Admissibility of Expert Evidence
    Dr Matthew Weait Reader in Socio-Legal Studies Birkbeck College University of London With thanks to Edwin Bernard and Robert James
    
    1.
    
    Scope of Submission
    
    In this brief submission I shall be focusing on two issues. These are (1) the admission of expert testimony to explain phylogenetic evidence in cases involving the alleged transmission of HIV;1 and (2) expert testimony as to the relevance of viral load. 2. 2.1 The Issues Phylogenetic evidence2
    
    In English law, a person is criminally liable if he intentionally or recklessly transmits HIV or another serious STI to a partner.3 In order to prove the actus reus of the offence, the prosecution must establish to the criminal standard that the defendant was the cause of the harm to the complainant. To do this, the prosecution relies on evidence gained by phylogenetic analysis. Phylogenetic analysis is a method that can establish whether the sub-type of HIV in the defendant’s body is the same as that in the complainant’s. In a number of cases involving HIV transmission the defendants pleaded guilty having been confronted with this evidence. However, in one of the cases where defendants pleaded not guilty and challenged this evidence they were acquitted on the direction of the trial judge.4 In the other three acquittals the phylogenetic evidence is barely discussed in the transcripts. In one the charge was already dismissed by the judge prior to the hearing and in another phylogenetic analysis had not even taken place, but the inability of the prosecution to find all of the complainant’s previous partners led to the judge accepting the defence application to dismiss. The reason for this is that phylogenetic analysis evidence cannot prove the timing, route or source of transmission.5 The only evidential value that phylogenetic
    1
    
    I limit the focus to HIV here, though the points are similarly applicable to cases involving the alleged transmission of other STIs. See, for example, the case of Ercan Yasar, who pleaded guilty in November 2008 to recklessly transmitting Hepatitis B (HBV) to a sexual partner. The “DNA” evidence (the status of which is unclear) was – because of the plea – not subject to cross-examination: transcript of the Trial of R v Ercan Yasar. Indictment No T20080252, Gloucester Crown Court, 18/11/2008 (in author‘s possession). See also the very unsatisfactory case of R v Peace Marangwanda [2009] EWCA Crim 60 (transmission of gonorrhoea). 2 A fuller discussion of the issues is provided in the Appendix. 3 Sections 18 and 20 of the Offences Against the Person Act 1861. See R. v Dica, [2004] 2 Cr App R 28; R. v Konzani [2005] 2 CrApp R 14. 4 R v Matthew Collins. In the other three acquittals the phylogenetic evidence is barely discussed in the transcripts. In one the charge was already dismissed by the judge prior to the hearing and in another phylogenetic analysis had not even taken place, but the inability of the prosecution to find all of the complainant’s previous partners led to the judge accepting the defence application to dismiss. For discussion see Weait, M. (2007) Intimacy and responsibility: the Criminalisation of HIV Transmission, Abingdon: Routledge-Cavendish, pp 32-33. 5 Bernard, E., Geretti, A-M, van Damme, A., Azad, Y. and Weait, M. (2007) ‘HIV forensics: pitfalls and acceptable standards in the use of phylogenetic analysis as evidence in criminal investigations of
    
    1
    
    analysis evidence can have, therefore, is to exclude the possibility of transmission, or to support other evidence that tends towards proving guilt. Put more strongly, a defendant cannot, unless there is no possibility whatsoever that the complainant could have contracted the same sub-type from another source, know that he or she is that source; nor is it possible for the prosecution to prove that she or he is in fact the source. The use of phylogenetic evidence to prove the transmission of HIV points to a paradox that I believe is not fully acknowledged in the Consultation Paper. In Part 3 of the Paper it is rightly emphasised that expert evidence will only be admissible if it meets the tests of relevance and reliability. The use of phylogenetic evidence suggests that the converse can also be the case. It is precisely because the evidence being admitted in these cases is irrelevant and unreliable that people are being convicted of serious offences and sentenced to immediate custody. At 6.6 the Paper states, in its core proposal, that the trial judge would have first to determine whether the evidence is logically relevant to a disputed matter. The answer to this is yes (to the extent it may support other evidence). However, at 6.10, where it sets out its proposals for a statutory test, the Paper suggests that it will be for the party seeking to have the evidence admitted to show that it is sufficiently reliable. In kinds of case discussed in this submission, it is the prosecution that obtains the evidence and seeks to have it admitted. If the points made in this submission are accepted, phylogenetic analysis evidence should never be admitted as a means of proving the actus reus of the offence because it does not meet the reliability criterion. This, however, does not prevent people being charged, prosecuted and pleading guilty because they are misled into believing that the evidence is reliable. It would be good if the Law Commission could find a way of addressing this important problem. My suggestions are that, • Where proof of the actus reus cannot be established scientifically judicial notice (para 6.54) should be taken of that fact and the jury directed as to this fact; or, alternatively Where scientific evidence is necessary to prove the actus reus of the offence o it should not be open to a court to accept a guilty verdict without crossexamination of an expert; and o the generally accepted limitations of the probative value of that expert evidence should be clearly explained to the jury in all cases. Viral load
    
    •
    
    3.
    
    The viral load, or viraemia, of a person living with HIV and not suffering from any other STI who is being treated with Anti-Retroviral Therapy (ART) may be negligible. In such cases the risk of onward transmission is substantially reduced. Indeed, in a recent Swiss case the conviction of a man for exposing a partner to the risk of transmission (a criminal offence in that country) was overturned after expert evidence to this effect was accepted by the court.6 (Evidence was provided by
    HIV transmission’ HIV Medicine, 8, 382–387; Pillay, D. and Fisher, M. (2007) ‘Primary HIV infection, phylogenetics, and antiretroviral prevention’, The Journal of Infectious Diseases, 195: 924-926. 6 Vernazzaa, P., Hirschel, B., Bernasconic, E., Fleppd, M., ‘Les personnes séropositives ne souffrant d’aucune autre MST et suivant un traitement antirétroviral efficacene transmettent pas le VIH par voie
    
    2
    
    Professor Bernard Hirschel, one of the authors of the Swiss Federal AIDS Commission's statement, published in January 2008, regarding the non-infectiousness of HIV-positive individuals on successful treatment (“The Swiss statement”).) In England and Wales (though not in Scotland) there can be no criminal liability merely for exposing someone to the risk of transmission. However, it is suggested that the Swiss statement does have relevance in the context of this submission. The reason for this is that people living with HIV who know that their viral load is negligible may believe that they are not reckless in the legal sense if they engage in sex which carries with it the risk of transmission. There are two possible situations which may confront a court where such a person is charged with recklessly transmitting HIV (setting aside the evidential problems set out in section 2 above). First, such a person may use a condom, but the condom fails. Second, they may choose not use a condom. CPS guidance states that: Evidence that the defendant took appropriate safeguards to prevent the transmission of their infection throughout the entire period of sexual activity, and evidence that those safeguards satisfy medical experts as reasonable in light of the nature of the infection, will mean that it will be highly unlikely that the prosecution will be able to demonstrate that the defendant was reckless.7 The Swiss statement was not published until after the CPS had published its guidance, and it is clear from the context of that guidance that “appropriate safeguards” was understood as the use of condoms. In light of the Swiss statement, it is suggested that: • Courts should admit expert evidence tendered by the defence to the effect that a person on ART with a negligible viral load, not suffering from any other STI, should not be treated as infectious; If such evidence is tendered, it would be for the judge to determine whether there is a case to answer. If the judge concludes in the light of this evidence that there is no evidence on which the jury can find the defendant reckless, she or he should direct an acquittal; If the judge concludes in the light of this evidence that there is a case to answer, she or he should direct the jury on the meaning of recklessness taking into account the honest belief which the defendant may have had as to his or her risk of onward transmission.
    
    • •
    
    •
    
    In the context of the Consultation Paper, this part of the submission is probably too fact-specific to be accommodated; but I have included it as an example of a case where the existence of reliable expert evidence may have an impact on determining questions of mens rea (something that the Consultation Paper does not otherwise address).
    
    sexuelle’ Bulletin des médecins suisses 2008;89: 5, 165-169 (http://www.saez.ch/pdf_f/2008/200805/2008-05-089.PDF). A discussion of the case may be found at http://www.aidsmap.com/en/news/CEFD90F2-34F1-4570-B9CF-1F0DB462AC9D.asp. 7 Crown Prosecution Guidance on the prosecution of sexually transmitted infections (http://www.cps.gov.uk/publications/prosecution/sti.html).
    
    3
    
    Appendix (from Weait, M. Intimacy and Responsibility: the Criminalisation of HIV Transmission, Abingdon: Routledge-Cavendish) 4.1 Causation, HIV and Phylogenetic Analysis8
    
    Given that it is not possible for the defendant to know, as I have been using that term, that he was the cause of the infection which forms the actus reus of the offence under s 20 of the OAPA 1861, resort must be had to scientific evidence that tends towards proving that he was. The evidence that has been used in criminal trials, both in England and other jurisdictions where liability depends upon proof of transmission, is based on what is known as phylogenetic analysis. Simply stated, phylogenetic analysis (‘PA’) is a way of determining the relationship between two samples of HIV from different sources, by establishing their genetic distance from each other.9 Using complex computational methods, it identifies differences between the genetic material in these sources and is thus able to give an idea of the likelihood, or probability, that the two sources are genetically related, and – if so – how close that relationship is. Because HIV, unlike DNA, is not unique to an individual, and because HIV undergoes genetic mutation inside its human host, phylogenetic analysis can only provide an estimate of the relatedness of the samples rather than a definitive match. It does not, and cannot (as we shall see below) provide a certain answer to the critical question in an HIV transmission case ‘Is the virus in the complainant the same as that in the defendant’? Despite these difficulties, phylogenetic analysis has been admitted as evidence in a number of criminal cases, both in the UK and elsewhere, and used by the prosecution to argue that the defendant caused the complainant’s infection.10 The first occasion was in 1992, when defence counsel for a man who had been convicted of the rape of a female complainant and for deliberately infecting her with HIV, requested a virological report on the relatedness of the strains in each party. The conviction was
    
    See, generally, Bernard, 2007; Bernard et al, 2007.. Bernard et al explain that ‘Types, groups, subtypes, recombinants, and quasispecies are scientific terms used to classify different strains of HIV, from the global, regional and country level (types, groups, subtypes/recombinants) to the individual level (quasispecies). There are two types of HIV that infect humans – HIV-1 and HIV-2. Both HIV-1 and HIV-2 are descendants of SIV (simian immunodeficiency virus) found in wild chimpanzees in Cameroon, in western Africa. HIV-1 is the type of HIV that is seen globally, whereas HIV-2 is limited predominantly to the areas around Cameroon. HIV-1 is further classified into three main groups called M, N and O. Again, it is group M that is seen globally, whereas groups O and N are limited predominantly to the areas around Cameroon. group M viruses are again further classified into subtypes (represented by letters of the alphabet, e.g. A, B, C) and recombinants, which are a combination of two subtypes. Recombinants are officially known as circulating recombinant forms (CRF) and represented by a number followed by the two combined subtypes (e,g. CRF01_AE, is a combination of subtypes A and E)’ (Bernard et al, 2007: 10). 10 Before its first successful use in a criminal trial, phylogenetic analysis had been used in the US to determine the relatedness between samples of virus from two dentists and their respective patients. See (case 1) Center for Disease Control (CDC),1990; Cieselski et al, 1992; Ou et al, 1992; (case 2) Jaffe et al, 1994; Myers, 1994. In the first case a CDC review determined that the dentist, who along with some of his patients subsequently died of AIDS-related illnesses, may have been the source of infection (though criminal charges were never brought). In the second it was determined that the dentist was definitely not the source.
    9
    
    8
    
    4
    
    upheld on appeal on the basis of the virological and other evidence. In his account of the way phylogenetic analysis was used, the virologist commented on its limitations: It is important to stress that even though our investigation showed that the strains carried by the male and the female were epidemiologically linked, we could not determine the direction of transmission, nor could we formally rule out the possibility that both the male and the female were infected by a third party. Thus, it was essential that the results from our sequence investigation be used in conjunction with other epidemiological information in the case. (Albert et al, 1994) The first time that phylogenetic analysis was used at trial, rather than as the basis for an appeal, was in the US in 1997. In the case of State of Louisiana vs Richard J. Schmidt11, the defendant – a doctor – was accused of attempting to murder his former mistress by injecting her with blood taken from his patients and which was infected with HIV and Hepatitis C. At a preliminary hearing phylogenetic analysis was ruled admissible in evidence, a decision that was confirmed by the Louisiana Court of Appeal.12 Dr Schmidt was convicted of attempted second-degree murder, and his appeals to both the Louisiana Supreme Court and the US Supreme Court were both unsuccessful.13 Since the Schmidt case, phylogenetic analysis has been used in a number of other jurisdictions (including England and Wales) as evidence tending towards supporting the prosecution’s case in charges relating to the transmission of HIV.14. 4.2 The Limitations of Phylogenetic Analysis Although the identification of specific transmission events with high statistical confidence through phylogenetics will always be subject to significant caveats (e.g., cannot exclude a common, unknown source of 2 similar infections, cannot identify direction of transmission, and ignores the potential of superinfection), as well as ethical concerns (consent for use to identify potential sources of infection), the potential to provide insights into transmission dynamics within communities is now being realized. (Pillay and Fisher, 2007: 924) It was explained earlier that, because of the nature of HIV and its genetic structure, phylogenetic analysis cannot establish definitive proof that HIV positive person X was the source of the infection in HIV positive person Y. All it can do is provide evidence of the degree of relatedness of the viral samples taken from each. This is done by building what is known as a phylogenetic tree – a visual representation of the results derived from PA. A phylogenetic analysis tree diagram provides a way of
    15th Judicial District Court, Lafayette Parish, LA, Criminal Docket No. 73313, Reasons For Ruling of Louisiana State 15th Judicial District Court, Judge Durwood Conque (1997). 12 State of Louisiana vs. Richard J. Schmidt, 699 So. 2d 488, K97–249 LA Court of Appeal, 3rd Circuit (1997); writ denied 706 So. 2d 451, 97–2220 LA (1997). 13 For the virologist’s account of the use of phylogenetic analysis in this case see Metzker et al, 2002. 14 These include an Australian case where the defendant was charged with ‘knowingly and recklessly’ transmitting HIV during the rape of an intellectually disabled man (Birch et al, 2000); a man sentenced to six years imprisonment in Denmark for sexually abusing and infecting a 12 year-old boy (Machuca et al, 2001; and a man prosecuted for raping and transmitting HIV to six women in Belgium (Lemey et al, 2005).
    11
    
    5
    
    demonstrating the relatedness of independently sourced viral strains by representing the extent to which the samples under investigation share common ancestry. Take, for example, the following, simplified, tree:15
    
    Here, viruses B, C, D and E are all descended from virus A, but viruses C and D are more closely related to each other than B and E. All this means is that there is a higher probability that C and D are infected with the same virus than would be a defendant and complainant who were infected with, respectively, viruses B and E (or B and D, or B and C). This raises two further problems, as regards proof of causation in a criminal case, even if we accept the higher probability value of the relatedness of viruses C and D. The first is that even if we have a defendant and complainant who are, respectively infected, with viruses C and D the relatedness of the viruses does not mean that D was infected by C or vice versa (in addition to not being able to provide definitive determination of the source a complainant’s infection, phylogenetic analysis is not able to provide proof of the route of transmission, or an accurate estimation of the timing of its occurrence, either). The difficulty here is fairly self-evident, and can be illustrated by the following, not entirely unrealistic, hypothetical. Suppose C and D start a sexual relationship at T¹. Six months later, C tests positive for HIV, but does not disclose this to D. Some time later, at T², D also tests positive for HIV. D confronts C, and demands to know whether, prior to T², he knew he was HIV positive. C admits that he did. D goes to the police and recounts the narrative from his perspective, alleging that C was the cause of his HIV infection. The police interview C and draw the inference that C was indeed the cause of D’s infection. He is arrested, charged, and phylogenetic analysis evidence is undertaken which shows that the strain of HIV that they each have is closely related. C, on the basis of the chronology of events as I have set them out here, pleads guilty – because he believes the source of his infection is a relationship that happened prior to meeting D, and is sentenced to an immediate term of imprisonment. The problem with this, and the potential for a miscarriage of justice, is easy to see. Why? Because it is perfectly possible that it was D who infected C. The fact that it was C who discovered his HIV positive status first, has nothing to do with – is logically unrelated to – who infected whom. The fact that D identifies himself as the victim, understandably given the moral dimensions of the narrative, establishes a criminal justice logic that is, unless the limitations of phylogenetic analysis evidence are properly understood by all parties, difficult to dislodge. D becomes a viable and credible complainant and is treated as such by the police and the courts, and C becomes a viable defendant against whom it is easy to build a case. If, however, C had suspected that D was the source of his infection he would have been able to confront D with that possibility, and D might then have
    15
    
    This is taken from Bernard et al, 2007.
    
    6
    
    established his (hitherto unknown) HIV positive status. On these alternative facts, it is unlikely – I would suggest – that D would go to the police. Furthermore, it should be noted that C would not have the option of bringing a charge against D even if he wanted to because, assuming that D did infect him, D had no knowledge of his HIV positive status at the relevant time and so cannot be reckless within the legal meaning of that term. A second problem, related to the one just set out, is that there is the possibility that either or both C and D were infected by an independent third party, or by third parties. Where C and D are in a non-monogamous relationship (or where one of them is nonmonogamous) it is possible that neither is the source of the other’s infection, although the narrative as related above might tend towards creating the impression that this is so (either subjectively from C’s and D’s perspective, or objectively as viewed by a jury charged with determining causation). The problem is compounded by the fact that where C and D share a geographically limited pool of sexual partners with HIV, the probability of them being infected independently by the same strain is higher than would otherwise be the case. Phylogenetic analysis is insufficiently accurate to be able to establish whether two people who share the same strain are the source of each other’s infection, or whether both or either have been infected independently by a person, or by persons, also infected with that strain. Of course, it is possible that where there is compelling evidence that D has been sexually faithful to D, and where there is no evidence that D had symptoms associated with HIV infection prior to starting a sexual relationship with C, the fact that phylogenetic analysis cannot by itself establish that C was the source of D’s infection will be less problematic. As explained above, phylogenetic analysis operates within a broader and more complex evidential narrative. However, where C and D are non-monogamous it will, it is suggested, be extremely difficult, for the prosecution to establish guilt on C’s part; and any judge called upon to direct a jury on the weight to which they should give phylogenetic analysis evidence should, it is submitted, be extremely careful both in explaining its limitations and in pointing out the ways in which the chronology of testing and (non)-disclosure in a particular relationship might lead them to draw false inferences about the route and source of transmission. (There is a wide range of possible reasons for C and D being infected with closely related strains of HIV, each of which poses potential difficulties in establishing causation: see Figure 2). Figure 216
    
    (i)
    
    C
    
    D
    
    (ii)
    
    D
    
    C
    
    (iii)
    16
    
    A
    
    C
    
    This is adapted from Bernard et al, 2007.
    
    7
    
    A
    
    D
    
    (iv)
    
    A B
    
    C D
    
    (v)
    
    C
    
    A
    
    D
    
    (vi)
    
    D
    
    A
    
    C
    
    Explanation In these examples, C and D are two individuals with closely related HIV strains. A and B are unknown third parties sharing the same strain. The arrows indicate possible sources and routes of transmission that may explain the reason for the relatedness of the HIV strain that C and D possess.
    
    4.3
    
    The Use of Phylogenetic Analysis in English HIV Transmission Cases, and Best Practice Guidance
    
    Given the limitations associated with phylogenetic analysis evidence, it is perhaps somewhat surprising that its use and interpretation has not been more widely challenged in the prosecutions that have been brought before the courts. In all but one of the cases in England and Wales, defendants have either pleaded guilty or, where there has been a trial on a not guilty plea, their defence has been that their partner(s) consented to the risk of transmission. In each of these cases, therefore, the defendants accepted the phylogenetic analysis evidence that they were the cause of the complainant’s infection. It was not until the summer of 2006 that a defence team sought to challenge the prosecution evidence on this point. At a trial in Kingston upon Thames Crown Court, Matthew Collins had been charged with recklessly transmitting HIV to his partner and pleaded not guilty. There was evidence before the court that his partner had been sexually active with other men, some of whom were HIV positive. The prosecution used phylogenetic analysis evidence to support its contention that he was the source of his partner’s infection but, after initially directing the jury to retire to consider its verdict, the judge directed an acquittal on the basis that they could not, on the evidence presented, be sure that this was so. Commenting on the case afterwards, the expert virologist in the case – Anna-Maria Geretti, who was brought in by the defence to challenge to prosecution case17 – stated:
    
    17
    
    Dr Geretti is one of the authors of Bernard et al, 2007.
    
    8
    
    Virological evidence should be seen in the context of other facts. You should not build a case around this type of evidence alone. There could be a chain of transmission, where four or five people are infected with a similar virus, so it is impossible to tell whether transmission has occurred between two people with a related virus.18 The acquittal in the Kingston case is significant for two distinct reasons. Firstly, it has highlighted the importance of ensuring that where phylogenetic analysis evidence is used, both the evidence gathering process and its presentation / interpretation must be done with the utmost care so as not to mislead the jury as to its significance. There is a number of measures that can be taken to ensure that this is so, and it is important that they are understood.19 1. It must be acknowledged by all participants in a criminal trial that HIV is a virus that exists and migrates within communities, rather than one that infects disconnected individuals. Everyone infected with HIV has been infected by someone else with HIV. Phylogenetic analysis alone (because of its inherent approximations and error rate) cannot prove that X infected Y, though it is able to exclude this possibility. The phylognetic analysis used in criminal cases is, or to date has been, undertaken in research rather than forensic laboratories. This is because very few of the latter have the equipment necessary to undertake it. Where a request for phylogenetic analysis is made to a research laboratory, it is critical that the samples are properly tracked (to eliminate the risk of cross-contamination, and that double-blind testing is carried out (to eliminate bias). (For example, the person undertaking the testing should not know in advance which sample is suspected of being the source of infection.) The samples also need to be blindly tested at two different time points, ideally in different laboratories, and the results should be consistent as between those times. In order to establish the relatedness of the samples, the viral gene sequences that they contain need to be compared to gene sequences in independent control samples. There is a risk, if these controls are taken from populations that are temporally and spacially distant from the test samples (e.g. in relation to two samples from London in 2006 using controls from Australia in 2003), that the test samples will appear to be, and can be represented as, more closely related to each other than they are to the controls. In order to minimise the inference that this means that the defendant (sample A) is likely to be the source of the infection in the complainant (sample B), the controls ought ideally to be taken from a setting that is socially, geographically and temporally relevant to the case under consideration. Thus, if the defendant and complainant share a similar socio-sexual network, the controls should be drawn from that network. This will ensure greater confidence in the relatedness of the test samples where this is apparent, although it does raise complex consent and data protection questions relating to the use of control samples. Where it is not possible to derive controls as indicated, the risk of
    
    2.
    
    3.
    
    18
    
    For further commentary on this case, see O’Connor, 2006. These points are developed from Bernard et al, 2007). For another discussion of the limitations of phylogenetic analysis in a forensic context see Budowle and Harmon, 2005.
    19
    
    9
    
    4.
    
    5.
    
    6.
    
    over-interpreting the relatedness of the test samples as compared with the control samples should be made clear. The choice of phylogenetic tree model constructed for the purpose of demonstrating the relatedness of the samples being tested should be based on its reliability and fitness for purpose and for no other reason (e.g. ease of explanation to a lay jury). If phylogenetic analysis suggests a genetic relatedness between the samples from two individuals, analysis of multiple genetic clones from each can strengthen evidence of that relationship. It is important that phylogenetic analysis involves the sequencing of reasonable length (≥ 500 nucleotides – sub-units of RNA – depending on the particular gene being investigated) of two or more genetic regions. The choice ought to target genes with different biological functions, different evolution rates, and different selective pressures (i.e. genes that are subject to different environmental factors). If using the pol region, it should be recognised that similar anti-retroviral treatment being taken by the individuals whose samples are being analysed can lead to similar mutations in the virus and so give the impression for that reason alone of being closely related.
    
    If these are the practical measures necessary to minimize the risk of injustice in cases where there is the possibility that the defendant was not the cause of the complainant’s infection, and which were highlighted as the result of the Kingston case, the second potential lesson of that trial, and of this science, is of a different kind. It is that it will be far easier to dispute the prosecution’s assertion that the defendant is guilty, all other things being equal, where there is evidence that the complainant was, or could have been HIV positive before starting a sexual relationship with him or her (such evidence including, for example, a history of other sexually transmitted infections), or where the complainant has been non-monogamous during the currency of the instant relationship. Raising evidence of each of these will inevitably be distressing for those complainants who wish to bring a case against a partner (although one might reasonably characterize them as a necessary cost). The more problematic, and politically divisive, issue is the effect of non-monogamy. If, as is seems clear from the Kingston case (and from one other, the details of which are less clear20), sex with others potentially sharing the same HIV strain is a fact that makes it harder to assert with anything like certainty that the defendant is the cause of a complainant’s infection, then it is possible that the CPS will be more likely to prosecute those whose partners have been sexually faithful, and / or who have a clean STI history. The requirement that public prosecutions are only brought where there is a reasonable prospect of conviction means that any credible challenge to the phylogenetic analysis evidence (which a defendant whose partner has been nonmonogamous will be able to do) will tend towards non-prosecution. We are therefore confronted by a crime whose effective prosecution in the courts will be influenced not simply by the availability of strong evidence – that is common to all offences and is unremarkable – but, as it has been in rape, by the sexual promiscuity or otherwise of the complainant. But whereas in rape a woman’s sexual history has typically been used as a way of undermining her credibility as regards non-consent, in the HIV
    20
    
    The case occurred in early 2007 in Preston Crown Court. It would appear that after establishing that the complainant had a sexual partner other than the defendant who was unwilling to provide a sample of blood for testing, the judge threw the case out on the basis that the forensic evidence presented by the prosecution was wholly inadequate (source: personal correspondence).
    
    10
    
    transmission context its effect is rather different. A man may traditionally have found it easy to adduce evidence that his partner’s sexual promiscuity cast doubt on her evidence of non-consent in an act of sexual intercourse with him, but this was not an argument that could be won without the evidence being admitted and explored in cross-examination at trial. It was an argument about the credibility of the complainant’s version of events – not about her biological status, about something amenable to objective independent verification, and is in any event less easy to do now than it once was.21 In contrast, phylogenetic analysis evidence is precisely about the biological ‘truth’ of the complainant, or something approximating it, and is something that can more readily be evaluated by prosecutors in advance of any trial ‘on the papers’. Although it remains the case that other evidence about the complainant, his or her shared history with the defendant, and the nature of their sexual relationship will be important where the relatedness of the samples they give is close, any possibility that the complainant’s infection may have come from a third party is likely to undermine the chance of a conviction and result in the abandonment of the prosecution. References Albert J., Wahlberg, J., Leitner, T., Escanilla, D. and Uhlen, M. (1994) ‘Analysis of a rape case by direct sequencing of the HIV-1 pol and gag genes’, Journal of Virology, 68: 5918-5924 Bernard, E. (2007) Criminal HIV Transmission, London: National AIDS Manual. Cieselski, C., Marianos, D., Ou, C.Y. (1992) ‘Transmission of human immunodeficiency virus in a dental practice’, Annals of Internal Medicine, 116: 798-805. CDC (1990) ‘Possible transmission of human immunodeficiency virus to a patient during an invasive dental procedure’, Morbidity and Mortality Weekly Report, 39: 489-93. Bernard, E. Azad, Y., Geretti, A.-M., van Damme, A.M. and Weait, M. (2007) HIV Forensics: The Use of Phylogenetic Analysis as Evidence in Criminal Investigation of HIV Transmission, London: National AIDS Manual and National AIDS Trust. Jaffe H.W., McCurdy, J.M., Kalish, M.L, et al (1994) ‘Lack of HIV transmission in the practice of a dentist with AIDS’, Annals of Internal Medicine, 121(11): 855-859. Ou, C.Y., Cieselski, C.A., Myers, G., et al ‘Molecular epidemiology of HIV transmission in a dental practice’, Science, 256: 1165-1171. Myers G. (1994) ‘Molecular investigation of HIV transmission’, Annals of Internal Medicine, 121(11): 889-890.
    The position in England and Wales is somewhat different now. A person charged with rape under s 1 of the Sexual Offences Act 2003 is one charged with a sexual offence within the meaning of the Youth Justice and Criminal Evidence Act 1999. Ss 41-43 of this latter legislation place restrictions on the admissibility of, or cross-examination on, evidence of previous sexual history. Unless there is leave of the court, no evidence may be adduced or questions asked in cross-examination by or on behalf of the accused about any sexual behaviour of the complainant. Furthermore, such evidence will only be permitted provided statutory criteria are met and the court considers that it may reach an unsafe conclusion on an issue to be decided in the case if such evidence were not to be heard. Defence counsel are not permitted to engaging in ‘fishing expeditions’ or ask general, character-related questions. Any questions asked or evidence that the defence wish to adduce must relate to a particular event or instance of behaviour on the part of the complainant.
    21
    
    11
    
    Machuca, R. Jorgensen, L.B., Theilade, P. (2001) ‘Molecular investigation of transmission of human immunodeficiency virus type 1 in a criminal case’, Clin Diagn Lab Immunol, 8(5): 884-90. Lemey P., van Dooren, S., van Laetham, K. et al (2005) ‘Molecular testing of multiple HIV-1 transmissions in a criminal case’ AIDS, 19(15): 1649-1658. Budowle, B. and Harmon R. (2005) ‘HIV legal precedent useful for microbial forensics’, Croat Med J, 46(4): 514-521. O’Connor, C. (2006) ‘Law and Disorder’, Positive Nation, 126 (October).
    
    12

Add Comment

To add a comment, you need to login or signup (you can signup without leaving this page)