complainant made a complaint of rape to the police. The police interviewed the defendant. Following the advice of his solicitor he declined to answer questions. He read a prepared statement in which he asserted in very general terms that "she was never against this sexual relationship that we were having".
19. According to the statement of facts and issues it is the defendant's case that:
20. On 8 December 2000 a preparatory hearing took place pursuant to section 29 of the Criminal Procedure and Investigations Act 1996. Counsel for the defendant applied for leave to cross-examine the complainant about the alleged previous sexual relationship between them and to lead evidence about it. Relying on the provisions of section 41 of the Youth Justice and Criminal Evidence Act 1999 the judge ruled: (i) that the act of consensual sexual intercourse with the friend could be put to the complainant in cross-examination; (ii) that the complainant could not be cross-examined, nor could evidence be led, about her alleged sexual relationship with the defendant; (iii) that the prepared statement could not be put in evidence.
21. The judge observed that this ruling would prima facie result in a breach of the right to a fair trial under article 6 of the Convention for the Protection of Human Rights and Fundamental Freedoms as scheduled to the Human Rights Act 1998. Pursuant to section 35 of the 1996 Act the judge gave leave to the defendant to appeal to the Court of Appeal. The defendant exercised that right.
II. The decision of the Court of Appeal
22. The defendant appealed against the judge's rulings. In giving the judgment of the Court of Appeal Rose LJ pointed out that the judge's first ruling, viz giving leave to cross-examine the complainant about sexual intercourse with the friend of the defendant, was made in error. No such leave had been sought.
23. The judgment was, however, principally concerned with the rulings by the judge that the complainant could not be asked whether, nor could the defendant give evidence that, she had sexual intercourse with the defendant on occasions during the previous three weeks: R v Y, The Times, 13 February 2001. Rose LJ recorded a concession by the Crown, rightly made in his view, that the questioning and evidence in relation to the complainant's alleged prior sexual activity with the defendant was admissible under section 41(3)(a) of the 1999 Act in relation to the defendant's belief in the complainant's consent: see section 1 of the Sexual Offences (Amendment) Act 1976. It followed that the judge's ruling in entirely excluding such evidence was wrong. On the other hand, Rose LJ concluded that the effect of the Act is that the alleged previous sexual relationship is inadmissible on the issue of consent. On this supposition Rose LJ further stated that the Crown accepted that the trial judge will, in due course, have to direct the jury that the evidence of the complainant's consensual activity with the defendant during the period before the alleged rape is solely relevant to the question of the defendant's belief as to consent and is not relevant to the question of whether the complainant in fact consented. However, Rose LJ was of the view that such a direction might lead to an unfair trial because a previous sexual relationship may be relevant to the issue of consent as well as belief in consent.
24. Allowing the appeal the Court of Appeal observed:
On 31 January 2001 the Court of Appeal certified the following question:
At the same time the Court of Appeal granted the Crown leave to appeal to the House of Lords.
III. The Secretary of State's intervention
25. Counsel for the defendant indicated that on the appeal to the House he would invite the House to read down section 41 of the 1999 Act in accordance with section 3 of the Human Rights Act 1998 so that section 41 could be given effect in a way that was compatible with the fair trial guarantee under article 6 of the Convention, and if that was not possible, he would invite the House to make a declaration of incompatibility. In these circumstances the Secretary of State for the Home Department applied for leave to intervene at this stage. An Appeal Committee recommended that leave be given to the Secretary of State to intervene. In its 31st Report of 7 March 2001 drafted by Lord Hope of Craighead the Appeal Committee observed  1 WLR 789, 792-793:
In the result the House has had the advantage of submissions not only on behalf of both the Director of Public Prosecutions and the defendant but also on behalf of the Secretary of State. On the hearing of the appeal counsel for the Secretary of State referred to parts of the preceding Parliamentary debates but he made clear that he was not doing so as an aid to construction of the statutory language under the rule in Pepper v Hart  AC 593. Instead he used this material, together with other materials, to identify the mischief which led to the enactment of the statute.
26. On the hearing of the appeal counsel for the Director of Public Prosecutions informed the House that the same issue arises in 13 other criminal cases. It is therefore a matter of some urgency.
IV. The context of section 41
27. Following the Second World War the general principle of the equality of men and women in all spheres of life has gradually become established. In the aftermath of the sexual revolution of the sixties the autonomy and independence of women in sexual matters has become an accepted norm. It was this change in thinking about women and sex which made possible the decision of the House of Lords in R v R  1 AC 599 that the offence of rape may be committed by a husband upon his wife. It was a dramatic reversal of old fashioned beliefs. Discriminatory stereotypes which depict women as sexually available have been exposed as an affront to their fundamental rights. Nevertheless, it has to be acknowledged that in the criminal courts of our country, as in others, outmoded beliefs about women and sexual matters lingered on. In recent Canadian jurisprudence they have been described as the discredited twin myths, viz "that unchaste women were more likely to consent to intercourse and in any event, were less worthy of belief": R v Seaboyer (1991) 83 DLR (4th) 193, 258, 278C per McLachlin J. Such generalised, stereotyped and unfounded prejudices ought to have no place in our legal system. But even in the very recent past such defensive strategies were habitually employed. It resulted in an absurdly low conviction rate in rape cases. It also inflicted unacceptable humiliation on complainants in rape cases.
28. In Director of Public Prosecutions v Morgan  AC 182 the House of Lords held that in a trial for rape a subjective belief by the defendant that the victim consented to sexual intercourse afforded a defence. Following this decision the Advisory Group on the Law of Rape was established. It produced the so-called Heilbron Report (1975) (Cmnd 6352). It treated previous sexual association between the complainant and the accused as potentially relevant but advised that in general the previous sexual history of the complainant with other men was irrelevant. Parliament enacted legislation which subjected the admission of evidence of the previous sexual experience of a complainant with third parties to a leave requirement. It did not touch on prior sexual contact between the complainant and the accused: section 2(1) of the Sexual Offences (Amendment) Act 1976. Section 2(2) provides that the judge shall only give leave "if and only if he is satisfied that it would be unfair to that defendant to refuse to allow the evidence to be adduced or the question to be asked." The statute did not achieve its object of preventing the illegitimate use of prior sexual experience in rape trials. In retrospect one can now see that the structure of this legislation was flawed. In respect of sexual experience between a complainant and other men, which can only in the rarest cases have any relevance, it created too broad an inclusionary discretion. Moreover, it left wholly unregulated questioning or evidence about previous sexual experience between the complainant and the defendant even if remote in time and context. There was a serious mischief to be corrected.
V. Section 41
29. Sections 41 to 43 of the 1999 Act imposed wide restrictions on evidence and questioning about a complainant's sexual history. These provisions are contained in Chapter III of Part II of the statute and appear under the heading "Protection of Complainants in Proceedings for Sexual Offences". The material part of section 41 reads:
Section 41 imposes the same exclusionary provisions in respect of a complainant's sexual experience with the accused as with other men. This is the genesis of the problem before the House. There are differences which need to be explored. In this task I have been greatly assisted primarily by the careful and incisive arguments of counsel but also by an as yet unpublished comprehensive review of the literature, comparative jurisprudence, and different legislative models and proposals for reform prepared by Neil Kibble of the Department of Law, University of Wales Aberystwyth "The Admissibility of Prior Sexual History with the Defendant in Sexual Offence Cases" (February 2001). My understanding is that in revised form it will be published in the Cambrian Law Review. It amplifies his earlier paper "The Sexual History Provisions, Charting a course between inflexible legislative rules and wholly untrammelled judicial discretion"  Crim LR 274.
VI. Sexual experience with the accused contrasted with sexual experience with other men.
30. Although not an issue before the House, my view is that the 1999 Act deals sensibly and fairly with questioning and evidence about the complainant's sexual experience with other men. Such matters are almost always irrelevant to the issue whether the complainant consented to sexual intercourse on the occasion alleged in the indictment or to her credibility. To that extent the scope of the reform of the law by the 1999 Act was justified. On the other hand, the blanket exclusion of prior sexual history between the complainant and an accused in section 41(1), subject to narrow categories of exception in the remainder of section 41, poses an acute problem of proportionality.
31. As a matter of common sense, a prior sexual relationship between the complainant and the accused may, depending on the circumstances, be relevant to the issue of consent. It is a species of prospectant evidence which may throw light on the complainant's state of mind. It cannot, of course, prove that she consented on the occasion in question. Relevance and sufficiency of proof are different things. The fact that the accused a week before an alleged murder threatened to kill the deceased does not prove an intent to kill on the day in question. But it is logically relevant to that issue. After all, to be relevant the evidence need merely have some tendency in logic and common sense to advance the proposition in issue. It is true that each decision to engage in sexual activity is always made afresh. On the other hand, the mind does not usually blot out all memories. What one has been engaged on in the past may influence what choice one makes on a future occasion. Accordingly, a prior relationship between a complainant and an accused may sometimes be relevant to what decision was made on a particular occasion.
32. In a balanced review of the voluminous critical literature in the United Kingdom between 1975 and 1999 Mr Kibble has shown that the principal focus throughout has been on the irrelevance and prejudicial impact of sexual experience of the complainant with other men. The target of the literature was the 1976 Act. When the issue of the relevance of sexual experience between a complainant and a defendant was raised there was broad agreement that such evidence is sometimes relevant (e.g. an ongoing relationship) and sometimes irrelevant (e.g. an isolated episode in the past). There was no case made out in the literature for the blanket exclusionary scheme incorporated in section 41 in respect of prior sexual experience between a complainant and accused. Not surprisingly the legislative technique adopted in section 41 has been criticised. Professor Diane Birch ("A Better Deal for Vulnerable Witnesses?"  Crim LR 223, 248), trenchantly commented:
It is difficult to dispute this assessment. After all, good sense suggests that it may be relevant to an issue of consent whether the complainant and the accused were ongoing lovers or strangers. To exclude such material creates the risk of disembodying the case before the jury. It also increases the danger of miscarriages of justice. These considerations raise the spectre of the possible need for a declaration of incompatibility in respect of section 41 under section 4 of the Human Rights Act 1998.
33. Counsel for the Secretary of State submitted that section 41 was based on the decision of the Supreme Court of Canada in R v Seaboyer 83 DLR (4th) 193. In that case a first attempt to introduce "rape-shield" provisions directed against the admissibility of sexual history evidence in rape cases was held to be invalid under section 7 of the Charter of Rights and Freedoms. By a majority the Supreme Court indicated what kind of provisions would be lawful. Following R v Seaboyer section 276 of the Criminal Code was amended. Subsequently the Supreme Court held that section 276 as amended was valid. As amended it was not viewed as a blanket exclusion: R v Darrach (2000) 191 DLR (4th) 539. Unfortunately, the Secretary of State's understanding of the Canadian position was flawed. R v Seaboyer is largely concerned with the irrelevance of sexual experience between the complainant and third parties. In her leading judgment McLachlin J placed general reliance upon an article of Galvin, who emphasises the probative value of prior sexual conduct between a complainant and an accused to the issue of consent: "Shielding Rape Victims in the State and Federal Courts: A Proposal for the Second Decade" (1986) 70 Minn LRev 763. Moreover, McLachlin J made a telling comment on prior sexual history with the accused. It is to the following effect, at 83 DLR (4th) 193, 280D:
R v Seaboyer does not justify the breadth of the exclusionary provisions of section 41 in respect of previous sexual experience between a complainant and a defendant. The amended section 276 of the Canadian statute is also in more flexible terms than section 41. Section 276 reads:
It will be observed that subsection (1) is directed at impermissible uses of the evidence. It is not a blanket prohibition. It has an inbuilt flexibility as appears from the balancing provision of subsection (2) and particularly the words of paragraph (c). The Canadian model is therefore in substantially less restrictive terms than section 41. Moreover, it is noteworthy that a law reform proposal in New South Wales explicitly accepts that the fact that the complainant engaged in sexual activity with the accused in the past may be relevant to the question whether she consented to sexual activity on the occasion in question: New South Wales Law Reform Commission Report (1998) (No 87) on section 409B of the Crimes Act 1900. A similar flexible approach is reflected in a discussion paper of the New Zealand Law Commission: "Evidence Law: Characters and Credibility" (1997) (Preliminary Paper 27) published in February 1997. Commonwealth developments do not support the breadth of the exclusionary provisions of section 41 in respect of the potential relevance of the sexual experience of a complainant with an accused.
VII. The interpretation of section 41
34. In order to assess whether section 41 is incompatible with the convention right to a fair trial, it is necessary to consider what evidence it excludes. The mere fact that it excludes some relevant evidence would not by itself amount to a breach of the fair trial guarantee. On the other hand, if the impact of section 41 is to deny the right to accused in a significant range of cases from putting forward full and complete defences it may amount to a breach.
35. Counsel for the Secretary of State has argued that unfairness to an accused will rarely arise because evidence of sexual experience between a complainant and an accused will almost always be admissible on the basis of the defence that the accused thought that the complainant consented. His argument has assumed that in practice an accused will almost invariably be able to put forward both defences. Counsel for the defendant has persuaded me that the defence of belief in consent would often have no air of reality and would in practice not be available, eg in cases where there are diametrically opposite accounts of the circumstances of the alleged rape, with the complainant insisting that it was perpetrated with great violence and the accused saying that the complainant took the initiative in an act of consensual intercourse. In any event, it does not meet the difficulty that the judge's direction to the jury would always have to be to the effect that the past experience between the complainant and the accused is irrelevant to the issue of consent. I would reject the submissions of counsel for the Secretary of State on this point. In these circumstances counsel for the Secretary of State accepts that, despite the interlocutory nature of the proceedings, the House must now grapple with the problem whether, measured against the guarantee of a fair trial, the breadth of the exclusionary provisions of section 41 in respect of sexual experience between a complainant and the defendant are justified and proportionate. The position of counsel for the Secretary of State on this point is realistic. To postpone the decision until after the conclusion of a number of pending trials, which raise the issue, would be unfair to individuals and contrary to the public interest.
36. Counsel for the Secretary of State further relied on the principle that, in certain contexts, the legislature and the executive retain a discretionary area of judgment within which policy choices may legitimately be made: see Brown v Stott  2 WLR 817. Clearly the House must give weight to the decision of Parliament that the mischief encapsulated in the twin myths must be corrected. On the other hand, when the question arises whether in the criminal statute in question Parliament adopted a legislative scheme which makes an excessive inroad into the right to a fair trial the court is qualified to make its own judgment and must do so.
37. The methodology to be adopted is important. In a helpful paper under the title "The Act of the Possible: Interpreting Statutes under the Human Rights Act"  EHRLR 665 Lord Lester of Herne Hill QC has summarised the correct approach, at p 674:
See also Bertha Wilson J, "The Making of a Constitution: Approaches to Judicial Interpretation" (1988) PL 370, 371-372; and David Feldman, "Proportionality and The Human Rights Act 1998" in The Principle of Proportionality in the Laws of Europe (1999), pp117, 122-123.
38. It is well established that the guarantee of a fair trial under article 6 is absolute: a conviction obtained in breach of it cannot stand. R v Forbes,  2 WLR 1, 13, para 24. The only balancing permitted is in respect of what the concept of a fair trial entails: here account may be taken of the familiar triangulation of interests of the accused, the victim and society. In this context proportionality has a role to play. The criteria for determining the test of proportionality have been analysed in similar terms in the case law of the European Court of Justice and the European Court of Human Rights. It is not necessary for us to re-invent the wheel. In de Freitas v Permanent Secretary of Ministry of Agriculture, Fisheries, Lands and Housing  1 AC 69 Lord Clyde adopted a precise and concrete analysis of the criteria. In determining whether a limitation is arbitrary or excessive a court should ask itself:
The critical matter is the third criterion. Given the centrality of the right of a fair trial in the scheme of the Convention, and giving due weight to the important legislative goal of countering the twin myths, the question is whether section 41 makes an excessive inroad into the guarantee of a fair trial.
39. Subject to narrow exceptions section 41 is a blanket exclusion of potentially relevant evidence. Section 41 must however be construed in order to determine its precise exclusionary impact on alleged previous sexual experience between the complainant and the accused. Two processes of interpretation must be distinguished. First, ordinary methods of purposive and contextual interpretation may yield ways of minimising the prima facie exorbitant breadth of the section. Secondly, the interpretative obligation in section 3(1) of the 1998 Act may come into play. It provides that "so far as it is possible to do so, primary legislation . . . must be read and given effect in a way which is compatible with the Convention rights". It is a key feature of the 1998 Act.
40. Three possible ways of minimising the excessive breadth of section 41 must be considered. The first possible gateway is to be found in section 41(3)(b), viz:
An example covered by this provision would be where it is alleged that the complainant invited the accused to have sexual intercourse with her earlier in the evening. In my opinion, however, neither ordinary methods of interpretation nor the interpretative obligation under section 3 of the 1998 Act enables one to extend the temporal restriction to days, weeks or months. Section 41(3)(b) acknowledges by its own terms that previous sexual experience between a complainant and an accused may be relevant but then restricts the admission of such evidence by an extraordinarily narrow temporal restriction.
41. The second gateway suggested by counsel for the Director of Public Prosecutions is the provision in section 41(5)(b) enabling evidence adduced by the prosecution to be rebutted or explained by or on behalf of the defence. The suggestion is that the Crown could adduce evidence which will enable the defence to lead evidence of previous sexual experience in rebuttal. This is not a coherent and satisfactory solution. It depends on the goodwill and co-operation of the prosecutor. A defendant has the right in a criminal trial to offer a full and complete defence. I would reject this suggested solution.
42. The third gateway is section 41(3)(c). It permits evidence where
This gateway is only available where the issue is whether the complainant consented and the evidence or questioning relates to behaviour that is so similar to the defence's version of the complainant's behaviour at the time of the alleged offence that it cannot reasonably be explained as a coincidence. An example would be the case where the complainant says that the accused raped her; the accused says that the complainant consented and then after the act of intercourse tried to blackmail him by alleging rape; and the defence now wishes to ask the complainant whether on a previous occasion she similarly tried to blackmail the accused.
43. Rightly none of the counsel appearing before the House were prepared to argue that on ordinary methods of interpretation section 41(3)(c) can be interpreted to cover, for example, cases similar to the one before the House where it is alleged that there was a previous sexual experience between the complainant and the accused on several occasions during a three week period before the occasion in question. Let me consider ordinary methods of interpretation in a little more detail. One could say that section 41(3)(c) is a statutory adoption of the striking similarity test enunciated in R v Boardman  AC 421. So interpreted section 41(3)(c) is a narrow gateway, which will only be available in rare cases. Alternatively, one could argue that section 41(3)(c) involves the test of high probative force of the evidence, which makes it just to admit it, in accordance with the principle stated in Director of Public Prosecutions v P  2 AC 447. Even if this approach was consistent with the language of section 41, the threshold requirement would be too high: often the evidence will be relevant but not capable of being described as having "high probative value". These ways of interpreting section 41(3)(c) cannot solve the problem of the prima facie excessive inroad on the right to a fair trial. It is important to concentrate in the first place on the language of section 41. Making due allowance for the words "in any respect" in section 41(3)(c), the test "that the similarity cannot reasonably be explained as a coincidence" is inapt to allow evidence to be admitted or questioning to take place that, for example, (i) the complainant invited the accused at an office party on a Friday to come to her flat on the Sunday to make love to her or (2) that the complainant and the accused had sexual relations on several occasions in the previous month. While common sense may rebel against the idea that such evidence is never relevant to the issue of consent, that is the effect of the statute. In my view ordinary methods of purposive construction of section 41(3)(c) cannot cure the problem of the excessive breadth of the section 41, read as a whole, so far as it relates to previous sexual experience between a complainant and the accused. Whilst the statute pursued desirable goals, the methods adopted amounted to legislative overkill.
44. On the other hand, the interpretative obligation under section 3 of the 1998 Act is a strong one. It applies even if there is no ambiguity in the language in the sense of the language being capable of two different meanings. It is an emphatic adjuration by the legislature: R v Director of Public Prosecutions, Ex p Kebilene  2 AC 326, per Lord Cooke of Thorndon, at p 373F; and my judgment, at p 366B. The White Paper made clear that the obligation goes far beyond the rule which enabled the courts to take the Convention into account in resolving any ambiguity in a legislative provision: see "Rights Brought Home: The Human Rights Bill" (1997) (Cm 3782), para 2.7. The draftsman of the Act had before him the slightly weaker model in section 6 of the New Zealand Bill of Rights Act 1990 but preferred stronger language. Parliament specifically rejected the legislative model of requiring a reasonable interpretation. Section 3 places a duty on the court to strive to find a possible interpretation compatible with Convention rights. Under ordinary methods of interpretation a court may depart from the language of the statute to avoid absurd consequences: section 3 goes much further. Undoubtedly, a court must always look for a contextual and purposive interpretation: section 3 is more radical in its effect. It is a general principle of the interpretation of legal instruments that the text is the primary source of interpretation: other sources are subordinate to it: compare, for example, articles 31 to 33 of the Vienna Convention on the Law of Treaties (1980) (Cmnd 7964). Section 3 qualifies this general principle because it requires a court to find an interpretation compatible with Convention rights if it is possible to do so. In the progress of the Bill through Parliament the Lord Chancellor observed that "in 99% of the cases that will arise, there will be no need for judicial declarations of incompatibility" and the Home Secretary said "We expect that, in almost all cases, the courts will be able to interpret the legislation compatibility with the Convention": Hansard (HL Debates), 5 February 1998, col 840 (3rd Reading) and Hansard (HC Debates), 16 February 1998, col 778 (2nd Reading). For reasons which I explained in a recent paper, this is at least relevant as an aid to the interpretation of section 3 against the executive: "Pepper v Hart: A re-examination" (2001) 21 Oxford Journal of Legal Studies 59. In accordance with the will of Parliament as reflected in section 3 it will sometimes be necessary to adopt an interpretation which linguistically may appear strained. The techniques to be used will not only involve the reading down of express language in a statute but also the implication of provisions. A declaration of incompatibility is a measure of last resort. It must be avoided unless it is plainly impossible to do so. If a clear limitation on Convention rights is stated in terms, such an impossibility will arise: R v Secretary of State for the Home Department, Ex p Simms  2 AC 115, 132A-B per Lord Hoffmann. There is, however, no limitation of such a nature in the present case.
45. In my view section 3 requires the court to subordinate the niceties of the language of section 41(3)(c), and in particular the touchstone of coincidence, to broader considerations of relevance judged by logical and common sense criteria of time and circumstances. After all, it is realistic to proceed on the basis that the legislature would not, if alerted to the problem, have wished to deny the right to an accused to put forward a full and complete defence by advancing truly probative material. It is therefore possible under section 3 to read section 41, and in particular section 41(3)(c), as subject to the implied provision that evidence or questioning which is required to ensure a fair trial under article 6 of the Convention should not be treated as inadmissible. The result of such a reading would be that sometimes logically relevant sexual experiences between a complainant and an accused may be admitted under section 41(3)(c). On the other hand, there will be cases where previous sexual experience between a complainant and an accused will be irrelevant, eg an isolated episode distant in time and circumstances. Where the line is to be drawn must be left to the judgment of trial judges. On this basis a declaration of incompatibility can be avoided. If this approach is adopted, section 41 will have achieved a major part of its objective but its excessive reach will have been attenuated in accordance with the will of Parliament as reflected in section 3 of the 1998 Act. That is the approach which I would adopt.VIII.
The task of trial judges
46. It is of supreme importance that the effect of the speeches today should be clear to trial judges who have to deal with problems of the admissibility of questioning and evidence on alleged prior sexual experience between an accused and a complainant. The effect of the decision today is that under section 41(3)(c) of the 1999 Act, construed where necessary by applying the interpretative obligation under section 3 of the Human Rights Act 1998, and due regard always being paid to the importance of seeking to protect the complainant from indignity and from humiliating questions, the test of admissibility is whether the evidence (and questioning in relation to it) is nevertheless so relevant to the issue of consent that to exclude it would endanger the fairness of the trial under article 6 of the convention. If this test is satisfied the evidence should not be excluded.
IX. Application of the interpretation adopted.
47. The appeal before the House concerns a concrete case. It involves the permissibility of questioning a complainant about an alleged recent sexual relationship between her and the defendant, and the admissibility of evidence on that point. These are matters for the trial judge to rule on at the resumed trial. But in my view he must do so on the broader interpretation of section 41(3)(c) required by section 3 of the 1998 Act.
48. I would decline to make the rulings sought by the Director of Public Prosecutions and the Secretary of State. Given the terms of this speech it is unnecessary to answer the certified question. I would dismiss the appeal.
LORD HOPE OF CRAIGHEAD
49. Rape is the most humiliating, distressing and cynical of crimes. It presents itself in various ways to the prosecutor. Sometimes it is accompanied by acts of extreme violence. In such cases proof that the crime has been perpetrated will be little more than a formality and the more difficult task is likely to be to prove the identity of the perpetrator. But more often than not very little, if any, violence is used, identity is not in issue as the parties were known to each other and the defendant admits that on the occasion in question he had sexual intercourse. The sole issue for the prosecutor in these cases will be whether it can be proved that the complainant did not consent to the sexual intercourse. The crime is constituted by proof of the fact of sexual intercourse with a person who at the time of the intercourse did not consent to it, accompanied by proof that at the time the defendant either knew that the person did not consent to the intercourse or was reckless as to whether that person consented to it: Sexual Offences Act 1956, as substituted by section 142 of the Criminal Justice and Public Order Act 1994. The absence of consent is, in these cases, the crucial issue. This is a question of fact, which must be resolved in the light of the evidence.
50. It is notorious that proof that the complainant did not consent to an admitted act of sexual intercourse raises difficult questions which, in the typical case, resolve themselves into issues of credibility. In its modern form the definition of the crime recognises that every woman has the right, on each and every occasion, to say "no". As Gonthier J put it in R v Darrach (2000) 191 DLR (4th) 539, 568, actual consent must be given for each instance of sexual activity. The crime has now been extended to the rape of a man by another man: Sexual Offences Act 1956, section 1(1) as substituted by section 142 of the Criminal Justice and Public Order Act 1994. So every man also has that right. But it is one thing for the law to recognise these essential facts. It is quite another for the law to put its principles into practice. That, in the final analysis, is what this case is about.
51. It is plain a balance must be struck between the right of the defendant to a fair trial and the right of the complainant not to be subjected to unnecessary humiliation and distress when giving evidence. The right of the defendant to a fair trial has now been reinforced by the incorporation into our law of article 6 of the European Convention for the Protection of Human Rights and Fundamental Freedoms by the Human Rights Act 1998. But the principles which are enshrined in that article have for long been part of our common law. The common law recognises that a defendant has the right to cross-examine the prosecutor's witnesses and to give and lead evidence. The guiding principle as to the extent of that right is that prima facie all evidence which is relevant to the question whether the defendant is guilty or innocent is admissible. As the fact that the act of sexual intercourse was without the consent of the complainant is one of the essential elements in the charge which the prosecutor must establish, the defendant must be given an opportunity to cross-examine the prosecutor's witnesses and to give and lead evidence on that issue. That is an essential element of his right to a fair trial.
52. But the extent to which a defendant may go in the exercise of his right to be given that opportunity is a matter to which the common law has failed to provide a satisfactory answer. The problem is at its most acute in cases where the parties to the alleged rape are known to each other and have had some kind of a relationship in the past. In their joint written intervention the Rape Crisis Federation of England and Wales, the Campaign to End Rape, the Child and Woman Abuse Studies Unit and Justice for Women state that the evidence is that this is the most frequent type of rape, the least likely to be reported to the police and, when proceedings are brought, the least likely to result in a conviction. The statistics to which they refer bear out this statement.
53. K Painter "Wife Rape, Marriage and Law: Survey Report, Key Findings and Recommendations" (Manchester University, 1991), reporting on a sample of 1007 women in 11 cities, stated that 1 in 4 of those interviewed said that they had been the victims of rape or attempted rape, that the most common perpetrators were current and ex partners and that 91 per cent of those interviewed had told no-one. Home Office Statistics quoted in Speaking Up for Justice, Report of the Interdepartmental Working Group on the Treatment of Vulnerable or Intimidated Witnesses in the Criminal Justice System (Home Office, June 1998) indicated that, while in 1985 35 per cent of reported rapes occurred within an intimate relationship and 30 per cent were by strangers, by 1997 these percentages had altered to 43 per cent and 12 per cent respectively. On the other hand the conviction rate for rape had decreased markedly over the same period. In 1985 24 per cent of rapes reported to the police resulted in a conviction. By 1996 the number of rape complaints to the police had trebled but the conviction rate had fallen to 9 per cent. Unpublished research for the Home Office in 1997 concluded that there was a link between the increased number of complaints involving intimates and former intimates and the decrease in the conviction rate: "The Processing of Rape Cases by the Criminal Justice System" (1997) (Jessica Harris).
54. To a substantial extent these studies may be thought to confirm what is already obvious. In an as yet unpublished paper "The Admissibility of Prior Sexual History with the Defendant in Sexual Offence Cases" (University of Wales Aberystwyth, February 2001) in which he conducted a review of the critical and reform literature on this subject in the United Kingdom between 1975 and 1999, Neil Kibble observed, at p 23, that the literature was concerned almost exclusively with the problems surrounding the admissibility of prior sexual history with third parties and that little systematic attention had been paid to the question of the relevance and admissibility of prior sexual history with the accused. But it is well known that women in general are deterred from making complaints that they have been raped by a person with whom they have or previously had a relationship. It is distressing enough for women to have to give evidence in these cases. They are unwilling to face the prospect of being further humiliated by questions directed to their previous or subsequent sexual history. The low conviction rate acts as a further deterrent. The humiliation for the woman is much increased if no conviction results after she has been subjected to that kind of questioning.
55. These and studies undertaken in other countries, many of which were referred to by L'Heureux-Dubé and Gonthier JJ in their partial dissent in R v Seaboyer  2 SCR 577, indicate that the balance between the rights of the defendant and those of the complainant is in need of adjustment if women are to be given the protection under the law to which they are entitled against conduct which the law says is criminal conduct. As McLachlin J said, at p 609B-E, in the judgment which she delivered on behalf of the majority in that case, it is fundamental to our system of justice that the rules of evidence should permit the judge and jury to get at the truth and properly determine the issues in the case. A law which prevents the trier of fact from getting at the truth by excluding relevant evidence runs counter to our fundamental conceptions of justice and what constitutes a fair trial. But there is a risk that juries may be diverted from the real issues in the trial by evidence about the complainant's sexual behaviour which is not directly relevant to the offence charged: R v Seaboyer  2 SCR 577, 634A-D; R v Darrach 191 DLR (4th) 539, 560-561, Kibble, p 41. A balance must be struck between the probative value of the evidence and its potential prejudice.
56. Section 41 of the Youth Justice and Criminal Evidence Act 1999 has been designed to achieve that adjustment. It is clear from the background against which that section was enacted and from its own terms that this is the mischief which it was intended to address. It is also clear from what has been happening in other jurisdictions where similar provisions have been introduced that there was a choice to be made as to how far the balance should be adjusted in favour of the public interest while preserving the right to a fair trial. A wide variety of measures to which I shall refer later, commonly known as "rape-shield" provisions, have been enacted to restrict the right of a defendant who is on trial for a sexual offence to cross-examine and lead evidence of the complainant's sexual conduct on other occasions.
57. Section 2 of the Sexual Offences (Amendment) Act 1976 left this matter to the discretion of the trial judge. The original Bill had contained complicated provisions which were designed to restrict the admissibility of such evidence, but these were removed and replaced by a general test of unfairness to the defendant. Section 2(2) of the Act provided that the judge should give leave if, and only if, he was satisfied that it would be unfair to the defendant to refuse to allow the evidence to be adduced or the question to be asked. But the statistics showed that the object of that measure, which was to protect complainants against unnecessary evidence and questions about their previous sexual experience, was not being achieved. They raised doubts as to whether it was satisfactory, in this very difficult and sensitive area, to leave the decision whether leave should be given entirely to the trial judge. The question which has been raised in this case is whether the new legislation, which greatly restricts the discretion given to the trial judge, is compatible with the defendant's Convention right to a fair trial.
58. I would take, as my starting point for examining section 41, the proposition that there are areas of law which lie within the discretionary area of judgment which the court ought to accord to the legislature. As I said in R v Director of Public Prosecutions, Ex p Kebilene  2 AC 326, 380-381E, it is appropriate in some circumstances for the judiciary to defer, on democratic grounds, to the considered opinion of the elected body as to where the balance is to be struck between the rights of the individual and the needs of society: see also Brown v Stott  2 WLR 817 per Lord Bingham of Cornhill, at p 835A-B, and Lord Steyn at p 842F-G. I would hold that prima facie the circumstances in which section 41 was enacted bring this case into that category. As I shall explain in more detail later (see paragraph 90, post), the right to lead evidence and the right to put questions with which that section deals are not among the rights which are set out in unqualified terms in article 6 of the Convention. They are open to modification or restriction so long as this is not incompatible with the right to a fair trial. The essential question for your Lordships, as I see it, is whether Parliament acted within its discretionary area of judgment when it was choosing the point of balance that is indicated by the ordinary meaning of the words used in section 41. If it did not, questions will arise as to whether the incompatibility that results can be avoided by making use of the rule of interpretation in section 3 of the Human Rights Act 1998, failing which whether a declaration of incompatibility should be made. But I think that the question which I have described as the essential question must be addressed first. As Lord Woolf CJ said in Poplar Housing and Regeneration Community Association Ltd v Donaghue  EWCA Civ 595, para 75, unless the legislation would otherwise be in breach of the Convention section 3 of the 1998 Act can be ignored. So the courts should always ascertain first whether, absent section 3, there would be any breach of the Convention.
59. I shall need to look at section 41 in more detail. But I must first set out briefly the facts which have given rise in this case to the question whether a sexual relationship between a defendant and a complainant may be relevant to the issue of consent so as to render its exclusion by that section a contravention of the defendant's right to a fair trial. The facts are important, because it is to the facts of the particular case as alleged by the defendant that any issues about any possible incompatibility with his Convention right to a fair trial must be directed.
60. The incident in which the respondent is alleged to have raped the complainant took place in the early hours of 14 June 2000 beside the river Thames as they were walking along the towpath. As they walked along the path the respondent fell down. The complainant states that when she tried to help him to his feet he pulled her to the ground and had sexual intercourse with her without her consent. The respondent's case is that on the occasion in question the complainant initiated consensual intercourse. He states that this was part of a consensual sexual relationship which covered a period of about three weeks prior to 14 June 2000 during which he had sexual relations with her, including sexual intercourse, in his flat on various occasions. The last of these was about one week before the alleged rape. In short, the respondent's case is that he did not rape the complainant because she consented to the act of intercourse. He seeks leave to cross-examine her and lead evidence about their previous relationship to support his defence that this was an act of consensual intercourse. No doubt any cross-examination which is directed to that relationship will tend to undermine her credibility on this vital issue.
61. The history is complicated by the fact that the respondent was sharing his flat with another man with whom the complainant is said to have formed an intimate relationship. It is said that she used to visit the other man at the flat, and that at about 9 pm on 13 June 2000 she had sexual intercourse with him in the flat when the respondent was not there. On the occasion of the alleged rape the respondent and the complainant were walking from the flat to a hospital where the other man had been taken after collapsing on his return from a picnic with the complainant that evening close to the river bank.
62. The case came before the trial judge for a preliminary hearing under sections 29 to 31 of the Criminal Procedure and Investigations Act 1996 on 8 December 2000. He was asked to rule on the extent to which cross-examination of the complainant would be permitted in the light of section 41 of the 1999 Act. He held that the complainant could be cross-examined about the act of sexual intercourse which took place between her and the other man a few hours before the occasion of the alleged rape and about any other sexual activity she may have had with other men at or about the same time under section 41(3)(b). But he said that cross-examination about her previous relationship with the respondent would not be permitted to any extent under either section 41(3)(b) or section 41(3)(c).
63. The respondent appealed against this ruling to the Court of Appeal (Criminal Division) under section 35(1) of the Criminal Procedure and Investigations Act 1996 with the leave of the trial judge. At the hearing of the appeal the respondent's counsel, Mr Rook QC, who did not appear in the court below, raised for the first time the question whether cross-examination and evidence directed to the complainant's prior sexual activity with the respondent would be admissible under section 41 in relation to a further defence that he honestly believed that she was consenting to intercourse: see Director of Public Prosecutions v Morgan  AC 182. The Crown conceded that cross-examination and evidence directed to this issue would be admissible under section 41(3)(a) of the Act: para 34 of the Court of Appeal's judgment. For his part, Mr Rook did not seek to suggest that there was such a degree of similarity in the present case as would permit questions to be asked under section 41(3)(c): para 19.
64. As to the merits of the appeal, the Court of Appeal  EWCA Crim 4 held that the judge was right to conclude that cross-examination and evidence about the complainant's recent consensual sexual activity with the respondent would not be admissible under section 41(3)(b): para 34. But the court held that the judge was wrong in saying that questions and evidence about the complainant's sexual behaviour with the respondent's friend or with other third parties on the night of the alleged rape would be admissible under section 41(3)(b) because the court was of the opinion that this material was not relevant: paragraph 35. The appeal was allowed however on the ground that the judge was wrong to hold that evidence and questions about the complainant's sexual behaviour with the respondent was inadmissible. This was because, as the Crown conceded, that evidence was permissible under section 41(3)(a) in relation to the defence of belief as to consent: para 35.
65. However, Rose LJ said that the court respectfully differed from the view that previous recent consensual intercourse between the complainant and the defendant was irrelevant to whether she consented on the occasion said to give rise to rape, and that it might be that a fair trial would not be possible if there could not be adduced, in support of the defence of consent, evidence as to the complainant's recent consensual activity with the defendant: paras 30, 33. It is those observations that have led to this appeal by the Crown, for which the Court of Appeal granted leave. On 7 March 2001 the House gave leave to the Secretary of State for the Home Department to be joined as a party to the appeal for the reasons given in the 31st Report from the Appeal Committee of that date:  1 WLR 789.
66. Your Lordships are not being asked in this case to reconsider the decision in Director of Public Prosecutions v Morgan  AC 182. The proper limits of that defence currently under examination by the government following an independent review, the results of which are set out in a consultation paper entitled "Setting the Boundaries, Reforming the Law on Sex Offences" (Home Office, July 2000), para 2.13. For the time being it may be noted, as it was pointed out in Jamieson v H M Advocate 1994 SLT 537, 541J-K by the High Court of Justiciary, that as the law stands difficult questions of fact may arise as to whether, if he can give no reasonable grounds for his belief, the accused genuinely believed at the time that the woman was consenting or was reckless or indifferent as to the matter of consent. For the purposes of this case it must be assumed that cross-examination and evidence which is directed to that issue will be permitted at the trial, in accordance with the concession by the Crown, under reference to section 41(3)(a) of the 1999 Act. The issue as to the respondent's honest belief that the complainant was consenting to intercourse is not an issue of consent.
67. All that needs to be said about this part of the respondent's case is that the extent to which the complainant may be cross-examined about her previous relationship with the respondent, and the extent to which the respondent may give evidence about it, for the purposes of the defence of honest belief will be subject at all times to control by the court under section 41(2)(b). The court has an overriding duty under that paragraph to ensure that any evidence or question for which leave is given is permitted only to the extent that to refuse leave would render a conclusion on any relevant issue in the case unsafe.
68. But the facts which the respondent wishes to elicit by cross-examination and to adduce in evidence in support of the defence of consent bring into sharp focus the following questions: (a) whether the questions and evidence will be admissible under section 41 when that section is construed according to ordinary common law principles and (b) if not, whether to exclude them would be compatible with his Convention right to a fair trial. If both of these questions are answered in the negative, two further questions will arise. The first is whether the critical parts of section 41 can be given a different meaning by using the techniques of statutory interpretation indicated by section 3 of the Human Rights Act 1998, which requires that the legislation must be read and given effect to, so far as it is possible to do so, in a way which is compatible with the respondent's Convention right. If that cannot be done, consideration will have to be given to the question whether to make a declaration of incompatibility under section 4 of the 1998 Act.
69. It may be noted in passing that a statement of compatibility was attached to the Bill before second reading that its provisions were compatible with the Human Rights Act 1998. Statements to that effect are now required by section 19 of the Act, which was brought into force on 24 November 1998: (SI 1998/2882). But Mr Pannick QC for the Secretary of State did not seek to rely on this statement in the course of his argument. I consider that he was right not to do so. These statements may serve a useful purpose in Parliament. They may also be seen as part of the parliamentary history, indicating that it was not Parliament's intention to cut across a Convention right: Lord Irvine of Lairg LC "The Development of Human Rights in Britain under an Unincorporated Convention on Human Rights" (1998) PL 221, 228. No doubt they are based on the best advice that is available. But they are no more than expressions of opinion by the minister. They are not binding on the court, nor do they have any persuasive authority.
The ordinary meaning of section 41
70. I propose in this section to examine in detail only those provisions of section 41 that are directly in issue in this case. It is not possible in this case to solve all the problems that may arise. But it may be helpful for me to state what I understand to be its basic structure.
71. Section 41 of the 1999 Act contains the following essential elements:(a)
it applies to any trial at which a person is charged with a sexual offence (see subsection (1) which extends, among other things, to a wide range of sexual offences involving children as well as those involving women who complain that they have been raped);(b)
it contains a general prohibition against the adducing by the accused of evidence or his asking of questions in cross-examination about any sexual behaviour of the complainant except with the leave of the court (see subsection (1), which is to be read with the definition of "sexual behaviour" in section 42(1)(c));(c)
it provides a requirement that leave be given only on an application made by or on behalf of the accused (see subsection (2), as to which section 43 lays down the procedure);(d)
it places a duty on the court to grant leave only if it is satisfied that the evidence or question falls within one or other of the two qualifying subsections (see subsection (2)(a), and subsections (3) and (5)); and(e)
it places an overriding duty on the court to grant leave only if to refuse to do so might have the result of rendering a conclusion on a relevant issue unsafe (see subsection (2)(b), which is to be read with the definition of "relevant issue" in section 42(1)(a)).
72. It is clear that this structure has been designed in such a way as to balance the competing interests of the complainant who seeks protection from the court and the accused's right to a fair trial. The section leans towards the protection of the complainant. The protection extends to questions and evidence about sexual behaviour after, as well as before, the event giving rise to the charge. It ends the assumption, widely held hitherto, that the complainant's prior sexual behaviour with the defendant is always relevant and admissible. The admissibility of the complainant's sexual behaviour with the defendant is to be determined under the same procedural provisions as those which apply to the admissibility of such behaviour with third parties. But the court is enabled, in the defendant's interest, to give leave in any case which falls within one or other of the two qualifying subsections where to do otherwise might render a conclusion on any issue falling to be proved in the trial by the prosecution or the defence unsafe.
73. Of the two qualifying subsections, the only one that is in play in this case is subsection (3). Subsection (5) applies where the purpose of the evidence or question is to rebut or explain evidence adduced by the prosecution. It was not suggested that the respondent's application was made in reliance upon this subsection. I would prefer not to speculate on the circumstances in which the subsection might be invoked. But it is reasonable to think that it was included with a view to the accused's right to a fair trial. The section places no restrictions on the evidence which may be led by the prosecutor. It would plainly be unfair if the prosecutor were, for example, to lead similar fact evidence to support the Crown's case of the kind described in Director of Public Prosecutions v P  2 AC 447 and the accused were not to be given an opportunity in cross-examination or by adducing evidence to rebut that evidence. Subsection (5) avoids this unfairness.
74. Subsection (3), which is the critical subsection in this case, comprises three qualifying conditions which are stated in the alternative. It requires careful analysis. First there are the opening words of the subsection. They provide that the subsection applies only if the evidence or question relates to a relevant issue in the case - that is, any issue falling to be proved by the prosecution or the defence at the trial: see section 42(1)(a). The wording of this part of the subsection reflects the general tenor of section 41, which is to protect the complainant against evidence or questions about his or her sexual behaviour other than as part of the event which is the subject matter of the charge. Put the other way round, the evidence or question will cross the threshold of subsection (3) if it relates to an issue which falls to be proved by the prosecutor or by the defence. In this respect at least the subsection has been designed to avoid the unfairness which would result if the accused were to be denied the opportunity to lead evidence or put questions directed to issues that were relevant at the trial. Thus far it does not infringe the defendant's right to make a full answer and defence to the charge.
75. But the threshold which is set by the opening words of subsection (3) is further qualified by subsection (4), which provides that for the purposes of subsection (3) - but not, it should be noted, for the purposes of the rebuttal provisions in subsection (5) - no evidence or question shall be regarded as relating to a relevant issue in the case if it appears to the court to be reasonable to assume that the purpose or the main purpose for which it would be adduced or asked would be to impugn the credibility of the complainant as a witness. At first sight this is a serious intrusion on the accused's right to a fair trial. In cases where the accused who is on trial for rape admits that he had sexual intercourse with the complainant on the occasion in question but says that it was with her consent the credibility of the two parties is likely to be the critical issue.
76. But the definition of "sexual behaviour" in section 42(1)(c) excludes for this purpose anything alleged to have taken place as part of the event which is the subject matter of the charge. It appears that subsection (4) is designed to address one of the two evils which lie at the heart of the mischief which forms the background to the enactment. These are the leading of evidence of sexual behaviour other than that which took place as part of the event which is the subject matter of the charge for the sole or main purpose of showing that, by reason of such sexual behaviour, the complainant (a) was more likely to have consented to the sexual conduct which is at issue in the trial or (b) was an unreliable or less than credible witness. These were described by McLachlin J in R v Seaboyer  2 SCR 577, 630G-H as the twin myths that may still inform the thinking of many but have no place in a rational and just system of law. As she put it, evidence of such behaviour cannot in itself be regarded as logically probative of either the complainant's credibility or consent. The evil which this subsection addresses in uncompromising terms is the drawing of impermissible inferences as to the complainant's credibility. I shall deal in the next section of this judgment (see paragraph 90 et seq, post) with the question whether by choosing to deal with this issue in this way the section has infringed the accused's Convention right to a fair trial.
77. Section 41 does not distinguish between evidence or questions about the complainant's sexual behaviour with the accused and the complainant's behaviour with persons other than the accused. The extent to which these two situations ought to be approached differently is left to the determination of the trial judge. There are strong reasons for imposing a narrower prohibition on the complainant's sexual behaviour with third parties. Evidence or questions about sexual behaviour with third parties is likely to be much harder to justify on grounds of relevancy than evidence about sexual behaviour with the defendant. Nevertheless I think that the draftsman was right to avoid laying down an absolute rule on this point. To have done so would have been to risk incompatibility with the accused's right to a fair trial. It is worth noting that the absolute prohibition in the original version of section 276(1) of the Canadian Criminal Code (RSC 1985, c C-46) which was held in R v Seaboyer to be incompatible with the defendant's rights under the Charter of Rights and Freedoms was directed solely to evidence about the sexual activity of the complainant with persons other than the accused. The section, in its original version, placed no restriction on the admissibility of evidence about sexual activity with the accused himself. Much of the discussion in that case is about the relevance or otherwise of the complainant's sexual activity with third parties. But McLachlin J, at p 633F, questioned whether evidence about other sexual activity with the accused should be automatically admissible, and in its revised form section 276(1) of the Code treats both kinds of sexual activity in the same way. In this respect, as counsel for the Secretary of State pointed out (in my view correctly), section 41 follows the Canadian example.
78. It was suggested during the hearing that questions about sexual behaviour with the accused would be less distressing and humiliating than questions about such behaviour with third parties. But to assent to that proposition would, I think, risk developing rules by reference to stereotypes. Each case is different, and there are sound reasons for thinking that complainants are likely to find evidence and questions about their sexual history distressing or humiliating whatever their subject matter. The only proper test is whether the evidence and questions relate to a relevant issue in the case.
79. Paragraph (a) of subsection (3) sets out the first qualifying condition. This is that the issue to which the evidence or question relates is not an issue of consent. The justification for enabling leave to be given in such cases was powerfully argued by McLachlin J in R v Seaboyer, at pp 613E-615B. The distinction which she drew was between impermissible generalisations about consent and specific inferences pointing to guilt or innocence. Examples of issues which will fall within this paragraph because the evidence of sexual behaviour is proffered for specific reasons are (a) the defence of honest belief, which McLachlin J defined for the purposes of her examination of the Canadian legislation as resting on the concept - which I consider to be consistent with that described in Director of Public Prosecutions v Morgan  AC 182 - that the accused may honestly but mistakenly (but not necessarily reasonably) have believed that the complainant was consenting to the sexual act; (b) that the complainant was biased against the accused or had a motive to fabricate the evidence; (c) that there is an alternative explanation for the physical conditions on which the Crown relies to establish that intercourse took place; and (c) especially in the case of young complainants, as in the Scottish case of Love v H M Advocate 1999 SCCR 783, that the detail of their account must have come from some other sexual activity before or after the event which provides an explanation for their knowledge of that activity. The fact that leave may be given for evidence and questions directed to these and similar specific issues under this paragraph is an important protection of the accused's right to a fair trial.
80. Paragraph (b) sets out the second qualifying condition. This is the first of the two qualifying conditions that relate to issues which are issues of consent. To qualify under this condition the evidence or questions must relate to sexual behaviour which is alleged to have taken place "at or about the same time" as the event which is the subject matter of the charge against the accused. The inclusion of the words "or about" give some, but not very much, latitude to the condition imposed by the paragraph. The overall effect is similar to that of the phrase "at or near his own place of work" in section 15(1) of the Trade Union and Labour Relations Act 1974, as substituted by section 16(1) of the Employment Act 1980, which was considered in Rayware Ltd v Transport and General Workers' Union  1 WLR 675. Nourse LJ said, at p 683C-D, that the word "near" is an expanding word, to be extended so far as to give effect to the intention of the legislature. As May LJ said in the same case, at p 682A-B, the question is in the end one of fact and degree.
81. As for the intention of the legislature in the case of section 41 of the 1999 Act, extensive reference was made to statements made by the Home Office ministers as reported in Hansard when the legislation was undergoing examination in Parliament. For the reasons which I explained in R v Secretary of State for the Environment, Transport and the Regions, Ex p Spath Holme Ltd  2 WLR 15, 48C-E, I consider that the effect of the exception to the rule that resort to Hansard is inadmissible for the purpose of construing an Act which was recognised in Pepper v Hart  AC 593 is that, strictly speaking, this exercise is available for the purpose only of preventing the executive from placing a different meaning on words used in legislation from that which they attributed to those words when promoting the legislation in Parliament. In expressing that view I wish to acknowledge the debt which I owe to my noble and learned friend Lord Steyn's valuable discussion of this point in "Pepper v Hart: A re-examination" (2001) 21 Oxford Journal of Legal Studies 59. But that situation does not arise in this case. In answer to a question which was put to him by my noble and learned friend in the course of the hearing counsel for the Secretary of State said in terms that he was not relying on this material as an aid to construction. So the proper course is to construe the words used according to their ordinary meaning without reference to what the ministers said about them in the course of the debates in Parliament.
82. But I think that it is legitimate to refer for the purposes of clarification to the notes to this section in the explanatory notes to the Act prepared by the Home Office. I would use it in the same way as I would use the explanatory note attached to a statutory instrument: see Coventry and Solihull Waste Disposal Co Ltd v Russell (Valuation Officer)  1 WLR 2093, 2103D-G. The relevant note states that it is expected that the phrase "at or about the same time" will generally be interpreted no more widely that 24 hours before or after the offence. The use of the words "or about" avoids the trap of placing a straightjacket around a matter that has to be determined according to the facts and circumstances of each case. It is sufficient for the purposes of this case to say that the previous sexual behaviour of the complainant, including acts of sexual intercourse, about which the respondent wishes to ask questions and lead evidence falls outwith the scope of the phrase "at or about the same time" according to the ordinary meaning of those words. The last act of consensual sexual intercourse which he alleges took place about one week before the alleged rape.
83. Paragraph (c) sets out the third qualifying condition. It is the second of the two qualifying conditions that relate to issues which are issues of consent. The broad concept to which it is addressed is that of similar fact evidence. As the cases of R v Boardman  AC 421 and Director of Public Prosecutions v P  2 AC 447 demonstrate, the principle on which the admissibility of similar fact evidence is based is that evidence which falls into this category may so strongly support the truth of the offence charged that it is fair to admit it notwithstanding its prejudicial effect: per Lord Mackay of Clashfern LC in Director of Public Prosecutions v P, at pp 462H-463A. This qualifying condition recognises that the accused may wish to rely on the same principle in order to support his defence of consent. The similarities which it permits are expressed in two alternatives, which are best examined separately. But very precise limits are set on the extent to which the principle may be used in this context. These are indicated by the concluding words of the subsection, which provides that the condition will not be satisfied unless the similarity "cannot reasonably be explained as a coincidence."
84. The first alternative is that on which the respondent seeks to rely in this case. It relates to the complainant's sexual behaviour on some other occasion which is alleged to have been so similar to any sexual behaviour of the complainant which took place as part of the event charged that it cannot reasonably be explained as a coincidence. In two respects the scope which is given by this provision for the giving of leave to put questions or adduce evidence is quite wide. The alternative is widely enough expressed to cover sexual behaviour with third parties as well as with the accused. And it is widely enough expressed to cover sexual behaviour after as well as before the event charged. To this extent the condition avoids the risk of unfairness to the accused. But the requirement that the similarity cannot reasonably be explained as a coincidence imposes a precisely expressed restriction which is significantly tighter than that which the Crown must satisfy under the rule established in Director of Public Prosecutions v P.
85. On the limited version of the facts of this case which has so far been made available, no similarity is alleged as to the complainant's sexual behaviour with the respondent on previous occasions to any behaviour on her part which took place as part of the event charged except for the bare fact that it included occasions when she is alleged to have had consensual sexual intercourse with him. Mr Rook did not seek to suggest to the Court of Appeal that there was such a similarity as would enable evidence to be adduced or questions asked under section 41(3)(c): see para 19 of the Court of Appeal's judgment. No attempt appears to have been made to investigate the facts to the level of detail that section 41(3)(c) demands.
86. For this reason the respondent's allegations seem to me to invite the criticism that they are based on one of the two evils which lie at the heart of the mischief which the section seeks to address: the myth that simply because the complainant consented to sexual intercourse on previous occasions she was more likely to have consented to sexual intercourse on this occasion. The scope of the requirement that the similarity cannot reasonably be explained as a coincidence is therefore not, as matters stand, the critical issue in this case. In my opinion the application fails on the ground that no similarity other than the bare fact of alleged previous consensual intercourse with the respondent has been demonstrated.
87. On the other hand the question whether the requirement that any similarity that may be alleged cannot reasonably be explained as a coincidence may yet arise in this case, if the respondent is given an opportunity to explain the basis for his application in greater detail. So I would add these comments. The test which this phrase lays down appears to have been taken from R v Boardman  AC 421 and in particular from Lord Salmon's observations where he said, at p 462C-D:
88. It is not easy to see how that dictum, which is taken from the context of criminal sexual conduct, can be applied to conduct on which the accused wishes to rely as a defence to the charge which has been laid against him. I do not think that it is helpful to speculate as to what kinds of sexual conduct will satisfy this test. Each case will have to be approached on its own facts. But on any view it has been deliberately framed in such a way as to indicate, according to the ordinary meaning of the words used, that it will not be easy to satisfy. It has been modified slightly from the strict test which Lord Salmon described because the phrase "cannot be explained as a coincidence" is qualified by the word "reasonably". Nevertheless it leans strongly in favour of the protection of the complainant. I shall deal in the next section of this judgment (see paragraph 90 et seq, post) with the question whether it leans too far.
89. The second alternative in paragraph (c) relates to the complainant's sexual behaviour on some other occasion which is alleged to be so similar to any other sexual behaviour of the complaint which took place "at or about the same time" as the event charged that it cannot reasonably be explained as a coincidence. The scope to be given to this phrase, according to the ordinary meaning of the words used, is the same as that to be given to the same phrase in paragraph (b). As in the case of the first alternative, the sexual behaviour is not limited to sexual behaviour with the accused before the event charged. It can include within its scope sexual behaviour with third parties as well as sexual behaviour with the accused or with third parties which took place after the event. But, as in the case of the first alternative, the scope to be given to this alternative is qualified by the requirement that the similarity cannot reasonably be explained as a coincidence.
The Convention right to a fair trial
90. The right of an accused under article 6(1) of the Convention is to a fair trial. As I observed in Brown v Stott  2 WLR 817, 851C, this is a fundamental and absolute right, to which the rights listed in articles 6(2) and 6(3) are supplementary. The rights listed in article 6(3) include the accused's right to examine or have examined witnesses against him and to obtain the attendance and examination of witnesses on his behalf under the same conditions as witnesses against him: see paragraph (d). There is no doubt that Parliament, by placing restrictions on the questions that may be asked and the evidence that may be adduced by or on behalf of the accused was entering upon a very sensitive area.
91. But article 6 does not give the accused an absolute and unqualified right to put whatever questions he chooses to the witnesses. As this is not one of the rights which are set out in absolute terms in the article it is open, in principle, to modification or restriction so long as this is not incompatible with the absolute right to a fair trial in article 6(1). The test of compatibility which is to be applied where it is contended that those rights which are not absolute should be restricted or modified will not be satisfied if the modification or limitation "does not pursue a legitimate aim and if there is not reasonable proportionality between the means employed and the aim sought to be achieved": Ashingdane v United Kingdom (1985) 7 EHRR 528, 547, para 57. A fair balance must be struck "between the demands of the general interest of the community and the requirements of the protection of the individual's fundamental rights": Sporrong and Lönnroth v Sweden (1982) 5 EHRR 35, 52, para 69. The general principles described in the Ashingdane case were restated in Lithgow v United Kingdom (1986) 8 EHRR 329, 393, para 194 and again in Fayed v United Kingdom (1994) 18 EHRR 393, 429, para 65; see also Brown v Stott  2 WLR 817, 851. The question whether a legitimate aim is being pursued enables account to be taken of the public interest in the rule of law. The principle of proportionality directs attention to the question whether a fair balance has been struck between the general interest of the community and the protection of the individual.
92. In my opinion the placing of restrictions on evidence or questions about the sexual behaviour of complainants in proceedings for sexual offences serves a legitimate aim. The prevalence of sexual offences, especially those involving rape, which are not reported to the prosecuting authorities indicates a marked reluctance on the part of complainants to submit to the process of giving evidence at any trial. The rule of law requires that those who commit criminal acts should be brought to justice. Its enforcement is impaired if the system which the law provides for bringing such cases to trial does not protect the essential witnesses from unnecessary humiliation or distress.
93. It seems to me that the critical question, so far as the accused's right to a fair trial is concerned, is that of proportionality. The points of particular concern which I have identified in my analysis of section 41 are (a) the exclusion by section 41(4) of evidence and questions for the purpose of impugning the credibility of the complainant as a witness (see paragraph 76, ante) and (b) the requirement in section 41(3) that any similarity cannot reasonably be explained as a coincidence (see paragraph 83, ante). The impact of these provisions on the right to a fair trial is highlighted by the fact that they are binding on the trial judge. They are mandatory. He has no discretion to admit the evidence or to allow the questioning if he thinks that it is in the interests of justice to do so.
94. The question is whether these provisions have achieved a fair balance. This will be achieved if they do not go beyond what is necessary to accomplish their objective. That is the essence, in this context, of the principle of proportionality. Furthermore, to ask oneself whether they are fair to the defendant is to address one side of the balance only. On the other side there is the public interest in the rule of law. The law fails in its purpose if those who commit sexual offences are not brought to trial because the protection which it provides against unnecessary distress and humiliation of witnesses is inadequate. So too if evidence or questions are permitted at the trial which lie so close to the margin between what is relevant and permissible and what is irrelevant and impermissible as to risk deflecting juries from the true issues in the case. The high rate of acquittals in rape cases before section 41 was introduced suggests that juries are not immune from temptation, and that they are quite likely to draw inferences from evidence about a complainant's sexual behaviour on occasions other than that of the alleged rape which the law now recognises they should not draw.
95. A prohibition of evidence and questions about the complainant's sexual behaviour on other occasions whose purpose, or main purpose, is to elicit material to impugn the credibility of the complainant as a witness seems to me to strike the correct balance. If the sole purpose is to impugn credibility, the defendant has no rights in the matter at all. The complainant's sexual behaviour on other occasions is irrelevant. No inferences can properly be drawn about her credibility from the mere fact that she has engaged in sexual behaviour on other occasions. I would hold that the words "or main purpose" which qualify the words "the purpose" in section 41(4) do not widen the prohibition to an extent which, when regard is had to the public interest, is unfair.
96. The effect of the requirement in section 41(3)(c) that any similarity cannot reasonably be explained as a coincidence is more difficult to assess. It seems to me that the assessment might best be approached in stages by asking these questions: (1) does a proportionate response to the legitimate aim entitle the legislature, in principle, to restrict the extent to which evidence may be adduced and questions asked about the complainant's other sexual behaviour where the issue is one of consent? (2) if so, are the restrictions in section 41(3)(c) so unfair that it can be said that no defendant who wishes to adduce such evidence or ask such questions can ever have a fair trial because its effect is to exclude relevant evidence whose probative value is not clearly outweighed by the prejudice which it may cause? (3) if not, has it been shown that it will cause such unfairness in this case?
97. It is not necessary to dwell on the first or on the last of these three questions. Some limit must be placed on the extent to which evidence may be adduced and questions asked if the legitimate aim is to be achieved. That point is not in dispute. As far as this case is concerned, I have already mentioned the fact that no attempt appears yet to have been made to investigate the facts to the level of detail that section 41(3)(c) demands. It is not yet possible to say that there is any relevant evidence about similar sexual behaviour by the complainant which would be excluded by the restrictions. So I do not think that it can yet be said that, if the restrictions are not caught by the second question, they are so unfair in this case as not to be proportionate.
98. There remains the second question. I agree with Mr Pannick QC for the Home Secretary that if the restrictions are likely to cause unfairness in isolated cases only, of which this is not one, the better course is to deal with them later and one by one as they arise. The point of the second question is that if it is answered in the affirmative the incompatibility which will result will be capable of being invoked by every defendant whose defence is directed to the issue of consent. That, in effect, is the position which the respondent adopts. He says that there is no point in attempting the exercise required by section 41(3)(c) because the restrictions are so tightly drawn that there is no reasonable prospect of overcoming them.
99. It is plain that the question is in the end one of balance. Has the balance between the protection of the complainant and the accused's right to a fair trial been struck in the right place? As I indicated earlier in this judgment (see paragraph 58, ante), I think that, if any doubt remains on this matter, it raises the further question whether Parliament acted within its discretionary area of judgment when it was choosing the point of balance indicated by section 41. The area is one where Parliament was better equipped than the judges are to decide where the balance lay. The judges are well able to assess the extent to which the restrictions will inhibit questioning or the leading of evidence. But it seems to me that in this highly sensitive and carefully researched field an assessment of the prejudice to the wider interests of the community if the restrictions were not to take that form was more appropriate for Parliament. An important factor for Parliament to consider was the extent to which restrictions were needed in order to restore and maintain public confidence.
100. Some assistance in finding an answer to this question may be gained by looking at the solutions that have commended themselves to other jurisdictions. Rape-shield legislation in the United States has been classified into four different models: H Galvin "Shielding Rape Victims in the State and Federal Courts: A Proposal for the Second Decade" (1986) 70 Minn L Rev 763; see also the helpful summary in Neil Kibble's paper, at pp 25-26. The Michigan model is the one followed most widely in the United States. It has been adopted in New South Wales (section 409B(3) of the Crimes Act 1900) and was adopted by Canada until it was held in R v Seaboyer  2 SCR 577 to be unconstitutional. It imposes a general prohibition on the introduction of evidence of prior sexual behaviour, subject to certain specific exceptions but permits evidence of prior sexual behaviour between the complainant and the defendant. The New Jersey model leaves the matter almost entirely to the discretion of the trial judge, but it provides for the question whether to admit the evidence to be determined at a pre-trial hearing. The Federal model follows the Michigan model to the extent that it imposes a general prohibition on the introduction of prior sexual behaviour with specific exceptions one of which relates to the complainant's behaviour with the accused, but it gives the trial judge a general residual discretion to admit the evidence if it would be contrary to the interests of justice to exclude it or to do so would violate the defendant's constitutional rights. A similar model is in force in Western Australia (sections 36B, 36BA and 36BC of the Evidence Act 1906). The California model prohibits evidence of prior sexual behaviour to prove consent unless the evidence is of prior sexual conduct between the complainant and the defendant, while evidence with respect to credibility is admissible at the discretion of the court.
101. To these four models there now fall to be added two more. The first of these is the revised Canadian model. Section 276 of the Canadian Criminal Code (RSC 1985, c C-46) was redrafted following the decision in R v Seaboyer to give statutory effect to the guidelines which the Supreme Court of Canada laid down in that case for the reception and use of sexual conduct evidence. It starts by providing that evidence of other sexual activity, whether with the accused or with any other person is not admissible to support an inference that, by reason of the sexual nature of that activity, the complainant is more likely to have consented or is less worthy of belief. In R v Darrach 191 DLR (4th) 539, 560 the Supreme Court of Canada held that this is an evidentiary rule that excludes such evidence because it is irrelevant. There are then three exceptions to that rule which allow the evidence to be admitted if the judge determines that the evidence is of specific instances of sexual activity, that it is relevant to an issue at the trial and that it has significant probative value that is not significantly outweighed by the danger of prejudice to the proper administration of justice. Guidelines are included to assist the judge in determining whether the evidence is admissible. In R v Darrach the court held that the procedure created by the revised section 276, taken as a whole, was consistent with the principles of fundamental justice and protected the defendant's constitutional rights.
102. Lastly there is the Scottish model. It was first enacted by section 36 of the Law Reform (Miscellaneous Provisions) (Scotland) Act 1985 and is now to be found in sections 274 and 275 of the Criminal Procedure (Scotland) Act 1995. Questioning designed to show that the complainer is not of good character in sexual matters, that she is a prostitute or that she has at any time engaged with any person in sexual behaviour not forming part of the subject matter of the charge is excluded by section 274. But section 275 provides that such questioning or evidence may be allowed where the court is satisfied that it is designed to explain or rebut other evidence, is questioning or evidence as to sexual behaviour which took place on the same occasion as the sexual behaviour forming the subject matter of the charge or is relevant to the defence of incrimination (that is, that the crime was committed by some other named individual) or that it would be contrary to the interests of justice to exclude it. A study by Dr B Brown "Sexual History and Sexual Character Evidence in Scottish Sexual Offence Trials" (University of Edinburgh, 1992) concluded that, while there were a number of positive features in this legislation, it fell short of achieving its aim in practice. It was suggested that, while there were other possibilities, a more certain remedy would be to modify the discretionary character of the exceptions and to identify instead specific types of circumstances in which sexual history or character evidence would be relevant to key issues in the trial.
103. It is reasonably clear from this brief review that there is no one single answer to the problem as to how best to serve the legitimate aim. There are choices to be made. There are indications from the wording and structure of section 41 that close attention was paid to the more recent Canadian and Scottish models. But in significant ways it has departed from both of them. The element of judicial discretion has been reduced to the minimum. There are risks involved in that choice. It has deprived the judge of the opportunity, in the last resort, of preventing unfairness to the defendant in circumstances where to do this would not significantly prejudice the proper administration of justice.
104. But two important factors seem to me to indicate that prima facie the solution that was chosen was a proportionate one. The first is the need to restore and maintain public confidence in the system for the protection of vulnerable witnesses. Systems which relied on the exercise of a discretion by the trial judge have been called into question. Doubts have been raised as to whether they have achieved their object. I think that it was within the discretionary area of judgment for Parliament to decide not to follow these systems. The second is to be found in a detailed reading of the section as a whole. As I have tried to show in my analysis of the various subsections, it contains important provisions which preserve the defendant's right to ask questions about and adduce evidence of other sexual behaviour by the complainant where this is clearly relevant. While section 41(3) imposes very considerable restrictions, it needs to be seen in its context. I would hold that the required level of unfairness to show that in every case where previous sexual behaviour between the complainant and the accused is alleged the solution adopted is not proportionate has not been demonstrated.
105. I emphasise the words "every case", because I believe that it would only be if there was a material risk of incompatibility with the article 6 Convention right in all such cases that it would be appropriate to lay down a rule of general application as to how, applying section 3 of the Human Rights Act 1998, section 41(3) ought to be read in a way that is compatible with the Convention right or, if that were not possible, to make a declaration of general incompatibility. I do not accept that there is such a risk. This is because I do not regard the mere fact that the complainant had consensual sexual intercourse with the accused on previous occasions as relevant to the issue whether she consented to intercourse on the occasion of the alleged rape.
106. For these reasons I consider that it has not been shown that, if the ordinary principles of statutory construction are applied to them, the provisions of section 41 which are relevant to the respondent's case are incompatible with his Convention right to a fair trial. I would hold that the question whether they are incompatible cannot be finally determined at this stage, as no attempt has been made to investigate the facts to the required level of detail to show that section 41 has made excessive inroads into the Convention right. It seems to me that it is neither necessary nor appropriate at this stage to resort to the interpretative obligation which is described in section 3 of the Human Rights Act in order to modify, alter or supplement the words used by Parliament. I think that it would only be appropriate to resort to surgery of that kind in this case if the words used by Parliament were unable, when they were given their ordinary meaning, to stand up to the test of compatibility. But that cannot, in my view be said of the allegations which the respondent makes as to the complainant's sexual behaviour with him prior to the incident of the alleged rape. All he appears to be relying upon at present is the mere fact that on various occasions during the previous three weeks she had had consensual sexual intercourse with him in his flat. As I have said, I consider that this fact alone - and nothing else is alleged about it - is irrelevant to his defence of consent. So I would hold that the exclusion of evidence and questions which relate to it in regard to that defence (but not that of honest belief ) is not incompatible with his right to a fair trial.
107. This does not mean that the question whether or not the respondent will have a fair trial is at an end. I agree with Mr Perry for the Crown that it will only be in rare and isolated cases, that the question of fairness will be capable of being determined before the trial. It was clearly right that this case should have been brought before your Lordships on appeal in view of the important issues of principle that were raised and the risk of exposing vulnerable witnesses to the risk of having to give evidence at a new trial. But now that these issues have been resolved the case must go back to the Crown Court for trial. The question whether the respondent did in the event have a fair trial will be open for consideration after the trial is over if he is convicted.
108. I should like to add however that I would find it very difficult to accept that it was permissible under section 3 of the Human Rights Act 1998 to read in to section 41(3)(c) a provision to the effect that evidence or questioning which was required to ensure a fair trial under article 6 of the Convention should not be treated as inadmissible. The rule of construction which section 3 lays down is quite unlike any previous rule of statutory interpretation. There is no need to identify an ambiguity or absurdity. Compatibility with Convention rights is the sole guiding principle. That is the paramount object which the rule seeks to achieve. But the rule is only a rule of interpretation. It does not entitle the judges to act as legislators. As Lord Woolf CJ said in Poplar Housing and Regeneration Community Association Ltd v Donogue  EWCA Civ 595, section 3 does not entitle the court to legislate; its task is still one of interpretation. The compatibility is to be achieved only so far as this is possible. Plainly this will not be possible if the legislation contains provisions which expressly contradict the meaning which the enactment would have to be given to make it compatible. It seems to me that the same result must follow if they do so by necessary implication, as this too is a means of identifying the plain intention of Parliament: see Lord Hoffmann's observations in R v Secretary of State for the Home Department, Ex p Simms  2 AC 115, 131F-G.
109. In the present case it seems to me that the entire structure of section 41 contradicts the idea that it is possible to read into it a new provision which would entitle the court to give leave whenever it was of the opinion that this was required to ensure a fair trial. The whole point of the section, as was made clear during the debates in Parliament, was to address the mischief which was thought to have arisen due to the width of the discretion which had previously been given to the trial judge. A deliberate decision was taken not to follow the examples which were to be found elsewhere, such as in section 275 of the Criminal Procedure (Scotland) Act 1995, of provisions which give an overriding discretion to the trial judge to allow the evidence or questioning where it would be contrary to the interests of justice to exclude it. Section 41(2) forbids the exercise of such a discretion unless the court is satisfied as to the matters which that subsection identifies. It seems to me that it would not be possible, without contradicting the plain intention of Parliament, to read in a provision which would enable the court to exercise a wider discretion than that permitted by section 41(2).
110. I would not have the same difficulty with a solution which read down the provisions of subsections (3) or (5), as the case may be, in order to render them compatible with the Convention right. But if that were to be done it would be necessary to identify precisely (a) the words used by the legislature which would otherwise be incompatible with the Convention right and (b) how these words were to be construed, according to the rule which section 3 lays down, to make them compatible. That, it seems to me, is what the rule of construction requires. The court's task is to read and give effect to the legislation which it is asked to construe. The allegations about the complainant's previous sexual behaviour with the respondent are so exiguous that I do not think that it would be possible for your Lordships in this case with any degree of confidence to embark upon that exercise. I would leave that exercise to be undertaken by the trial judge in the light of such further information about the nature and circumstances of his relationship with the complainant that the respondent can make available if and when he renews his application. If he finds it necessary to apply the interpretative obligation under section 3 of the Human Rights Act 1998 to the words used in section 41(3)(c) of the 1999 Act, he should do so by construing those words, so far as it is possible to do so, by applying the test indicated in paragraph 46 of the speech of my noble and learned friend Lord Steyn.
111. The Court of Appeal reversed the decision of the trial judge on the question whether evidence and questions about the complainant's sexual behaviour with third parties would be admissible. I agree with that part of their decision. They also reversed the trial judge on the question whether evidence and questions about the complainant's sexual behaviour with the defendant would be admissible in relation to the defence of honest belief. With that part of their decision I also agree. But I am not satisfied that the respondent and his legal advisers have yet applied their minds sufficiently to the detailed requirements which must be met if his application for leave is to fall within the first alternative of the qualifying condition laid down in section 41(3)(c) in relation to the defence of consent.
112. By its order of 15 January 2001 the Court of Appeal allowed the appeal and reversed the ruling of the trial judge. In the formal petition to the House the appellant asks that that order should be reversed, but in the event at the hearing of the appeal your Lordships were not invited to set aside the order by either party. For those reasons I would dismiss the appeal. But I would hold that the respondent should be given an opportunity, in the light of the decision of this House, to renew his application to the trial judge for leave to be given under section 41(3)(c).
113. This appeal raises important and difficult questions about the meaning and effect of section 41 of the Youth Justice and Criminal Evidence Act 1999. That section is one of three sections comprising Chapter III of Part II of the Act under the title "Protection of Complainants in Proceedings for Sexual Offences." Section 41 deals with certain limitations on the leading of evidence and the asking of questions in cross-examination at the trial of a person charged with a "sexual offence", a phrase which is defined in section 62 and includes a charge of rape.
114. The protection of complainants, particularly in rape cases, has for a considerable time been a matter of very real concern. In 1976 some attempt was made to relieve the situation by the passing of the Sexual Offences (Amendment) Act 1976. Section 2 of that Act provided that where any person charged with rape pled not guilty
In terms of section 2(2) the judge was only to grant leave if he was satisfied "that it would be unfair to that defendant to refuse to allow the evidence to be adduced or the question to be asked". It is to be noted that this provision dealt only with cases of rape and did not relate to sexual experience of a complainant with the defendant. It left the leading of the evidence and the questioning as matter for the discretion of the trial judge. That might on the face of it appear to provide a flexible and reasonable solution to a problem which can in so many cases turn upon the particular circumstances of the case. But the very flexibility came to be seen as too uncertain a safeguard for the protection of the complainant.
115. So another attempt has been made to resolve this difficult and delicate problem. Section 2 of the 1976 Act has been repealed and section 41 of the 1999 Act has taken its place. The present appeal is concerned with construction of this new provision. I do not take up space in repeating the full terms of the section here but the form which has been adopted deserves comment. Subsection (1) of section 41 to an extent echoes the approach adopted in section 2(1) of the 1976 Act. But apart from other less important differences, the new subsection is not confined to cases of rape but extends to a "sexual offence", a term which is explained in section 42(1)(d), and instead of being limited as section 2(1) was to "any sexual experience of a complainant with a person other than that defendant", the new section refers generally to "any sexual behaviour of the complainant". When one comes to the cases where leave to lead evidence or ask questions about the complainant's sexual behaviour, instead of simply relying on the discretion of the trial judge, as was the approach taken in the 1976 Act, the new section seeks to specify with some particularity the matters in relation to which the judge may or may not grant leave.
116. Both in legislation and in academic writings various forms of provision have been conceived to provide the desired restraint on the giving of evidence or the asking of questions about the complainant's sexual history. The forms vary from discretionary provisions to inflexible rules. Diverse opinions have been expressed about the merits of one form as against others. For example Temkin ("Evidence in Sexual Assault Cases: The Scottish Proposal and Alternatives" (1984) 47 MLR 625, 635) expressed the view that experience in jurisdictions outside Scotland "mostly suggests that the discretionary method is ineffective in controlling the use of sexual history evidence". McColgan ("Common Law and the Relevance of Sexual History Evidence" (1996) 16 Oxford Journal of Legal Studies 275, 307) expressed the view that a discretionary approach was wholly inadequate and considered that it was necessary to regulate the admission of sexual history evidence by means of tightly drawn legislation. On the other hand attempts to legislate with a blanket exclusion and strictly defined exceptions have been criticised. The constraints of what has been described as a "pigeon-hole approach" may not prove satisfactory. Sopinka J observed in R v Morin  2 SCR 345, 370-371, that
In the Home Office Report "Speaking Up For Justice" (1998) (para 9.72) under recommendation 63 the working group favoured the Scottish approach subject to consultation on the precise formulation. Various formulations have been described and discussed by Kibble in a valuable paper ("The Admissibility of Prior Sexual History with the Defendant in Sexual Offence Cases"). But while it is interesting to explore these alternatives the problem raised in the present case is not how the legislation should be drafted, but how section 41 is to be construed. It is desirable next to set out the facts.
117. The respondent, A, has been charged with rape. The offence is said to have been committed in the early hours of 14 June 2000. His defence is that sexual intercourse took place with the consent of the complainant. It appears that the complainant had formed an intimate relationship with a friend of the respondent, Y, who was sharing a flat with him, and that at about 9.00 pm on 13 June 2000 the complainant and Y had had sexual intercourse at the flat in the absence of the respondent. At a preparatory hearing counsel for the respondent explained to the court that the defence contention was that there had been a sexual history between the respondent and the complainant for some three weeks prior to the date of the alleged rape and that he had had consensual sexual intercourse with her on occasions during that period. Counsel applied to cross-examine the complainant and to lead evidence about the alleged history. The judge ruled that the act of consensual intercourse with Y could be put to the complainant in cross-examination, but that the complainant could not be cross-examined nor could evidence be led about her alleged sexual relationship with the respondent, nor could a prepared statement by the respondent which outlined his alleged relationship with her be put in evidence. The judge also held that his ruling was prima facie in breach of article 6 of the European Convention for the Protection of Human Rights and Fundamental Freedoms.
118. The respondent sought to challenge the judge's ruling by way of appeal to the Court of Appeal. There the focus of the attack was on section 41(3)(b). The Court of Appeal  EWCA Crim 4 took the view that it was impossible to construe that provision as applying to events even days before the alleged rape. They considered that a fair trial might not be possible if there could not be adduced in support of the defence of consent evidence about previous sexual activity on the part of the complainant with the defendant. But they allowed the appeal on the ground that the judge should not have excluded questioning and evidence about such activity altogether since, as was conceded before the court, that was permissible in relation to the issues of an honest belief of consent. So also the statement by the defendant should have been admissible for that same purpose. They also held the judge to have been wrong in admitting questions about the alleged sexual intercourse with Y on the night before the alleged rape. That point was conceded by the respondent.
119. The question which has been certified by the Court of Appeal is:
Some preliminary observations fall to be made in relation to the question. First, the admissibility of evidence is in itself a matter to be determined by the national law and not by article 6. Article 6 guarantees the right to a fair trial and does not lay down rules as to the admissibility of evidence (Schenk v Switzerland (1988) 13 EHRR 242, 265-266, para 46). The question under the article is whether the proceedings as a whole were fair (Khan v United Kingdom, The Times, 23 May 2000, para 34). On the other hand the exclusion of evidence which is relevant to the defence may well render the trial unfair.
120. Secondly, the House is being asked to determine a matter of compliance with the Convention in advance of the trial. Of course there are strong practical reasons for the taking of such a course and it has been adopted on other occasions. Brown v Stott  2 WLR 817 is but one example. But the course has this consequence that the issue can only be answered on a very preliminary basis at this stage. The alleged previous sexual history of the complainant is claimed to be relevant to the matter of consent. The critical question which arises as matters presently stand is whether that evidence is excluded by section 41. If it is relevant to the defence and necessarily made inadmissible by the statute, then it would be difficult to resist the conclusion that the statute was incompatible with article 6 of the Convention.
121. Thirdly, while in construing the section one may well first adopt the ordinary approach of looking to the meaning of the words and the purpose of the provision, where one is faced with a challenge to the compatibility of the provision with the Convention resort may well require to be had to the rule of interpretation embodied in section 3(1) of the Human Rights Act 1998, namely that "so far as it is possible to do so, primary legislation and subordinate legislation must be read and given effect in a way which is compatible with the Convention rights". If that cannot be done a declaration of incompatibility may require to be made in terms of section 4. It is for Parliament to devise a provision which strikes a proper balance between the interests of the complainant and the interests of the defendant. The question is whether a proper balance has been struck by section 41.
122. The recent development of the law in Canada in regard to the present problem is instructive. Section 276 of the Criminal Code had been so drafted as to provide a blanket exclusion of any evidence on behalf of the accused concerning the sexual activity of the complainant with any person other than the accused unless it was led to rebut evidence led by the prosecution, or it was evidence seeking to incriminate another person as guilty of the offence, or it related to the occasion of the alleged offence and related to the consent which the accused alleged he believed was given by the complainant. The provision came before the Canadian Supreme Court in R v Seaboyer  2 SCR 577 and by a majority was held to be unconstitutional. In the course of the leading judgment McLachlin J identified several examples of evidence which the provision would exclude but which could be critical to the defence of the accused. These comprised evidence to support a defence of honest belief, evidence to show a bias or motive to fabricate the evidence on the part of the complainant, evidence to explain physical evidence on which the prosecution might rely to show intercourse or the use of force, and evidence of a pattern of conduct. However legitimate the aims of the legislation might be it was cast so widely as to deprive the accused of the ability to mount lines of defence which were relevant and legitimate. The section was thereafter revised. The form of the revised version was to exclude evidence of sexual activity of the complainant with the accused or any other person to support an inference that by reason of that activity the complainant was more likely to have consented or was less worthy of belief. That provision sought to lay to rest the so called twin myths relating to the effect of general sexual behaviour on matters of consent and credibility. Further such evidence was to be excluded unless the judge determined that it was of specific instances of sexual behaviour, was relevant to an issue at trial and that it had significant probative value which was not substantially outweighed by the danger of prejudice to the proper administration of justice. The revised version of the provision came before the court in R v Darrach (2000) 191 DLR (4th) 539 and was held to be compatible with the accused's right to make a full answer and defence and was in conformity with the Constitution. The approach taken by the revised provision and the philosophy behind it deserve careful attention. The Canadian experience shows how important it is to secure a proper balance between the necessity to provide sufficient protection for the victim of a sexual offence at the trial of the person accused of the offence and the corresponding necessity to secure that the accused has the opportunity to present any relevant defence which he has to the charge. It is right that the victim be protected. But it is also right that an accused should be allowed a fair trial.
123. The question whether a complainant has consented to intercourse with the defendant is often a critical issue in a trial for rape. So the question what evidence is or is not relevant to the matter of the giving or withholding of consent becomes of particular importance. From at least the nineteenth century the belief seems to have been held that if the complainant had had sexual intercourse with some other man or men than the defendant she would be more likely to have consented to intercourse with the defendant. It seems also to have been believed that such a complainant would be less worthy of belief than a woman of unblemished chastity. These ideas have been labelled as myths in Canadian jurisprudence. In R v Seaboyer McLachlin J (p 630) considered that they had no place in a rational and just system of law.
124. Ideas which may have seemed sound in the nineteenth century should be discarded in the face of the very different society in which we now live. The respect which is due to women in society requires that a proper recognition should be given to their independence of mind and the autonomy which they undoubtedly should enjoy. General beliefs about the propensity of unchaste women to consent to intercourse or to be unworthy of credit which may have held sway in an age when the position of women in society bears little comparison with what it is today should now be seen as heresies and discarded as outmoded. Any general idea that because a woman has consented once she is likely to have consented on the occasion in dispute should be relegated to history. Evidence of a general reputation in relation to sexual matters is properly excluded under section 41(6). The only matters to which evidence or questioning may be permitted are specific instances of alleged sexual behaviour.
125. But there is one vital distinction which must be recognised among the generalities which are sometimes adopted in this context, and that is the distinction between a history of intercourse with the defendant and a history of intercourse with other men. To an extent that distinction has been recognised in the past. While questions could be asked in cross-examination of the complainant about someone other than the defendant evidence could not be called to contradict her answer since that would open the way to an inquiry into a multitude of collateral issues (R v Holmes (1871) LR 1 CCR 334 ). On the other hand evidence could be led to counter an answer where the question had been asked in relation to intercourse with the defendant (R v Riley (1887) 18 QBD 481). But the distinction should be recognised as going further. It may readily be accepted that some evidence at least relating to sexual behaviour with the defendant outside the particular event which is the subject of the trial may be relevant as casting light on the question of the complainant's consent. But I do not consider that evidence of her behaviour with other men should now be accepted as relevant for that purpose. As Lord Justice-Clerk Macdonald said long ago in the Scottish case, Dickie v H M Advocate (1897) 24 R(J) 82, 84:
126. The distinction was noted and stressed in the Report of the Advisory Group on the Law of Rape (The Heilbron Committee) (1975) (Cmnd 6325), p 17, paras 100-101 where it was noticed that cross-examination of the complainant's relationship with other men has always stood on a different footing from cross-examination about her relationship with the accused, and that it was the former which causes the real problem. The view is expressed in para 134 that questions and evidence as to the association with the accused will in general be relevant, but that the previous sexual history of the complainant with other men ought in general not to be introduced. Their recommendations (p 36 of the Report) extended to the allowance of some such evidence but subject to leave and in circumstances which they specified. But, as Kibble has pointed out (p 23), little attention has been paid in the literature in the United Kingdom to the relevance and admissibility of prior sexual history with the accused. The context for discussion has sometimes been the admissibility of sexual history with third parties, and general references to "sexual history" can mask the fact that it is behaviour with third parties which is principally being considered. The failure to draw the distinction can give rise to misunderstanding. It is a distinction which has to be kept in mind when assessing the substance of the "twin myths".
127. It is then somewhat surprising to find that section 41 does not recognise this distinction. It deals throughout with sexual behaviour, an expression which is defined in section 42(1)(c) as meaning:
Of course the admission of any evidence of sexual behaviour with anyone is subject to the obtaining of the leave of the court in terms of section 41(1) as more fully detailed in subsection (2). But it should not be thought that evidence of sexual behaviour with other persons than the defendant is necessarily of equal standing in respect of relevance or of weight as sexual behaviour with the defendant himself. It may be noted in passing that since "sexual behaviour" in section 41(1) includes anything alleged to have taken place as part of the event which is the subject matter of the charge against the accused (see section 42(1)(c)), no restraint seems to be imposed under section 41(1) on evidence or questions about such behaviour, no leave being required. But that appears to conflict with the intention of section 41(5) which envisages some limitation being imposed on such evidence or questions.
128. It is, however, to be noticed that the phrases "sexual behaviour" and "other sexual experience" seem to be referring to matters of conduct or activity, to acts or events, of a sexual character and not to general considerations of the existence of a relationship or to the objective facts of acquaintanceship or familiarity or the lack of such factors. The background to the critical incident forming the subject of the trial is not excluded from the evidence by the section, at least to the extent that it does not trespass into the particular field of any sexual behaviour or experience on the part of the complainant. The converse of this point is made in section 41(6) where the evidence of behaviour which may be permitted under subsections (3) and (5) must relate to specific instances of behaviour and not to anything other than specific instances.
129. The present appeal is concerned strictly with the admission of evidence in relation to an issue of consent. We are thus not immediately concerned to explore section 41(3)(a). But it has to be noticed that on the face of that provision evidence of sexual behaviour of the complainant is not excluded by the statute under any limitations beyond that of the exercise of a judicial discretion under section 41(2). Evidence of sexual behaviour is thus potentially treated as admissible by the statute on any issue other than that of consent. To take one example which was raised during the hearing under reference to the recent Scottish case, Love v H M Advocate 1999 SCCR 783, evidence might be admitted about a sexual experience of the complainant, in that case a boy of fifteen, which had occurred two years after the abuse which was the subject of the charge but which could provide an alternative explanation for his evidence.
130. One example of such evidence which caught the attention of the Court of Appeal is that of evidence directed to showing that the defendant had an honest belief that the complainant was consenting. It was not disputed before us that such evidence falls under section 41(3)(a). That evidence of sexual behaviour with other persons than the defendant should be so allowed seems questionable, but the only safeguard is the discretion of the judge and that would have to be relied upon as excluding irrelevant material. On the other hand evidence of sexual behaviour with the defendant might well be relevant to a defence that he honestly believed that the complainant was consenting on the occasion in question. That might seem to let in by a back door evidence which, apart from the express statutory limitations, might be excluded if presented as evidence relating to an issue of consent. Where evidence of sexual behaviour is admitted under section 41(3)(a) the jury then require to be directed by the judge that they may look to such evidence for the purpose of determining the issue of honest belief but must ignore the evidence for the purpose of determining consent. Such a summing up was thought in the Court of Appeal to be worthy of Wonderland. But such mental ability as the distinction requires may not constitute an impossible feat for a jury. Nor may they be required to engage in it in so many cases. The recognition of the element of honest belief as a factor to be excluded if the necessary mens rea in a charge of rape is to be proved was established in Director of Public Prosecutions v Morgan  AC 182. But even there, as Lord Hailsham of St Marylebone pointed out (p 207) the question was wholly unreal:
In practice the incidence of cases where a real issue of honest belief arises and calls for a delicate analysis may be fewer than may be feared.
131. But in the present case we are concerned only with evidence sought to be presented as relating to an issue of consent. Further the issues as presented in the appeal relate only to evidence of sexual behaviour with the accused. As I have already stated the section recognises the possibility that sexual behaviour with others than the defendant are not necessarily to be excluded. But that is not the question in the present appeal. It is necessary to consider the two categories of case where evidence may be admitted in relation to an issue of consent as set out in section 41(3)(b) and (c). That provision is made for the admission of evidence of specific instances of sexual behaviour in these two paragraphs reflects a recognition that such instances may be relevant to the resolution of a dispute on the issue of consent, that is to say, that the sexual history of the complainant may be relevant to the issue of consent and admissible in evidence.
132. The former provision, section 41(3)(b), relates to the possible admission of evidence of behaviour "alleged to have taken place at or about the same time as the event which is the subject matter of the charge against the accused". There is a degree of elasticity in this provision, but it cannot be strained so far as to allow evidence of incidents remote in time. The problem is to assess the permissible degree of remoteness, and that is a question which will have to be determined finally in light of the circumstances of each particular case. Certainly it appears that incidents must be outwith anything which has taken place as part of the event which is the subject matter of the charge. Section 42(1)(c) excludes such things from the definition of "sexual behaviour" so far as section 41(3)(b) is concerned. But how far backwards or indeed forwards one may go beyond the borders of the event remains for the judge to determine. In State of Missouri v Murray (1992) 842 SW (2d) 122 it was held that incidents bearing on consent which had occurred 3½ and 6½ months before the alleged rape should have been admitted, but the statutory test there was that the incidents be "reasonably contemporaneous" which is a different and broader test than that in section 41(3)(b). It would be undesirable to prescribe any test in terms of days or hours, but where the point of reference is to the time of the event it may be difficult to extend that to a period of several days. Certainly the period which the defendant in the present case seeks to explore could not be treated as within the scope of section 41(3)(b).
133. I turn next to section 41(3)(c). Some use was made in the hearing of the phrase "striking similarity" in relation to this particular provision. That phrase was no doubt used because of the comparison which can be made between the drafting of section 41(3)(c) and the language used in particular by Lord Salmon in R v Boardman  AC 421, 462. But Parliament has chosen not to use that phrase and the standard which it has selected is to my mind something short of a striking similarity. R v Boardman was concerned with the problem of similar fact evidence used for the purpose of establishing that the accused had committed the offence with which he was charged. An example can be seen in R v Butler (1986) 84 Cr App R 12. In the present context the defence is seeking to fortify its proposition that the complainant was consenting to intercourse. The context and the purpose of the evidence is not so much to show from past events that history has been repeated, as to indicate a state of mind on the part of the complainant towards the defendant which is potentially highly relevant to her state of mind on the occasion in question. The language used is thus not quite the language of R v Boardman.
134. The provision seeks to make a comparison between the sexual behaviour of the complainant on some other occasion or occasions and the sexual behaviour of the complainant on the occasion of the alleged rape. That latter occasion is more particularly detailed as being either (i) of sexual behaviour which took place as part of the event which is the subject matter of the charge (the definition of "sexual behaviour" in section 42(1)(c) being expressly qualified for this purpose) or (ii) of any other sexual behaviour of the complainant which took place at or about the same time as that event but not as part of the event which is the subject matter of the charge. It may be that the scope of the latter may coincide with the scope of section 41(3)(b), but that is not a point for decision here. The former occasion or occasions, that is to say the other occasion or occasions of similar behaviour, are not limited by reference to time or to place. In particular it may be noticed that behaviour subsequent to the critical event might be admissible where it bears upon the issue of an alleged earlier consent.
135. The essentials for the application of the provision are that there should be a similarity in any respect between the two incidents of sexual behaviour which cannot reasonably be explained as a coincidence. It is only a similarity that is required, not an identity. Moreover the words "in any respect" deserve to be stressed. On one view any single factor of similarity might suffice to attract the application of the provision, provided that it is not matter of coincidence. That the behaviour was with the same person, the defendant, must be at least a relevant consideration. But if the identity of the defendant was alone sufficient as the non-coincidental factor, that would seem to open the way in almost every case for a complete inquiry into the whole of the complainant's sexual behaviour with the defendant at least in the recent past, and that can hardly have been the intention of the provision. What must be found is a similarity in some other or additional respect. Further the similarity must be such as cannot reasonably be explained as a coincidence. To my mind that does not necessitate that the similarity has to be in some rare or bizarre conduct. So long as the particular factor is of a significance which goes beyond the realm of what could reasonably be explained as a coincidence, it should suffice. Something about the sexual behaviour of the complainant on each of the occasions, such as something said or done by him or her which is not so unremarkable as to be reasonably explained as a coincidence has to be found. I would not attempt any kind of definition and in any event it is the words of the statute which matter, but the language seems to me to be looking for some characteristic or incident of the complainant's sexual behaviour which can reasonably be seen to have a significance beyond the fact that it is contemporaneous with the behaviour and which bears some kind of connection or relationship with the behaviour which on a reasonable view is not a mere matter of chance. The task for the judge is to examine the evidence proposed to be led and see if such a similarity in some respect can be found. The matter comes eventually to be one of the circumstances of the particular case.
136. It may well be that the problem can be resolved in practice without any straining of the language of section 41(3)(c). But if the situation is such that the case cannot readily be fitted within the ordinary meaning of the words, then it would be necessary to resort to the special rule of construction embodied in section 3 of the Human Rights Act 1998. If a case occurred where the evidence of the complainant's sexual behaviour was relevant and important for the defence to make good a case of consent, then it seems to me that the language would have to be strained in order to avoid the injustice to the accused of excluding from a full and proper presentation of his defence. Evidence of other occasions in which the accused had engaged with the complainant in sexual behaviour may not always be relevant. It may, for example, be so remote in time as to make the drawing of any inference from it impossible. But in many cases it may be highly relevant, and a fair trial would be endangered if in such a case it was excluded. I do not consider that the terms of the section are so rigid as to make it impossible for such evidence to be admissible and so to create an incompatibility with the Convention.
137. The question then is whether or not the evidence which is sought to be led in the present case can fall within the scope of section 41(3)(c). This will depend upon a careful assessment of the presence or absence of a similarity beyond coincidence between the previous and the critical occasions. That is properly a matter for the trial judge to determine. In interpreting the section he must bear in mind that he may require to adopt the special standard laid down in section 3 of the Human Rights Act. From the sparse facts which are before us, if not on the ordinary approach to construction, at least if one was to follow the instruction contained in section 3, I am not able to affirm that the evidence in question may not fall within the scope of section 41(3)(c). I consider that the judge should be afforded a fresh opportunity of considering the matter in light of the present appeal.
138. I do not propose to consider the provisions of subsections (4) and (5) in any depth. It seems to me that the former will require a very fine analysis in its practical application. Issues of consent and issues of credibility may well run so close to each other as almost to coincide. A very sharp knife may be required to separate what may be admitted from what may not. The purpose of subsection (4) may be taken to be the abolition of the false idea that a history of sexual behaviour in some way was relevant to credit. The recognition of that myth as heresy is to be welcomed. But the subsection may have to be carefully handled in order to secure that that myth remains buried in the past and at the same time secure the availability of evidence of sexual behaviour which is properly admissible as bearing on the issue of consent.
139. Subsection (5) opens the way for the admissibility of questioning and evidence by or on behalf of the defendant for the sole purposes of rebutting or explaining evidence adduced by the prosecution about any sexual behaviour of the complainant. This includes behaviour forming part of the event which is the subject of the charge and is obviously necessary to give the defendant even a minimal right to present his defence. There may well be room for debate about the precise scope of the subsection. Indeed the respondent sought to argue that a denial of consent by the complainant would enable him to call evidence of the previous sexual relationship which he claims he had with the complainant during the few weeks before the alleged rape. I am not persuaded that the subsection can properly by any standard be construed as to lead to that result. But it is unnecessary to explore its scope for the purposes of the present appeal.
140. Critically this appeal has been concerned with section 41(3)(c) and it is important that there should be some general guidance on the application of that section. I have had the opportunity of seeing in draft the formulation which my noble and learned friend Lord Steyn has proposed in paragraph 46 of his speech. I agree that the effect of the decision in this case can be expressed in the statement which he has made.
141. I recognise the peculiarity of the way in which this problem has come before us, as has been described in the speech of my noble and learned friend Lord Hutton. The circumstances do not make it altogether easy to determine the formal outcome. In light of the view which I have taken and of the consequence of the decision reached by the Court of Appeal I do not consider that their reasoning can be allowed to stand without qualification. On the other hand the appellant has substantially failed in the submissions presented on its behalf. I consider that the proper course is to dismiss the appeal and enable the trial judge to have the opportunity of making a fresh ruling in light of the decision of this House after such further procedure as he may consider appropriate.
142. In a criminal trial there are two principal objectives of the law. One is that a defendant should not be convicted of the crime with which he is charged when he has not committed it. The other is that a defendant who is guilty of the crime with which he is charged should be convicted. But where the crime charged is that of rape, the law must have a third objective which is also of great importance: it is to ensure that the woman who complains that she has been raped is treated with dignity in court and is given protection against cross-examination and evidence which invades her privacy unnecessarily and which subjects her to humiliating questioning and accusations which are irrelevant to the charge against the defendant. The need to protect a witness against unfair questioning applies, of course, to all trials but it is of special importance in a trial for rape. Linked to the third objective is the further consideration that allegations relating to the sexual history of the complainant may distort the course of the trial and divert the jury from the issue which they have to determine.
143. It is the need to achieve both the objective of protecting an innocent defendant and the objective of protecting a woman complainant which gives rise to the difficult and important issue before the House on this appeal. The issue is difficult because in some cases where an innocent defendant wishes to give evidence that prior to the sexual intercourse which gives rise to the charge against him he had had consensual sexual intercourse with the complainant on previous occasions, it may be necessary to permit him to give such evidence and the complainant to be cross-examined on the matter in order to enable the jury to come to a just verdict. On the other hand there will be other cases where the adducing of evidence of the complainant's past sexual conduct and cross-examination about it will be unnecessary to ensure that justice is done and may prevent the conviction of a guilty defendant.
144. Parliament enacted section 41 of the Youth Justice and Criminal Evidence Act 1999 to give effect to the third objective. In the course of their submissions counsel referred to the statements of Ministers in Parliamentary debates in order to show the mischief at which the section was aimed. In the debate on the Bill in the House of Commons on 24 June 1999 a Minister of State at the Home Office, Mr Paul Boateng, said:
145. The thinking of the Government was explained as follows by Lord Williams of Mostyn QC, a Minister of State at the Home Office, during the third reading debate on the Bill in the House of Lords on 23 March 1999:
146. Section 41 provides:
147. The intent of section 41 is to counter what have been described as the two myths. One is that because a woman has had sexual intercourse in the past she is more likely to have consented to intercourse on the occasion in question. The other is that by reason of her sexual behaviour in the past she is less worthy of belief as a witness.
148. As regards the first myth it is important to recognise that the present case is one where the defendant wishes to make the case that the complainant had previously had consensual sexual intercourse with himhe does not wish to make the case that the complainant had previously had consensual sexual intercourse with other men and that therefore it was likely that she had consented to have sexual intercourse with him. This is an important distinction, and I propose to confine my observations to a case such as the present one where a defendant seeks to give evidence of the complainant having had previous consensual sexual intercourse with him.
149. The Supreme Court of Canada in R v Seaboyer (1991) 83 DLR (4th) 193 and R v Darrach (2000) 191 DLR (4th) 539 has given powerful and learned judgments demonstrating that the two myths have no place in modern law, but the judgments were given in cases where the defences which the defendants wished to advance related to the alleged sexual conduct of the complainants with other men and the judgments are primarily directed to this situation. In R v Seaboyer, at p 280D, McLachlin J accepted by implication that in some cases evidence of previous sexual conduct of the complainant with the accused might be admissible. The learned Justice considered, at p 279, the proposal of Professor Harriet Galvin in an article entitled "Shielding Rape Victims in the State and Federal Courts: A Proposal for the Second Decade" (1986) 70 Minn L Rev 763, 903:
McLachlin J then stated, at p 280C:
150. The first question which arises on this appeal is whether evidence that there had been previous consensual sexual intercourse between the complainant and a defendant is a relevant matter for the jury to consider. On this issue I would make two observations. The first is that in enacting section 41(3)(c) Parliament has accepted that there may be circumstances in which previous sexual behaviour on the part of the complainant is relevant to the issue of consent, because paragraph (c) permits evidence of previous sexual behaviour of the complainant where it is alleged to have been, in any respect, so similar to any alleged sexual behaviour of the complainant at the time of the alleged offence that the similarity cannot reasonably be explained as a coincidence.
151. The second observation is that whilst there can be no dispute that the Minister of State was correct to say, in the passage from the debate in the House of Lords which I have set out above, that "The fact that a complainant has consented previously does not mean that she will consent again", it does not follow, in my opinion, where there has been a recent affectionate relationship between a woman and a man, that one cannot say that the fact that she has consented previously is relevant in deciding whether she consented when there was intercourse with the same man a relatively short time later. I consider that there is much force in the statement of Professor Galvin, at p 807 of her article, that
152. In my opinion there will be some cases where evidence of previous consensual sexual intercourse between the complainant and the defendant would be clearly relevant, but there will also be cases where such evidence would not be relevant. Where there has been a recent close and affectionate relationship between the complainant and the defendant it is probable that the evidence will be relevant, not to advance the bare assertion that because she consented in the past she consented on the occasion in question, but for the reason given by Professor Galvin, which is that evidence of such a relationship will show the complainant's specific mindset towards the defendant, namely her affection for him. In relation to this point Professor Galvin, at p 786, cites the opinion of Dean Wigmore that such evidence shows: "an emotion towards the particular defendant tending to allow him to repeat the liberty." But where there had only been some isolated acts of intercourse, even if fairly recently, without the background of an affectionate relationship, it is probable that the evidence will not be relevant. But beyond stating that the test is that of relevance, I think that it is not possible to state with precision where the dividing line is to be drawnit will depend on the facts of the individual case as assessed by the trial judge.
153. An additional difficulty in the present case is that the information before the House as to the evidence which the defendant wishes to give is very limited. The agreed statement of facts records:
154. If the evidence were confined to those bare facts I would be of opinion that it would not be relevant to the issue of consent. But it may be that the defendant will be able to give more detailed evidence of his relationship with the complainant which would make his evidence of previous consensual intercourse relevant.
Is relevant evidence admissible under section 41?
155. Therefore on the basis that further evidence which the defendant may wish to give may be relevant I turn to consider whether such evidence is admissible under section 41. This gives rise to the following questions. The first is whether the evidence would be admissible under section 41 on ordinary principles of construction. If the answer to this question is in the negative the second question is whether the exclusion of the evidence infringes the defendant's right to a fair trial under article 6 of the Convention for the Protection of Human Rights and Fundamental Freedoms ("the Convention"). If the answer to this question is in the affirmative the third question is whether section 41 can be construed pursuant to section 3 of the Human Rights Act 1998 in such a way that it is compatible with article 6. If the answer to this question is in the negative it would be the duty of the House under section 4(2) of the 1998 Act to consider making a declaration that section 41 is incompatible with the Convention right given by article 6.
156. I consider it to be clear that if, as in this case, the defendant says that the most recent act of previous intercourse took place approximately one week before the date of the alleged offence, that sexual behaviour of the complainant cannot come within the scope of the words "at or about the same time as the event which is the subject matter of the charge against the accused" in section 41(3)(b). This is so whether ordinary rules of construction are applied or the more expansive approach to construction is taken under section 3 of the 1998 Act.
157. A more difficult question arises under section 41(3)(c) because that paragraph is, in itself, a difficult one to construe in the context of the section. In R v Boardman  AC 421 this House laid stress on the need for "striking similarity" between the criminal acts with which the accused was charged and other criminal acts which the prosecution sought to put in evidence against him. But in Director of Public Prosecutions v P  2 AC 447 the House stated that "striking similarity" is not an essential element in permitting evidence to be given of other conduct and Lord Mackay of Clashfern LC (with whose speech all the other Members of the House concurred) stated, at p 460D:
158. It is to be noted that paragraph (c) does not contain the term "strikingly similar"it contains the less stringent words "so similar that the similarity cannot reasonably be explained as a coincidence." It also provides that it is sufficient if the sexual behaviour of the complainant is alleged to have been so similar "in any respect". Moreover section 42(1)(c) defines "sexual behaviour" as being "any sexual behaviour".
159. I turn to consider the following hypothetical case. A defendant wishes to give evidence that for a number of months prior to the date of the alleged offence he had had a close and affectionate relationship with the complainant and that he had had frequent consensual intercourse with her during that period. Before intercourse he would kiss her and she would return his kisses. At the time of the alleged offence, before having intercourse, affectionate behaviour took place between them as it had done on the early occasions. Is this evidence admissible under section 41(3)(c)? It can be argued that the similarity between the sexual behaviour of the woman on the earlier occasions and on the occasion in question cannot reasonably be explained as a coincidence: there is a causal connection which is that the woman was fond of the defendant and attracted to him and that is why intercourse has taken place on all occasions. But can it be said that the behaviour is "so similar that the similarity cannot reasonably be explained as a coincidence"? Such behaviour is normal between a man and a woman and so it cannot be said to be "strikingly similar", but that is not what the paragraph requires. Therefore I think there is an argument that such evidence would be admissible under section 41(3)(c). However I consider that some weight must be given to the word "so", which I think was intended to emphasis that mere similarity was not sufficient. Moreover having regard to the way in which the mischief at which the section was directed was described by the Minister of State in the debate in the House of Lords, I do not think that Parliament intended that evidence such as that which I have described in the hypothetical case can be admitted under section 41(3)(c). Therefore I would hold that such evidence is not admissible under the paragraph.
Is section 41, on ordinary principles of construction, incompatible with the right to a fair trial given by article 6?
160. In Brown v Stott  2 WLR 817 the Judicial Committee of the Privy Council considered the circumstances in which a particular right given by article 6 may be qualified by considerations of the public interest which Parliament has taken into account in enacting the statutory provision under consideration; in that case the public interest being the need to address in an effective way the high incidence of death and injury on the roads caused by the misuse of motor vehicles. But it is clear that in relation to a fair trial certain rights are absolute and cannot be qualified. Lord Bingham of Cornhill stated, at p 825A:
And, at p 836B:
Lord Hope of Craighead stated, at p 851A:
161. In the type of case which I have instanced where a man, who may be innocent, wishes to give evidence of previous acts of sexual intercourse with the complainant in the course of a recent close and affectionate relationship, such evidence would be a central and essential part of his defence, and I consider that to deny him the opportunity to cross-examine the complainant and to give such evidence would compromise the overall fairness of the hearing and would deny him the essence of a fair trial. In my opinion the right of a defendant to call relevant evidence, where the absence of such evidence may give rise to an unjust conviction, is an absolute right which cannot be qualified by considerations of public interest, no matter how well founded that public interest may be. This right is well described in the argument of counsel for the appellants as set out by McLachlin J in R v Seaboyer 83 DLR (4th) 193, 260-261:
Therefore I would hold on ordinary principles of construction that section 41 is incompatible with the right to a fair trial given by aticle 6.
Under section 3 of the Human Rights Act 1998 can section 41 be read and given effect in a way which is compatible with the right to a fair trial given by article 6?
162. Section 3(1) provides:
As my noble and learned friend Lord Steyn stated in R v Director of Public Prosecutions, Ex p Kebilene  2 AC 326, 366B, this subsection enacts a strong interpretative obligation, and Lord Cooke of Thorndon, at p 373F, described the subsection as an adjuration. It is clearly desirable that a court should seek to avoid having to make a declaration of incompatibility under section 4 of the 1998 Act unless the clear and express wording of the provision makes this impossible.
163. In paragraph 159 I have observed that on ordinary principles of construction and having regard to the change in emphasis in Director of Public Prosecutions v P  2 AC 447 away from "striking similarity" to "probative force" there is a possible argument that relevant evidence of a previous close and affectionate relationship in which sexual intercourse took place is admissible under section 41(3). Therefore pursuant to the obligation imposed by section 3(1) that section 41 must be read and given effect in a way which is compatible with article 6, I consider that section 41(3)(c) should be read as including evidence of such previous behaviour by the complainant because the defendant claims that her sexual behaviour on previous occasions was similar, and the similarity was not a coincidence because there was a causal connection which was her affection for, and feelings of attraction towards, the defendant. It follows that I am in full agreement with the test of admissibility stated by my noble and learned friend Lord Steyn in paragraph 46 of his speech.
164. Therefore I consider that the matter should be remitted to the trial judge in the Crown Court to consider if the evidence which the defendant wishes to give (as amplified by him if he wishes to do so) is admissible under that test. Having regard to the terms of this speech I think it is unnecessary to answer the certified question.
165. In conclusion I wish to make some observations on the nature of the present "appeal". In this case there was an appeal by the defendant under section 35 of the Criminal Procedure and Investigations Act 1996 from the ruling of the trial judge on a preparatory hearing. The appeal was brought by the defendant to challenge the ruling of the trial judge that section 41 of the 1999 Act prohibited him from giving evidence of, or addressing questions to the complainant, about previous consensual sexual intercourse between himself and the complainant. The Court of Appeal gave two principal rulings. One ruling was that the trial judge was in error in excluding such evidence and questioning entirely, because they were permissible under section 41(3)(a) in relation to the defence of belief as to consent. The other ruling was, however, that the trial judge was correct to rule that such evidence and questioning were not admissible on the issue of consent. Because of the first ruling the Court of Appeal held that the appeal was allowed. Therefore, in discussion with the Court of Appeal after the judgment had been given, counsel for the defendant accepted that he could not seek to appeal to this House as the Court of Appeal had allowed the defendant's appeal.
166. Moreover the Crown did not wish to appeal against the ruling of the Court of Appeal upholding the decision of the trial judge that the evidence and questioning were inadmissible under section 41 on the issue of consent, because this was the result for which the Crown contended.
167. In the course of delivering the judgment of the Court of Appeal Rose LJ stated, at para 15:
However later in the judgment, at para 37, Rose LJ stated in an obiter dictum:
168. This dictum caused the Crown concern because it was its view that the exclusion of evidence of previous sexual behaviour by the complainant with the defendant would not render the trial unfair under article 6. Accordingly the Crown sought leave from the Court of Appeal to appeal to this House, and the Court of Appeal granted leave and certified the following question as being a point of law of general public importance which was involved in the decision to allow the interlocutory appeal:
169. Section 33 of the Criminal Appeal Act 1968 (as amended, inter alia, by section 36 of the 1996 Act) provides:
In the written statement of facts and issues before the House the final paragraph states:
And in its written case the Crown states:
170. Therefore it would appear to be clear that this was not "an appeal . . . at the instance of the . . . prosecutor from [a] decision of the Court of Appeal" rather the prosecutor accepted that the decision of the Court of Appeal was correct but wished to appeal against an obiter dictum of the Court of Appeal.
171. However the Court of Appeal granted leave to appeal and certified that a point of law of general public importance was involved in the decision, and therefore I think, on balance, that the House had jurisdiction to hear the matter. Moreover in this case it was of obvious importance that the issue whether section 41 is compatible with article 6 or whether the section can be read in a way which is compatible with the article should be resolved because of the number of cases pending in which the point arose and of the need to protect vulnerable complainants from the risk of having to give evidence at a second trial and, very understandably, the Court of Appeal was clearly influenced by this consideration.
172. Section 33 of the 1968 Act, however, makes it clear that leave should only be given to appeal from a decision of the Court of Appeal. If it is considered desirable that a party who does not wish to challenge the decision of the Court of Appeal but to obtain a ruling of the House on the correctness of an important obiter dictum in the course of the judgment on a point which affects other cases should be allowed to do so, I consider that the wording of section 33 would have to be amended by Parliament.
173. In the unusual circumstances of the present case where the Crown, as the appellant, is not challenging the decision of the Court of Appeal, the question whether to allow or dismiss the Crown's appeal gives rise to a point of some difficulty, but as I consider that the argument advanced by the Crown gave too narrow a construction to section 41 I would dismiss this appeal.