England and wales court of appeal

  1. On 26 October 2006, in the Crown Court at Bournemouth, before HHJ Jarvis and a jury, Benjamin  Bree , a 25 year old man of excellent previous character, was convicted of rape. There was no dispute that, after a very heavy evening drinking together, he had sexual intercourse with a young woman aged 19 years whom we shall identify as "M". This appeal required us to address the effect of voluntary heavy alcohol consumption as it applies to the law of rape. After the hearing on 13th March we quashed the conviction. These are our reasons.

  2. At the outset of the case, the prosecution invited the jury to convict the appellant on the basis that through self-induced voluntary intoxication, M was effectively unconscious throughout most of the sexual activity, and lacked the capacity to consent. Her evidence was to a different effect. We must therefore set it out in some detail.

  3. On 4th February 2006, the appellant visited his brother, Michael, who was a student at university in Bournemouth. Michael's flat was shared with five other students. One of them was the complainant, M, another student. Each had a separate room with its own en-suite facilities.

  4. The appellant and M had met on a previous occasion. She was invited to spend the afternoon with the appellant and Michael, but was not interested. Later she agreed to join them, and Michael's girlfriend, Holly, for the evening. They all drank a considerable amount of alcohol, apparently buying round for round. M drank two pints of cider, and, over the evening, between four and six drinks of vodka mixed with Red Bull. The appellant, who had been drinking earlier in the day, drank two pints of lager, and then he too moved on to vodka and Red Bull.

  5. Michael and Holly were the first to leave, but they were followed soon afterwards by the appellant and M. They returned to her flat. CCTV coverage showed the appellant and M returning to her flat, arm in arm M could not remember very much about the return journey, but accepted that she must have been conscious as she walked home, and she had all the necessary fobs, keys and passes which she used to gain entrance for them both.

  6. Both girls were badly affected by drink. Holly was sick in the kitchen. M was sick in the shower in her room. She recollected lying on the floor and vomiting. According to her evidence she had only once vomited before as a result of alcohol, and whereas on the previous occasion she had "thrown up" just the once, on this occasion she was continually throwing up.

  7. The appellant and Michael looked after the two girls. M remembered one of the two brothers (in fact it was the appellant) asking her where he could find some shampoo, and washing her hair. There was no suggestion of any sexual activity at this stage. The appellant was behaving unselfishly. After she was asked for the shampoo, M had no particular memory of her hair being washed, and by then, according to her evidence, she became unconscious.

  8. M's next memory, as narrated to the jury, was that she was lying on her bed, but unable to recollect how she got there. She said that the appellant was on the bed with her, his upper body on her lower body, his face between her legs, with his mouth and tongue on and in her vagina. She did not consent. "I did nothing or said anything in response. I felt as if I wasn't in my body. I hadn't recovered significantly from how I felt in the bathroom, and I didn't know how long his mouth was in my vagina. I remember his fingers in my vagina. I could just feel this. I don't know where his head was. The next thing I recall is his coming close by my face and asking if I had a condom. I said no". She said that she did not want to have sex, but she did not say so to him. She felt "like it wasn't happening. I knew I didn't want this but I didn't know how to go about stopping it." She was not co-ordinated in her body. She remembered his penis in her vagina, when she was on her back. She recalled penetration, and pain, and she said "ow". At another point she made some kind of noise which led the appellant to say "shush". To try and avoid sexual intercourse she turned over. She was curled in a ball facing the wall. Although his penis was withdrawn for a while, he penetrated her again. She had no idea how long intercourse lasted. When it ended she was still facing the wall. She did not know whether the appellant had in fact used a condom or not, nor whether he ejaculated or not. Afterwards he asked if she wanted him to stay. She said "no". In her mind she thought "get out of my room", although she did not actually say it. She didn't know "what to say or think, whether he would turn and beat me. I remember him leaving, the door shutting." She got up and locked the door and then returned to lie on her bed curled up in a ball, but she could not remember for how long.

  9. M accepted that her recollection of events was "very patchy". She did not know a great deal about what had gone on, and she agreed that she did not say "no" to sexual intercourse, although she did remember saying "no" when asked if she had a condom. She agreed that there were periods during the incident of which she had no recollection, and so she could not say whether she was responding to the appellant's advances or giving him encouragement. Her case remained that she was not consenting to sexual activity with him.

  10. M's next action was to telephone a friend, Naomi. It was about 4.52am. According to Naomi, the conversation was marked by tears and crying, and to begin with it was difficult to understand what M was trying to tell her. M gave some account of the evening's events, and complained that she had been "used". She said that when she woke up she was 100% sure that she had had sex. She did not use the word "rape". Without going into the details any further than necessary, she eventually telephoned her mother, who gave evidence of a complaint of rape, and there was a subsequent conversation with another witness on the telephone complaining that she had been raped.

  11. Medical evidence was put before the jury on an agreed basis. Based on a full examination of M, this evidence did not advance or undermine the Crown's case.

  12. After the incident was reported to the police, the appellant was arrested. He appeared to the arresting officer to be shocked and extremely upset, and could not believe that an allegation of rape had been made against him. His case throughout, both in police interview and in evidence to the jury, was that although the complainant may have become less inhibited because she was intoxicated, she was lucid enough to consent to sexual intercourse, that she did so, and that he reasonably believed that she was consenting.

  13. We need not set out the full summary of the police interview, which was before the jury. The appellant admitted that M was the worse for drink, and he described looking after her when she was sick. He described his return to the room, and the way in which sexual activity began. He said that he was "absolutely positive" that she was awake and conscious throughout and she seemed to be in a lot better state after she had been sick. She seemed keen, and responded to his touching positively by moaning quietly, and rolling on to her back and opening her legs. He said that she was encouraging him by her moans, and the situation simply escalated. She removed her own pyjama trousers before intercourse. After intercourse started, she asked him if he had a condom, and when he asked whether she had one, she said "no". At that point, reality took over and the heat of the moment ended and he thought better of it. So he climbed off her. He was pretty certain that he had not ejaculated.

  14. He was adamant that M had consented, and her reaction and her movements made her seem "pretty enthusiastic". Her response was encouraging. There was no "ow" or anything like it. At the end of the evening he was not sober, and on a scale of 1 - 10, 1 representing "sober" and 10 "complete drunkenness", he was at 7-6, probably 6. M was also drunk. She had been sick in the shower. At that stage she did not sound too healthy, but once she had got it off her chest, she started to come round and became a lot more coherent.

  15. The appellant's evidence of the incident in his evidence to the jury was to much the same effect. He described how, on their return to his brother's flat, they helped the two girls, who were ill. The complainant had been sick. The appellant brought her some water, cleaned her up and washed her hair. He gave her pyjamas. He left the room while she put them on.

  16. He then left the building for a cigarette. He returned about five minutes later. The time, confirmed by CCTV footage, was 3.20am. He went to the complainant's room to make sure she was alright and take her some water. When he came in she was awake, on her bed in her pyjamas. He put water and a bin near her bed. He sat on the edge of the bed, and started to stroke her. He insisted that M appeared to welcome his advances, which progressed from stroking of a comforting nature to sexual touching. She said and did nothing to stop him.

  17. He told the jury that one needed to be sure about consent which is why he stroked her for so long. The complainant could not gainsay that this foreplay lasted for some time. Eventually he put the top of his fingers inside the waistband of her pyjama trousers, which would have given her an opportunity to discourage him. She did not. She seemed particularly responsive when he put his hand inside her pyjama trousers. After sexual touching, he motioned for her to remove her pyjama trousers. He pulled them down slightly, then she removed them altogether. He removed his own trousers. They had "brief sex". She was moist, and became vocal, and he told her to "shush". As he had explained in his police interview, while they were having intercourse, she asked "do you have a condom?", and he said that he did not. He asked her whether she had one. She replied "no". She was concerned about having unprotected sex, so he stopped. He did not ejaculate. She never said or did anything to give the impression that she was in pain, or that she was not consenting. Afterwards he went to the bathroom and washed his face. He returned to M and asked her whether she wanted him to stay the night. She said not, so he kissed her shoulder and left the room.

  18. He agreed that on their return to the flat M was intoxicated and influenced by alcohol. She was not drunk and incapable. When he had used the word "drunk" in interview he had not used it in the sense of a definitive scale of intoxication. She was not that drunk, and given the way in which she had responded to him, he did not think it necessary to ask if it was alright to continue with sexual intercourse.

  19. An expert witness calculated that by 3.45am, the complainant's blood alcohol would have been 60mg per 100ml, rather under the legal limit for driving. However that calculation did not, and could not, take account of the fact that before intercourse took place, the complainant had vomited severely. It was formally admitted by the prosecution that "excessive alcohol consumption can produce marked sedation and may also impair memory, which may result in "blackout"".

  20. As we have indicated, at the start of the trial the prosecution alleged that the appellant raped M when her level of intoxication was so great that she was effectively unconscious. She lacked the capacity to consent, and therefore did not consent. However, by the end of the evidence, the prosecution case against the appellant had changed. The jury were no longer invited to conclude that M had been unable to consent to intercourse because she was unconscious, rather, the prosecution accepted that the gaps in her recollection were probably the result of intoxication, and lack of memory, rather than unconsciousness. The prosecution case, therefore, was not that the complainant lacked the capacity to consent, but that she did not in fact consent to intercourse. Her ability to resist was hampered by the effects of alcohol, but her capacity to consent remained. She knew what was happening. She knew that she did not want to have sexual intercourse, and so far as she could, made that clear. The appellant's case, as we have indicated, was unchanged from start to finish, that notwithstanding, and perhaps because of drink, M was consenting. He reasonably believed that she was.

Discussion: Intoxication And Consent

  1. Section 1 of the Sexual Offences Act 2003 (the 2003 Act) provides the current definition of rape.

"(1) A person (A) commits an offence if -

(a) he intentionally penetrates the vagina, anus or mouth of another person (B) with his penis,

(b) B does not consent to the penetrations, and

(c) A does not reasonably believe that B consents

...2) Whether a belief is reasonable is to be determined having regard to all the circumstances, including any steps A has taken to ascertain whether B consents.

(3) Section 75 and 76 apply to an offence under this section.

(4) A person guilty of an offence under this section is liable, on conviction on indictment, to imprisonment for life."

  1. Section 74 of the 2003 Act defines consent:

"…a person consents if he agrees by choice, and has the freedom and capacity to make that choice".

One of the objectives of the Sexual Offences Act 2003, which came into force on 1 May 2004, was to bring coherence and clarity to the meaning of consent. The provisions relating to consent represented the result of substantial discussion and Parliamentary debate about the principles which should apply to the acutely sensitive and intensely personal area of sexual relationships, whether they arise in the context of a long established marriage, or partnership, or a casual sexual encounter between total strangers. Arguments about consent abound just because consent to sexual intercourse extends from passionate enthusiasm to reluctant or bored acquiescence, and its absence includes quiet submission or surrender as well as determined physical resistance against an attacker which might expose the victim to injury, and sometimes death. The declared objective of the White Paper, Protecting the Public (Cm. 5668, 2002) was to produce statutory provisions relating to consent which would be "clear and unambiguous". As enacted, the legislation on this topic has not commanded totally uncritical enthusiasm. For some it goes too far, and for others not far enough. The law in the area, and our decision, must be governed by the definition of consent in section 74.

  1. Neither "freedom", nor "capacity", are further defined or explained within section 74 itself, nor indeed in sections 75 and 76, which create evidential presumptions relating to consent. We note the analysis in the illuminating article, The Sexual Offences Act 2003, Rape, Sexual Assault and the Problems of Consent,(2004)CLR 328 by Professor Temkin and Professor Ashworth, that "it might be thought that "freedom" and "choice" are ideas which raise philosophical issues of such complexity as to be ill-suited to the needs of criminal justice - clearly those words do not refer to total freedom or choice, so all the questions about how much liberty of action satisfies the "definition" remains at large". Notwithstanding these philosophical difficulties, it is clear that for the purposes of the 2003 Act "capacity" is integral to the concept of "choice", and therefore to "consent".

  2. Section 75 and section 76 of the 2003 Act address the issue of consent in practical situations which arise from time to time in cases of alleged sexual offences including rape. They are not, however, exhaustive. The presumptions in section 75 are evidential and rebuttable, whereas those in section 76 are irrebuttable and conclusive. In this appeal we are not concerned with either of the conclusive presumptions relating to consent specified in section 76. The common characteristic of the particular situations covered by the evidential presumptions in section 75 is that they are concerned with situations in which the complainant is involuntarily at a disadvantage. Section 75 (2) (f) is plainly adequate to deal with the situation when a drink is "spiked", but unless productive of a state of near unconsciousness, or incapacity, this paragraph does not address seductive blandishments to have "just one more" drink. Section 75 (2)(d) repeats well established common law principles, and acknowledges plain good sense, that, if the complainant is unconscious as a result of her voluntary consumption of alcohol, the starting point is to presume that she is not consenting to intercourse. Beyond that, the Act is silent about the impact of excessive but voluntary alcohol consumption on the ability to give consent to intercourse, or indeed to consent generally.

  3. It is perhaps helpful to identify a number of features of the law relating to consent which although obvious are sometimes overlooked. On any view, both parties to the act of sexual intercourse with which this case is concerned were the worse for drink. Both were adults. Neither acted unlawfully in drinking to excess. They were both free to choose how much to drink, and with whom. Both were free, if they wished, to have intercourse with each other. There is nothing abnormal, surprising, or even unusual about men and women having consensual intercourse when one, or other, or both have voluntarily consumed a great deal of alcohol. Provided intercourse is indeed consensual, it is not rape.

  4. In cases which are said to arise after voluntary consumption of alcohol the question is not whether the alcohol made either or both less inhibited than they would have been if sober, nor whether either or both might afterwards have regretted what had happened, and indeed wished that it had not. If the complainant consents, her consent cannot be revoked. Moreover it is not a question whether either or both may have had very poor recollection of precisely what had happened. That may be relevant the reliability of their evidence. Finally, and certainly, it is not a question whether either or both was behaving irresponsibly. As they were both autonomous adults, the essential question for decision is, as it always is, whether the evidence proved that the appellant had sexual intercourse with the complainant without her consent.

  5. Before the 2003 Act, it was not difficult to identify the relevant legal principles, and for a judge to explain the law relating to the voluntary consumption of alcohol (or drugs) by a complainant. Thus, for example, in R v  Malone [1998] 2 CAR 447 the Court of Appeal upheld the direction:

"She does not claim to have physically resisted nor to have verbally protested. She says the drink has disabled her from doing either….she has told you she did not consent….you must be sure that the act of sexual intercourse occurred without (her) consent. Submitting to an act of sexual intercourse, because through drink she was unable physically to resist though she wished to, is not consent. If she submits to intercourse because of the drink she cannot physically resist, that, of course, is not consent. No right thinking person would say that in those circumstances she was genuinely consenting to what occurred. What occurred….not wishing to have intercourse but being physically unable to do anything about it…would plainly, as a matter of common sense be against her will. It would be without her consent".

  1. We record this direction as illustrative of what was regarded as an appropriate direction in the circumstances of an individual case to a particular jury, rather than a learned disquisition of the law of consent as applied to rape. We should however highlight  R v  Lang [1976] 62 CAR 50 which summarised the relevant principle. The jury sought guidance from the judge on the question of whether the complainant's alcohol consumption may have vitiated her consent to sexual intercourse. The court observed

"…there is no special rule applicable to drink and rape. If the issue be, as here, did the woman consent? the critical question is not how she came to take the drink, but whether she understood her situation and was capable of making up her mind. In Howard [1965] 50 CAR 56 the Court of Criminal Appeal had to consider the case of a girl under 16. Lord Parker CJ….said:… "in the case of a girl under 16 the prosecution…must prove either that she physically resisted, or, if she did not, that her understanding and knowledge was such that she was not in a position to decide whether to consent or resist ". In our view these words are of general application when ever there is present some factor, be it permanent or transient, suggesting the absence of such understanding or knowledge…. None of this was explained to the jury. Their attention was focussed by the judge upon how she came to take drink, not upon the state of her understanding and her capacity to exercise judgment in the circumstances."

  1. In the context of the statutory provision in section 74, it is noteworthy that Lang decided thirty years or so ago, directly focussed on the "capacity" of the complainant to decide whether to consent to intercourse or not. These are the concepts with which the 2003 Act itself is concerned.

  2. We are not aware of any reported decisions which deal with this aspect of the new legislation. We should however refer to the much publicised case of  R v  Dougal, heard in Swansea Crown Court, in November 2005. Having heard the evidence of the complainant, the Crown decided to offer no further evidence. Before the jury counsel for the Crown explained:

"the prosecution are conscious of the fact that a drunken consent is still a consent and that in the answer, in cross examination, she said, in terms, that she could not remember giving her consent and that is fatal to the prosecution's case. In those circumstances the prosecution will have no further evidence on the issue of consent. This is a case of the word of the defendant against that of the complainant on that feature It is fatal to the prosecution's case…"

  1. The judge (Roderick Evans J) directed the jury that as the prosecution was no longer seeking a guilty verdict, there was only one verdict which could be returned, and that was an acquittal. He added that he agreed with the course the prosecution had taken.

  2. Without knowing all the details of the case, and focusing exclusively on the observations of counsel for the Crown in Dougal, it would be open to question whether the inability of the complainant to remember whether she gave her consent or not might on further reflection be approached rather differently. Prosecuting counsel may wish he had expressed himself more felicitously. That said, one of the most familiar directions of law provided to juries who are being asked to conclude that the voluntary consumption of alcohol by a defendant should lead to the conclusion that he was too drunk to form the intention required for proof of the crime alleged against him, is that a drunken intent is still an intent. ( R v  Sheehan and Moore [1975] 60 CAR 308 at 312). So it is, and that we suspect is the source of the phrase that a "drunken consent is still consent". In the context of consent to intercourse, the phrase lacks delicacy, but, properly understood, it provides a useful shorthand accurately encapsulating the legal position. We note in passing that it also acts as a reminder that a drunken man who intends to commit rape, and does so, is not excused by the fact that his intention is a drunken intention.

  3. Some of the hugely critical discussion arising after Dougal missed the essential point. Neither counsel for the Crown, nor for that matter the judge, was saying or coming anywhere near saying, either that a complainant who through drink is incapable of consenting to intercourse must nevertheless be deemed to have consented to it, or that a man is at liberty to have sexual intercourse with a woman who happens to be drunk, on the basis that her drunkenness deprives her of her right to choose whether to have intercourse or not. Such ideas are wrong in law, and indeed, offensive. All that was being said in Dougal was that when someone who has had a lot to drink is in fact consenting to intercourse, then that is what she is doing, consenting: equally, if after taking drink, she is not consenting, then by definition intercourse is taking place without her consent. This is unexceptionable.

  4. In our judgment, the proper construction of section 74 of the 2003 Act, as applied to the problem now under discussion, leads to clear conclusions. If, through drink (or for any other reason) the complainant has temporarily lost her capacity to choose whether to have intercourse on the relevant occasion, she is not consenting, and subject to questions about the defendant's state of mind, if intercourse takes place, this would be rape. However, where the complainant has voluntarily consumed even substantial quantities of alcohol, but nevertheless remains capable of choosing whether or not to have intercourse, and in drink agrees to do so, this would not be rape. We should perhaps underline that, as a matter of practical reality, capacity to consent may evaporate well before a complainant becomes unconscious. Whether this is so or not, however, is fact specific, or more accurately, depends on the actual state of mind of the individuals involved on the particular occasion.

  5. Considerations like these underline the fact that it would be unrealistic to endeavour to create some kind of grid system which would enable the answer to these questions to be related to some prescribed level of alcohol consumption. Experience shows that different individuals have a greater or lesser capacity to cope with alcohol than others, and indeed the ability of a single individual to do so may vary from day to day. The practical reality is that there are some areas of human behaviour which are inapt for detailed legislative structures. In this context, provisions intended to protect women from sexual assaults might very well be conflated into a system which would provide patronising interference with the right of autonomous adults to make personal decisions for themselves.

  6. For these reasons, notwithstanding criticisms of the statutory provisions, in our view the 2003 Act provides a clear definition of "consent" for the purposes of the law of rape, and by defining it with reference to "capacity to make that choice", sufficiently addresses the issue of consent in the context of voluntary consumption of alcohol by the complainant. The problems do not arise from the legal principles. They lie with infinite circumstances of human behaviour, usually taking place in private without independent evidence, and the consequent difficulties of proving this very serious offence.

  7. The summing up

The striking feature of the summing up, which is criticised in a number of different ways, is that it does not directly address either the general problems to which this kind of case may give rise, nor their specific application to the present case.

  1. The jury were rightly directed that an essential requirement before the appellant could be convicted was that M did not consent to intercourse. They were told that "a person consents if he agrees by choice and has the freedom and capacity to make that choice". The statutory definition having been read, no further elucidation was given. Our attention was drawn to  R v Olugboga [1981] 73 CAR 344, decided after the enactment of the Sexual Offences (Amendment) Act 1976. As Professor Temkin and Professor Ashworth explain, the report Setting the Boundaries: Reforming the Law on Sexual Offences(2000) which echoed a much earlier report by an advisory group chaired by Heilbron J in November 1975, suggested that the broad approach to consent and submission adopted in Olugboga should be abandoned. In our view, even if these criticisms are justified, the judgment contains passages of continuing value. The court rejected the submission on behalf of the Crown that a trial judge was required "merely to leave the issue of consent to a jury in a similar way to that in which the issue of dishonesty is left in trials for offences under the Theft Act". Because of the myriad circumstances in which the issue of consent may arise, the judgment continued, "We do not think that the issue of consent should be left to a jury without some further direction. What this should be will depend on the circumstances of each case."

  2. In this case the jury should have been given some assistance with the meaning of "capacity" in circumstances where the complainant was affected by her own voluntarily induced intoxication, and also whether, and to what extent they could take that into account in deciding whether she had consented. Moreover, the judge did not address the changed way in which the prosecution put its case against the appellant. There is a significant difference between an allegation that the complainant was unconscious and for that reason not consenting to intercourse, and an allegation that, although she was capable of giving consent, despite her state, she was not in fact consenting to intercourse and was giving clear indications that she was rejecting the appellant. The potential for confusion was compounded by the fact that the complainant herself asserted, more than once, that she was unconscious at different stages of the encounter. At the same time the Crown conceded that what she believed to be and said were periods of unconsciousness should for the purposes of the trial be treated as moments of memory deficit caused by drink. Of course if the Crown was not contending that she was unconscious, that at least was consistent with the appellant's case that she was indeed conscious throughout.

  3. The jury were not provided with any assistance about how properly to address these problems. Thus, when summing up, the judge referred more than once to the complainant's evidence of occasions when she had been unconscious without reminding them of the Crown's concessions. For example, he reminded the jury that M said "she had no memory particularly of her hair being washed after the shampoo was asked for. She was unconscious. (Her) next memory is being of her being on her bed …." It is therefore at least possible that the jury proceeded on the basis that the complainant was indeed unconscious, contrary to the prosecution case in its developed form, but as she herself had asserted. If so, the conclusion that she was not consenting to intercourse would have followed without much difficulty. In a situation like this, the approach in Olugboga, that the issue of consent and capacity should be directly addressed, applied with yet greater force.

  4. The problem was further compounded by the way in which the judge actually addressed the issue of voluntary intoxication. He rightly pointed out that "drink has played a dominant feature in the evidence in this case", but that was said in the context that the appellant's "self induced intoxication can never be a factor which can properly be taken into account when considering [sic] Mr Bree does not reasonably believe that M consented". That attempted to address, whether adequately or not, the effect of drink on the appellant. So far as the complainant was concerned, the direction was even more limited. The judge pointed out that M accepted that "she had drunk a great deal during the evening, and you must, of course, consider that and its potential impact upon her reliability…. It is something which I am sure you have in mind in any event, and it is fair that you bear that in mind when considering her evidence and the whole of the evidence in this case".

  5. In short, the only specific feature of the complainant's alcohol consumption identified by the judge was its possible relevance to her reliability as a witness. Beyond that, if the jury were able to derive anything from what the judge said, it was vague in the extreme. The context, after all, was that although the appellant conceded that the complainant had been drunk, it was a fundamental part of his defence that she was conscious throughout and did in fact consent to sexual activities and intercourse with him. From the defence point of view, the drink she had consumed was a factor which may have led her to behave in a way which, if sober, she would not. She had drunk far more that she was accustomed to. This critical aspect of the case was not sufficiently addressed in the summing up, indeed it was not addressed at all. The questions whether she might have behaved differently drunk than she would have done sober, and whether, although and perhaps because drunk, she might have behaved as the appellant contended, and the way in which the jury should consider these important issues, were not mentioned at all.

  6. A number of further features of the summing up were criticised, but it is unnecessary to deal with them. In a trial in which the issues of consent and voluntary intoxication were fundamental to the outcome, the jury were given no or no sufficient directions to enable the verdict which they reached to be regarded as safe. Accordingly the conviction was quashed