DNA Fingerprinting: A review of the criticisms of DNA evidence. Is it really the absolute identification evidence?

     Yet even with such a method at hand, the possibility of error -either through human mistake, or through human interference- may not be excluded. That means, that the Courts still will have to value the DNA evidence accordingly to the other facts of a case, and not rely only at the results of the DNA examination.

1.2      Analysis of Samples in the Forensic Laboratory

     Following the collection of the DNA sample from the crime scene, the DNA sample must go through the profiling process in the forensic lab. Suspects DNA samples will be collected and after the end of the profiling technique the profiles from the suspects and from the crime scene will be compared. If there is no match between these profiles then the profile from the crime stain will be compared with other DNA profiles existing in the National DNA database.

     In England the first DNA profiling process, known as RFLP, had been heavily scrutinised. The technical analysis itself was not always accurate and experience has shown that great dangers lie in ambush resulting from its long and complicated nature, as always with new scientific methods;

“The process of DNA profiling starts with DNA being extracted from the crime stain and also from a sample taken from the suspect. In each case the DNA is cut into smaller lengths by specific enzymes. The fragments produced are sorted according to size by a process of electrophoresis. This involves placing the fragments in a gel and drawing them electromagnetically along a track through the gel. The fragments with smaller molecular weight travel further than the heavier ones. The pattern thus created is transferred from the gel onto a membrane. Radioactive DNA probes, taken from elsewhere, which bind with the sequences of most interest in the sample DNA are then applied. After the excess of the DNA probe is washed off, an X-ray film is placed over the membrane to record the band pattern. This produces an auto-radiograph which can be photographed. When the crime stain DNA and sample DNA from the suspect have been run in separate tracks through the gel, the resultant auto- radiographs can be compared.”

     Firstly, there is a possibility that discrepancies will occur in the number and place of the bands on the comparable auto-radiographies of the two DNA profiles. This has been a ground for challenging the reliability of DNA evidence in several cases. Where discrepancies are found in a match, the expert must give a satisfactory explanation with respect to their seriousness.  If he fails so, the court may lead to the conclusion DNA evidence is unreliable.

     In R v Deen the above scenario was demonstrated. Deen was convicted for rape. One of the reasons that made the Court of appeal to order a retrial was that DNA evidence which linked Deen with the rape was inaccurate. In the instant case, a DNA sample was taken from the vaginal of the victim and one from the appellant’s blood. After the two profiles had been compared a match was declared.

     However on appeal, the DNA evidence was challenged by the defence counsel, whose expert argued that there were two bands in the profiles which were discrepant. Although, professor Roberts did explain the discrepancy of the first band to be the result of stringency, he could not however, explain the discrepancy of the other band and therefore claimed that the two DNA profiles produced a mis-match.

     The Court of Appeal in its judgement did not doubt that this appeal would be allowed. Lord Taylor accepted that “Even if a number of bands correspond exactly, any discrepancy between the profiles, unless satisfactory explained, will show a mis-match and will exclude the suspect from complicity.” Consequently, a retrial was ordered.

     Secondly, the phenomenon of degradation may be responsible for inconclusive results. DNA is an organic compound and therefore liable to recycling into the environment. According to Inman and Rudin (1997, p.11) degradation occurs when a DNA molecule breaks into smaller fragments and as a result the bands of DNA are less than the minimum for the needs of the profiling process. Factors that can cause degradation of DNA include time, temperature, humidity, light and the presence of insects.

     As a fact, degradation can have effect on the ability to obtain a useful result from DNA typing. Degraded samples may be responsible for another phenomenon known as “band-shifting.” Accordingly, degraded DNA may not run in the gel at the same speed as it would in ideal DNA samples. This makes the bands shifting up and down in relation to each other. Therefore, even if both profiles have come from the same perpetrator, the band will have different placements on the autobiographies. Nevertheless, even if band-shifting occurs the scientists may still be able to declare a match as soon as the bands have not had a significant shift.

     As it has been presented, the procedure is quite dangerous as there is no critical approach by both lawyers and scientists. As hall (1990) criticises, DNA evidence becomes a black box in which scientific evidence are put from the one side and the verdict is being produced at the other. The reason for such a similitude is that simply both lawyers and jurors may not be able to keep up with the scientific analysis of DNA evidence. The Jurors will have to prepare their verdict by considering the results of a quite complex process.

     Following the Dean, case new methods have been invented to combat the many criticisms of the RFLP process. The new methods are based on the STR technique. This pioneering process looks only at specific areas of the DNA that are known to vary widely between people, and therefore are extremely useful for the purposes of criminal investigation. STR, in antithesis with RFLP, is immune to problems occurring from discrepancies in the profiles or from degradation and therefore it is a more stable and reliable.  

     Nevertheless, there might be the case where the scientists declare a match between a DNA profile from the crime stain and from a suspect who is not the true source of the sample. DNA evidence is not unique for each person and therefore it is possible for two people chosen randomly to have the same DNA profile. Thus, there is a probability that a match can be declared due to random match probability. This probability is calculated by population geneticists from existing DNA databases.

     Furthermore, it has been argued that examiner bias may lead the scientist who compares the profiles to declare a match. Moreover, it might be a person related to the offender. It has been argued that brothers have quite similar DNA profiles and therefore if a scientist does not pay the necessary attention he might declare a match negligently. In a conversation in the House of Commons, Dr. Werrett reported that he had faced instances were a suspect alleged the brothers defence and additionally he said that the suspect had six brothers who were not willing to give a sample.

     Although, as discussed above, the DNA technology has been developed and such errors are not likely to occur frequently, in principle there is still a possibility for such errors as human interpretation is still required. 

     Of course, there might be the case where an innocent defendant is the true source of a match report is the true source of a match. A study carried out carried out by J. Tobasco showed that it is very possible for DNA to be transferred following actions as simple as a kiss. In his study the participants required to kiss each other for a period of 5 seconds and then smoke a cigarette. Analysis of the cigarette butts showed that 67% had of them had DNA from the person they were kissing. This study purports that it is possible that an innocent girlfriend of murdered victim would be incriminated if the victim has kissed her and afterwards has smoked a cigarette before his murder.

1.3 Estimation of Match Probability

     Following the declaration of a match, the scientist must estimate the random match probability or the likelihood ratio which reflects the probability that the DNA profile of a random chosen person matches the crime stain. This depends on the number of the bands that match between two profiles and the genes frequency which appears in the relevant population of such band matching, as it was explained in R v Gordon.

     Nonetheless, the statistical evaluation of a match has been hazardous area of DNA Profiling, up to date. Swenson and Coleman (1995) regard it “as the most contentious debate in respect of DNA evidence”.

     One of the reasons of criticism is that this process is quite complex. The statistical interpretation requires the scientists to conclude into a discriminative result, after comparing the DNA profile with the other samples from the available database. And whilst population studies have shown that there exist significant differences in the frequency of genes between ethnic groups, other studies have shown that there are a lot of similarities in all populations.  However, it has been proved that the differences in the frequency of genes are much more than the similarities.

     Currently in England there are four databases available to the Forensic Science Service with separate sub-sets for the Afro-Caribbean, Asian and Caucasian populations. It has been argued that the systems used to calculate such differences may be biased in favour of the suspect producing unreliable results, where the suspect belongs to a subpopulation. For instance if the defendant is Sierra-Leonean, the afro-Caribbean database might not have enough DNA samples of that ethnic origin to produce an accurate estimation.

     The reliability of the DNA database concerned the court In R v Musa-Gbengba, where a new jury direction was given in respect to the statistical significance of the Crown DNA evidence. Although, the defendant’s origin was Sierra Leonean, the prosecution used an Afro-Caribbean database to estimate the match probability.

     The defence counsel argued that the match probability produced by the general Afro-Caribbean database was inaccurate and therefore its weight should be reduced. This argument was empowered by evidence from the cross-examination of the prosecutions’ expert, who testified that in the chosen database there was no representation of the defendant’s origin. Thus, the Judge gave a direction to the Jury to change the match probability if it accepted that the defence’s submission was correct. Following the Judgement in R v Musa-Gbengba, in R. v. Alan James Doheny, the court ruled that the Forensic Science Service should make available to the defence expert, the databases that were used to estimate the match probability.

PART B “Retention of DNA evidence and the Future”

The police power to retain DNA fingerprint samples has been a matter of concern for the English courts of justice, particularly for challenging the admissibility of DNA evidence. Since 1984, the police powers to obtain samples from persons involved in criminal investigations have been increasing. However, such powers of the police have been heavily opposed by civil liberties supporters. There have also been cases in which the legislative bodies where claimed to be in breach of the right to privacy and the right non-discrimination. Nevertheless, since the establishment of the National DNA database in 1994, the English government tends to support the development of the database.

2.1        The Development of Law

The Police and Criminal Evidence Act 1984 was the first body which enabled the police to take samples from a suspect who was connected with the investigation of an offence. The PACE separated such evidential samples into two categories; the intimate samples and the non-intimate samples.

Following the Pace, the Criminal Justice and Public Order Act 1994 made significant changes to the PACE in respect with the conditions that a sample maybe taken from suspects in order to produce a DNA profile for the purposes of criminal investigation.

Accordingly, the police was given the power to take an intimate sample from a detained person as soon as an officer with the minimum rank of a superintendent had reasonable grounds for suspecting such person for a recordable offence. Recordable offences include all imprisonable offences. It has to be reminded, that ALL offences are imprisonable, with the exception of "obstruction of the Highway" and "breach of the peace". So practically, the police had the power to take samples from any detained suspect person.

.

The 1994 Act also allowed the police to take non-intimate samples from a person detained or in custody with his consent, if the police officer of the minimum rank of the superintendent could reason that the person is suspected for a recordable offence. However the consent of the person was not necessary if the person was charged with a recordable offence or was informed that he will be reported for such an offence. Furthermore the act made provision for the samples to be compared with other samples from crime scenes as soon as the person is informed about the reason that the search is being made.

Non-intimate was defined a sample is a) a sample of hair other than public hair, b) a sample taken from the nail or under the nail, c) a swab taken from a persons body, including the mouth but not any other body orifice, saliva, d) a footprint or a similar impression of any part of a persons body other than a part of his hand.

R v Cooke was a rape case where the issue was identification. The defendant was convicted for rape after a DNA profile created from a hair pulled from his scalp, matched the profile from the crime stain. However, the defendant contended that the judge was in error in admitting the DNA evidence. The ground for his appeal was that the hair sample which the police took from him did not fall within the meaning of the PACE Act 1984 and therefore it would have an adverse effect on the fairness of the proceedings, under s.78 of the PACE 1984.

As a fact, each hair is surrounded by a sheath which consists of leaving tissue. Thus, the appellant argued that whereas a hair is a non-intimate sample within the meaning in the 1984 Act, the sheath is a separate entity and does not come within the acts definition of a non-intimate sample. However the court of appeal dismissed his claim and decided that what is obtained by plucking a hair is “hair” for the purposes of the Act.

Five years later, new changes were made by the Criminal Justice and Police Act 2001 which provided the police with powers to take samples from persons that were acquitted or prosecuted, as soon as the samples have been obtained lawfully. This extension of powers was a direct result of the House of Lords ruling in Attorney-General’s Reference.

In 1997, an old woman was raped anally by a man who broke into her house. A swab was taken from semen found on her body which produced a DNA profile that after was afterwards placed in the National DNA Database. In 1998 D was charged with burglary and a swab was taken from him without his consent. However, although he was acquitted of the offence, his DNA profile was not destroyed. Instead it was placed in the National DNA database and indeed it linked him with the rape of the old woman in 1997.

The man was arrested but he denied of being involved in rape. In his trial the judge ruled that the DNA evidence that linked him with the rape was inadmissible by reason of section 64(3)b of the PACE Act 1984 as amended, which prohibited the use of information derived from a sample ought to have been destroyed. Consequently, the crown did not offer any other evidence and the man was acquitted.

 

The attorney-general referred the question to the Court of Appeal as to whether a judge in such circumstances had the discretion confined to him under section 78 of the PACE to admit the relevant evidence in spite of the terms of section 64.1 of the PACE 1984. Following a negative answer from the court of appeal the question went to the House of Lords. There, their Lordships reversed the decision of the Court of appeal.

Accordingly, it was ruled that s.64(3B)(b) did not specify if evidence adduced as result of a prohibited investigation should be inadmissible, and furthermore it did not make provision for the effects of a breach of a prohibition of investigation. Nonetheless, it was held that subsection 3(b) should be read with section 78 of the PACE which confers the judge a discretionary power to exclude the evidence if it would be unfair. The House of Lords concluded that the question of admissibility was a matter for regulation under national law.

The most recent development was the amendments made by the Criminal Justice Act 2003. The Criminal Justice Act 2003 (Commencement No.3 and Transitional Provisions) Order 2004 provided that from the 5th of April 2004, section 10 of the Criminal Justice Act 2003 should come into force. Section 10 expands more the police powers and provides that the taking of non-intimate samples can be made without the appropriate consent of the person as soon an that person is in police detention following his arrest for a recordable offence.

2.2   The Retention of DNA and Civil Liberties

     In the case of Marper both applicants have been arrested and charged for different offences and DNA samples were taken from them. However, they were not convicted and requested their DNA profiles to be destroyed.  Nevertheless, the Chief constable of the south Yorkshire police station refused to destroy the samples and acted with regard to s.64 of the PACE 1984. Both parties applied for judicial review, citing s. 6 of the Human Rights Act 1998, that s.64 of the PACE had to be interpreted in a compatible way with article 8 and 14 of the European convention of Human rights.

     Under article 8(1) every individual has a right to private and family life, home and correspondence. DNA is a biological compound which contains personal genetic information about each person’s appearance and health. Therefore it was accepted by Lord Wolf that there was a concern under article 8(1). The police would continue to retain the DNA profiles, only if the circumstances of retention were justified under the principle of proportionality in accordance with article 8(2).

     The Select Committee agreed however that the retention of samples according to the PACE were not in breach of rights to private life or the right to non-discrimination.

     As it has been analysed, in part A of this work, DNA evidence offers great benefits to our society. DNA evidence is a powerful identification which helps police to resolve serious crimes and also it has been helpful in discovering previous miscarriages of Justice.

     On the other hand, Mahendra had also argued that the police, by starting the search for culprits from the DNA records rather than go into the world, misunderstands fundamentally the purposes of an efficient and fair criminal investigation.

     In England civil liberties organisations have been disturbed by the Judgement in Marper. In the United States the same policy to obtain non-intimate samples from persons that have been arrested, for the purposes of DNA profiling has been applied since 1999.  Civil libertarians from the American Civil Liberty Union have expressed their opposition to such policies.

     Nevertheless, the Court -after weighing the benefits of DNA evidence to a democratic society, against the protected rights of the defendants- dismissed the application holding that despite there was an interference with one's right to privacy, under article 8(1), the intercourse was justified under art 8(2). Therefore, the police had the right to retain DNA samples, as soon as they were used for the prevention and detection of crime, the conduct of a prosecution or the investigation of a crime.

     The other issue that was considered was the compatibility of the PACE with the right to non-discrimination. The applicants in Marper argued that s.64 of the PACE categorised innocent individuals into two separate categories, those that have been absolutely innocent and those that been suspected, but at the end they were found innocent. They argued that such a diversification was incompatible with the presumption of innocence, a fundamental principle and right of the defendant.

Lord Woolf and Lord Waller held that the provisions did not discriminate between the two categories of innocent people. They ruled that there was “a clear objective distinction” between those that had been suspected but at the end were found innocent, and those that were absolutely innocent. However Lord Sedley had different opinion. He was of the view that there was discrimination between the two categories, but that such discrimination was objectively justified, provided that data would have been destroyed in cases in which investigation shouldn’t be initiated in the first place.

      It seems that discrimination would be no issue once the day comes, when every member of a population has been sampled on a DNA database; Sir Alec Jeffreys on a comment at the British Association's science festival in Leicester stated that “if we were all on the database, we’re all in exactly the same boat – the issue of discrimination disappears”. His other concern was, that the police looks at information on the physical characteristics of the profiles to find a clue on the appearance of the suspect, where no other evidence is available. With respect to that, Sir Jeffreys expressed his discomfort by saying that “The use of this sort of very private genetic information by the police does fill me with very considerable concern."

     Mahendra agrees that the only response to the concerns of discrimination will be to make a truly representative database, in which all the citizens’ DNA profiles will be included and everyone shall have a form of a national genetic identity card.

Part C “Inexpert Experts”

3.1 Fingerprint or Profile?

     When the application of DNA Profiling in criminal identification was discovered, Sir Alec Jeffreys termed it as “DNA fingerprint”.  However, this term was misleading, for the reason that it equated the power of DNA evidence with the Fingerprint evidence. The public and more specifically the jurors may have taken a wrong impression in respect of DNA evidence before their arrival in the court room.

        

     Although the media and press ought to use the term fingerprint more frequently to describe the DNA profiling process, the former differs significantly from our well known fingerprint. Fingerprint identification has been accepted by the courts world widely for more than a hundred years. On the other hand, DNA profiling is a new process and has generated much debate. Whilst fingerprint relies upon a straightforward lift of prints from surfaces and their subsequent examination – a process with a few steps and very little potential for error, DNA profiling is a complicated and long scientific process, which as it has been evaluated, has a great potential for error. In addition to that, another significant difference is their discrimination power. Whilst dermal fingerprint is unique for every individual and an expert when presenting it in court will state that he is certain that a mark in the crime scene was made by its originator, the same DNA profile ca be found in more than one individuals and therefore the expert will only give a numerical statement known as match probability.

     It has been argued that events that happen outside the court may influence jurors’ judgements. For example, a juror may read newspaper article, make a relevant discussion or observe an incident which will make him see things differently. This responds to the philosophical question whether it is possible for a juror to completely isolate himself from the influences induced to him by his upbringing, his "social class" and his surroundings, which the leading opinion in the philosophy of law teaches us is impossible. The juror could be affected by the above influences and it is not possible to deny them completely, although he may try to deny them as much as to be accepted by the society as an independent juror. As a fact, these events will vary unsystematically among jurors and may be responsible for unpredictable differences in judgement that cannot be discovered at voir dire.

     Generally, even in recent days, the media prefers to use the term of Fingerprint rather than profile when describing DNA evidence. The Independent published an article after the multiple trains bombing in Madrid that suggested that the victims could be identified by a process described as “DNA Fingerprint Evidence.” Furthermore, the Guardian used the term DNA fingerprinting to describe the identification technique that was used to clarify the identity of Sadam Husein after his capture. Moreover, academic writers have published various books, which include in their title the term “Fingerprint” when referring to DNA profile evidence.

     It seems that the term “Fingerprint” is quite dangerous and it might have prejudicial effects to the defendant, especially when his conviction is truly based on DNA evidence. Although, Forensic science avoids to use this terminology for DNA evidence, other sciences such us molecular biology use it quite commonly. Given that the term “fingerprint” has been and will be used in the future, for the shake of the defendant, appropriate measures must be taken to distinguish in the minds of the Jurors what it really means.

3.2 Bayes Theorem

     From the introduction of DNA evidence as an identification tool in criminal trials, the accepted method that the Jury had to follow in order to compare DNA evidence with non-scientific evidence was based upon Bayes’ theorem. However there has been great debate concerning its use in the criminal trials.

     On the one hand Alvin Goldman (1986) suggests that our doxastic attitudes are frequently better represented by binary beliefs rather than by probabilities. Additionally Finkelstein and Fairley (1970) propose that jurors should use Bayes’ theorem to aggregate pieces of information in some cases to evaluate identification evidence On the other hand, however, Zabell opposes to that proposal and argues that the technical Bayesian approach is more likely to mislead then to assist a lay jury.

     Furthermore, Tribe (1971) objects to the use of statistical evidence in jury trials. He asserts that if the uncertainties involved in evaluating evidence were exposed and made prominent at trial, it would be undermining for the both the jury and society’s belief that proof “beyond reasonable doubt” could be achieved.

     With respect to the Bayes theorem, the jury has the difficult task to evaluate the evidence which need to be compared with the DNA evidence in the following mathematical formula:  

[  P(G|E) :  P(nG|E)  ]     =     [  P(G) : P(nG)  ]     x     [  P(E|G) : P( E|nG )  ]

<=>

Posterior Odds               =          Prior Odds        x        Likelihood Ratio

     The first step for the Jurors is to estimate the probability that the defendant is guilty from the evidence drawn to their attention, before the likelihood ratio of the DNA was given to them. This result would represent the Prior Odds in the Bayes theorem. In the next step they will be given by the expert the likelihood ratio or the match probability of the profiling process. The product of Prior odds times the likelihood ratio will be the Posterior odds showing the defendants guilt after the incorporation of the probability given by the DNA evidence.

     If there is more evidence that need to be considered, the jury will have to apply the Bayes theorem separately each time. This means that if there were five pieces of evidence which had to be compared with the DNA evidence, then the Bayes theorem would have to be applied twenty times.

     It is very clear that Bayes theorem is based on a complex mathematical formula. Zabell argued that generally, human behaviour is not able to estimate the prescriptions of Bayesian model. He also commented that the Bayesian probability model provides certain consistency constraints on probabilities that may help us police our beliefs, but the theory says nothing about where probabilities come from in the first place. Furthermore, studies by Tversky and Kahneman have concluded that when people make assessment of probability they commit serious mathematical errors.

     Moreover Mark Kaplan has remarked that the Bayesian conception of subjective probability presumes that our mental architecture contains the equivalent of a black box stocked with subjective probabilities just waiting to be elicited.

     A similar approach took the Court of Appeal in R v Adams (Denis John) , where the issue was identification. In this case the prosecution relied entirely on DNA evidence.  The prosecution expert gave evidence that the chance of a randomly selected individual to match the crime stain was one in two hundred million. However the victim failed to identify the defendant on an identification parade, and furthermore at the committal of the proceedings she said that the appellant did not look like the man who raped her. On the other side, the appellant had an alibi which was also supported by his girlfriend.

     The prosecution expert alleged that the most appropriate method to compare non-DNA evidence with DNA evidence could be made using Bayes Theorem. As mentioned above, by using this method, the Jurors would have to apply the theorem several times, incorporating evidence such us the likelihood ratio that the perpetrator was a local man and the likelihood that the defendants alibi was true.

     After considering the defence arguments the court allowed the appeal ordering a retrial for several grounds. Firstly, there was a fear that the objectivity which encompasses the numerical figures that are used when applying Bayes theorem, might hide the element of judgement on which the calculation was based. Secondly, Bayes theorem was deemed to rigid for the jurors given that they had to assess the evidence separately. Thirdly, it was inherited that jurors must evaluate evidence by joint decision using their "common sense" and "experience of the world" and not by the means of an automated formula, since this would deny them the free human will. Fourthly, it was accepted that Jurors would have different figures to be attached to the evidence and therefore they wouldn’t be able to alter their views individually at a later stage if they decided so. The court concluded that the introduction of the Bayes theorem into a criminal trial would plunge the jury into inappropriate and unnecessary realms of theory and complexity which would have a negative effect in their task.

        

     Moreover it has been argued that the concept of Prior Odds is incompatible with the presumption of innocence. The presumption of innocence means that a defendant at the beginning of the trial is innocent. This mathematically expresses that the prior odds in the beginning of the trial would be zero. In fact this would render the process useless given that when a number is multiplied with zero equals zero. Nevertheless, the first prior probability usually for the sake of the process is a small figure close to zero commonly the likelihood ratio of the population that leave around the area of the crime. Even if this figure is not reflective a common response is that over time the evidence will wash out the differences in initial starting points.

     On the other hand, Jowett (2001) argues that the courts should accept Bayes theorem as the one mathematically sound method of fully understanding the implications of DNA matches. In order to attack the above criticism he suggests that the court should implement the Bayesian networks approach, which despite being based on Bayes theorem, it has the advantage that it bypasses the complex workings of Bayesian probability.

     Nevertheless, Jowett in continuance parallels Bayesian networks with computers. Just as people use computers without knowing exactly how they work, he believes that we too can use Bayesian networks. However his suggestion was inconsistent with the view of the court in Adams in respect of Bayes theorem. Using a Bayesian systems approach wouldn’t give the opportunity to jurors to alter their decisions.

3.3 The Prosecutor’s fallacy

     After a match has been found between two profiles, two distinct questions might be asked in the court. Firstly, what would be the possibility that an individual would match the DNA profile from the crime sample given that he was innocent and secondly what would be the possibility that an individual was innocent, given that he matched the DNA profile from the crime sample?

     If the expert answers the first question with the answer of the second question, then we have the phenomenon of the “prosecutor’s fallacy.” In other words, the prosecution’s fallacy is the danger of presenting the DNA evidence concerning the frequency of occurrence, the rarity of a profile, as a chance of the suspect not being the attacker.

     Where the prosecutor’s fallacy occurs, it will result in confusing the Jurors, in their estimation of probability. If the Judge in a trial concerning the interpretation of DNA evidence does not notice that the expert has committed the prosecutions fallacy in his testimony, it is almost certain that by in summing up the Jury will draw wrong inferences about the DNA’s weigh and therefore a verdict based on such factors will have prejudicial effects for the defendant.

     The prosecutor’s fallacy was the second ground for which the appeal in R v Deen was allowed. The expert of the prosecution, Mr. Michael Davey gave evidence that the probability that an individual chosen at random would match the crime scene sample was one in three million. However the court of appeal accepted that both the expert on cross-examination and the Judge on his summing up have been trapped by the prosecutor’s fallacy and ordered a retrial.

     Moreover, in the case of Doheny and Adams the court of appeal accepted that the both the expert and the Judge in his summing up have pronounced the prosecutor’s fallacy and allowed therefore the appeal. Indeed, the court introduced the guidelines for the proper procedure of interpreting DNA evidence in a criminal trial.

     Accordingly, the scientist should adduce in court the evidence arose from the DNA analysis along with his calculations of the match probability. He must also give to the defence expert any information needed to scrutinise the basis of the calculations, like the databases used. The expert must also explain to the jury the nature of the matching between the profile from the crime scene and the profile of the defendant and then give them the likelihood ratio. If necessary, the expert may also indicate how many people are likely to be found to have the same profile found in the crime scene, in the United Kingdom or in a more limited relevant sub-group, for instance the Caucasian.

     Afterwards, the Jury has to decide if it was the defendant who left the crime stain or someone else, taking into account all the relevant evidence. However, the expert  under no circumstances he may be asked his opinion or use such terminology in respect of the likelihood ratio that it was the defendant who left the crime stain, which may lead the jury to believe that he was expressing such an opinion.

     The judge has to give careful directions when summing up and explain to the jury how to use the likelihood ratio in arriving to their verdict. He may also draw attention to all relevant evidence which built the significance of the likelihood ratio. Judges should adapt a direction similar to this: “Members of the jury, if you accept the scientific evidence called by the Crown, this indicates that there are probably only four or five white males ion the United Kingdom from whom that semen stain could have come. The defendant is one of them. If that is the position, the decision you have to reach, on all the evidence, is whether you are sure that it was the defendant who left that stain or whether it is possible that it was one of that other small group of men who share the same DNA characteristics.”

     Last year, however, in the case of Pringle V the Queen the prosecutors’ fallacy was again pronounced. The appellant was convicted for murder and had received the mandatory death sentence. In that case the prosecution heavily relied on DNA evidence whilst the defendant did not have an expert to counter the DNA evidence.

     The expert in the trial asserted that DNA sample “could have come from the appellant” and also that “she was 99.999 per cent certain” for that. On appeal, the judicial board ruled that there were sufficient errors and inaccuracies both in the experts’ statements and in the judges summing up, for the jury to have been seriously misled. Thus, their lordships rendered the conviction unsafe and remitted the case to the Court of Appeal of Jamaica to consider whether there should be a retrial.

Conclusion

     This work has reviewed some of the criticisms of DNA evidence. It has showed that from the very early steps when the evidence is collected for the purposes of the criminal investigation, DNA samples maybe contaminated either by accident or deliberately. Following its arrival to the laboratory, the profiling procedure, despite the great development of science, is not always producing accurate and certain results. Namely, the complex nature of the process and the uncertain estimations of likelihood ratios resulting from the population databases blur more the picture of DNA evidence, before it arrives in the courts of justice.

     Furthermore, when the time comes to be presented in the criminal courts the inexpert jurors will have to evaluate a complicated mathematical formula known as the Bayes theorem by which they will decide if the defendants’ guilt is proved beyond reasonable doubt. It has been proved that such a complex formula it will more confuse than help the inexpert jury.

In addition to that, many of them will come to court with wrong inferences about the power of DNA evidence which may affect their verdict.

     Nevertheless, it would be unfair to forget the benefits that DNA evidence offers to criminal justice as an identification tool. Many who have been convicted for crimes that haven’t done, as soon as DNA evidence became a tool in the hands of justice was used to prove their innocence. In the opposite scenario, others who were responsible for serious crimes have been convicted.

     Thus it would be unjust to consider only the negative criticisms of DNA evidence in this conclusion and reject one of the greater developments of identification science to date. However, it would also be excessive and unreasonable to render it as “the absolute identification evidence”. One day this characterisation may be well grounded, given the rapid development of science. Today, however, as it has been showed DNA evidence is still a well-disputed subject.

     When juries create their verdicts from DNA evidence, the English courts can impose the life sentence while the American courts can impose the death penalty. Therefore, defence counsels need to examine DNA evidence with great care and from different aspects in order to provide the maximum for their clients. Of course, it is essential that the appropriate measures must be taken to educate the jurors in order to achieve the a reasonable standard in order to give the appropriate weight to DNA evidence and be able to realise if the defendants guilt is proved beyond reasonable doubt or at least examine if they are able to correspond to the needs of the evaluation of such complicated evidence.  

It has to be admitted, that the very high accuracy rates that DNA evidence achieves, among with the fact that there has to be sound and severe scientific background to understand the procedures in DNA evidence collection, from the crime scene and all the way to the Court, is a very comforting way to rely solemnly on this evidence and deny the importance of i.e. a live witness or other evidence, which may be considered subjectively affected.

This is considered by the conductor of this paper as the most important consideration, since this would not be the first time that a scientific method is presented to the public and to the legal science as the ultimate method to solve all cases and to withstand any objection as to its perfect results.

There has to be reminded, for instance, the suggested perfection of the Bertillon anthropometric data collection system, the French predecessor to the fingerprint analysis. It was supported that there cannot exist two different persons with the same Bertillon dimensions (although the possibility was admitted for twin brothers or sisters). And though, in the U.S. there existed two such persons who could only be told apart by fingerprint examination.

Another example is the so-called "Lie Detector" device which surprisingly aroused itself many questions about its eligibility as evidence, primarily because it would deny the defendant his right to lie or not tell the truth in order to protect himself from an accusation.

So the conclusion is that, it has to be reminded that the DNA evidence -once it is accepted in a case or a Court as a reliable method- will only give information about whether physical presence or physical contact has taken place. But the interpretation of such a statement, and the connection of such a statement to the other evidence in a case, will be the task of the Judge and/or the Jury.

BIBLIOGRAPHY

BBC News (2002) “Privacy Fears over DNA database” Thursday 12 September 2002, BBC News from http://news.bbc.co.uk/1/hi/in_depth/sci_tech/2002/leicester_2002/2252782.stm

Bourne C, (2002) “Retaining fingerprints and DNA samples” New Law Journal Vol. 152 No.7055 page 1693

Burn, S. (1999), “The Expert Witness in Court” Shaw & Sons, 2nd edition

Corker, D. (1996), “Disclosure in Criminal Proceedings” London, Sweet & Maxwell

Clayton and Dennis (2003), “50 Years of DNA” Nature Publishing

Dixon C. (2000),  “Super DNA test that spots any brush with law” Sunday Express, May 14

Ebisike, Norbert. (2000), “An Appraisal of Forensic Science Evidence in Criminal Proceedings” Greenway Publication

Evett I. (2000) “DNA Profiling: A discussion of issues relating to the reporting of very small match probabilities” Criminal Law Review p. 341

Farrington D. (1993), “Unacceptable Evidence” New Law Journal Vol. 143 No 6604 p 857

Finkelstein M and Fairley  (1970) “A Bayesian approach to identification evidence”, Harvard Law Rev Vol. 83 page 489.

Freeman, M. and Reece, H. (1998) “Science in Court”, Athenaeum Press

Hall A.  (1990), “DNA fingerprints – Black box or black hole?” New Law Journal, Vol. 140 No. 6443 p. 203

Hastie, R. (1998) “Inside the Juror – The Psychology of Juror Decision Making” Cambridge University Press

Inman K. and Rudin N. (1997) “An Introduction to Forensic DNA Analysis” 2nd Edition Hardcover, CRC Press

Jowett, C. (2001) “Sittin’ in the Dock with Bayes” New law Journal Vol. 151 No. 6971 p.2001

Kerr, N. and Bray, R. (1982) “The Psychology of the Courtroom”, Academic Press

Lachter K, (Unknown) “Science and the Law: the Implications of DNA Profiling” Dartmouth University from

Lee, H. & Tirnady, F. (2003) “Blood Evidence, How DNA is Revolutionizing the Way We Solve Crimes” Perseus Publishing

Liberty Human Rights Organization Press Release (2002), “Databasing the DNA of innocent people - why it offers problems not solutions” from http://www.liberty-human-rights.org.uk

Mahendra (2002) “Citizens, DNA, Civil Liberties” New Law Journal Vol.152 No.7049 page 1405

Mahendra, B. (2003) “Doc Brief” New Law Journal Vol.153 No.7078 p.665

Mason, D. (2002) “The last word on Hanratty” New Law Journal Vol. 152 No.7033 p.777

Murphy, P. (1998) “Evidence & Advocacy” Blackstone Press Limited 5th edition

Murphy P. (2003) “Murphy on Evidence” Oxford University Press, Eighth Edition

Public Accounts Committee Publications (1998-1999), seventh Report from

Reid Hastie (1983) “Inside the Jury”  Cambridge Publishing

Select Committee on Public Accounts (2003),  Minutes of Evidence from

Swenson E. and Coleman H. (1995), “DNA in the Courtroom: A Trial Watcher’s Guide” 1st Edition Genelex Publication

Tambasco A.  (2004), “When is a kiss just a kiss”  from magazine Profiles in DNA Volume 7, No. 1 from

The Forensic Science Service (2002), Fact Sheet on DNA LCN, 2002 from

The Journal of Criminal law (2002), Vol. 66 No. 4

The National DNA Database Annual Report 2002-2003 from

Townley L. and Ede R. (2004), “Forensic Practice in Criminal Trials” The Law Society

Tribe L. (1971) “Trial by Mathematics: Precision and Ritual in the Legal Process Harvard Law Review Vol. 84  No.1329

United States Civil Liberties Union “ACLU urges Attorney General to adopt panel's opposition to Mass DNA testing” from

CASES

R v Deen,  The Times, January 10, 1994 1993 WL 966047

R v Gordon [1995] 1 Cr App R 290

R v Doheny and Adams [1997] 1 Cr. App. R. 369

R v Adams (No.3) [1997] 1 Cr. App. R. 369

R v Musa Gbengba [1997] 1 Cr.App.R. 367.

R v Cooke 1995 1 Cr App R 318

Attorney-General’s Reference (No.3 of 1999) Re [2001] Crim.L.R. 394

R (on the application of Marper and another) v Chief Constable of South Yorkshire

Pringle V the Queen [2003] WL 116971

R v Adams (Denis John)  [1996] Crim. L. R. 898.

R. v Shirley 2003 WL 21554724

R v Hanratty [2002] 2 Cr. App. R. 30

---