Does English law need a statutory reliability test for  expert evidence in criminal proceedings? Reliability test in other jurisdictions have not successfully dealt with the thorny issue of reliability of expert evidence.

                  DR KS Dhillon. FRCS, LLM

        (1)Introduction

                                  

No criminal justice system can guarantee that miscarriage of justice will not occur but it can put in place procedures that will reduce the risk of wrongful convictions. The reality is making difficult choices between competing values. In the late 1980’s and early 1990’s a number of high profile miscarriages of justice came to light including the Birmingham Six, the Guildford Four, the Maguires, the Tottenham Three , the Cardiff Three and the Taylor Sisters.[1]  Two public inquiries were established to improve the English justice system and restore public confidence.[2]In fact on the same day as the conviction of the Birmingham Six was quashed, the Home Sectary, Kenneth Baker announced the establishment of the Royal Commission on Criminal Justice (Runciman Commission, 1993).The Commission had far reaching terms of reference but since the forensic evidence and expert testimony played a prominent role in several miscarriages of justice, one of the terms of reference was to address the issue of forensic science and expert evidence. It recommended improving the forensic science services by exposing it to market principles and the setting up of a Forensic Science Advisory Council. It also recommended a ‘modified and improved adversarial procedure for preparation and presentation of scientific evidence’.[3]

The May Inquiry (1994) into the circumstances surrounding the miscarriage of justice in relation to the Guildford Four and Maguire Seven raised considerable doubts about the forensic science evidence. The report ‘apportioned most of the responsibility to individuals, particularly the individual scientist’.[4]This individual failing according to the report was not due to weakness or fault of the criminal justice system and no rules in the system could provide complete protection from these failings.

More recently miscarriages of justice in Clarke (2003) and Cannings (2004), which sparked a public outcry, were also widely attributed to failings of the expert witness testimony. The House of Commons Science and Technology Committee lamented that the expert evidence in criminal proceedings was being admitted too readily and without sufficient scrutiny and the committee called for reforms.[5] In response the Law Commission first carried out a consultation in 2009[6]  and subsequently published its report, ‘Expert evidence in criminal proceedings in England and Wales’ in 2011.[7] Attached to the report was a draft bill which recommended a statutory reliability test for admission of expert evidence in criminal proceedings.

Reliability test in other common law jurisdictions have not successfully dealt with the thorny issue of expert evidence reliability. There have been claims that reliability tests in many common law jurisdictions have failed to keep incriminating scientific evidence out of the courts.[8] The criminal justice system’s dilemma is its heavy reliance on forensic science evidence. Unfortunately many kinds of forensic science evidence, with the exception of DNA evidence, lack the scientific foundation which is necessary for the formulation of useful reliability tests for admissibility of expert testimony. Contrary to what many believe even published and peer review studies have weakness which have been highlighted in recent years.

This article aims to analyse and highlight some of the current weaknesses in the state of law on admissibility of scientific medical and forensic evidence in criminal proceedings in major common law jurisdictions. A review of the current scientific and forensic evidence proffered in criminal trials reveals that there are many inherent reliability issues which make it difficult for the courts to approach admissibility requirements based on such evidence. Implementing strict reliability requirements would deprive the courts of useful relevant expert evidence which has yet to come up with methods of determining reliability. Furthermore strict reliability test may prevent the courts from keeping pace with scientific development. An analysis of some of the high profile miscarriages of justice shows that tightening the rules of admissibility of scientific/forensic evidence is unlikely to resolve the perennial problems associated with unreliable expert evidence. These weakness need to be addressed by other reforms. Tightening the rules of admissibility of expert evidence may reduce pragmatism, flexibility and judicial discretion which are essential for the law to develop in tandem with the development of the sciences.                     

                             

     (2) Admissibility Standards for Expert Evidence

The ‘relevancy test’ for expert testimony is consistently used in all common law jurisdictions. The ‘reliability test’ for admissibility of scientific evidence has been in existence in Australia since 1912, in the US since 1923 and in the UK since 1953, yet the rules of admissibility remain unsettled and are mired in controversies. The legal community is struggling with the issue of admissibility standards for expert evidence testimony in criminal proceedings. The United States with federal statutory reliability standards has not been able to steer clear of inconsistencies and controversies.

           (A)Standards in England and Wales

The principle of admissibility of scientific opinion in English law dates back to 1782[9]. The test for admissibility of expert evidence was defined in R v Silverlock [1894] 2 Q.B. 766.This case concerned the admissibility of hand writing expert testimony. Here the test of admissibility in relation to expert evidence was possession of the required skill (expertise) and the expertise need not be obtained by formal qualification. Whether the witness is an expert in the required field is for the tribunal to decide and for the witness to prove his expertise.[10] Bingham L J in R v Robb[11] reaffirmed this test of admissibility.[12] Bingham L.J went on to say that expert opinion can be given not only in established field of science but also in other areas such as handwriting, finger printing, and accident reconstruction as well as in literary fields of work. However the court will not admit the evidence of ‘an astrologer, a soothsayer or an amateur psychologist’. [13]The ‘relevancy test’ for admission of expert testimony was defined in R v Turner [1975] QB 834, where Lawton L J said,

[a]n expert's opinion is admissible to furnish the court with scientific information which is likely to be outside the experience and knowledge of a judge or jury. If on the proven facts a judge or jury can form their own conclusions without help, then the opinion of an expert is unnecessary.

The English Courts have been quite liberal in admitting incriminating expert opinions without consistently requiring ‘reliability’ as a prerequisite for admission of the evidence.[14] This has been seen in a number of cases in relation to facial mapping.[15] Issues such as unreliability, exaggeration or misrepresentation are of probative value. These are left to the tribunal of fact to take into consideration when deciding on the ultimate issue. In fact, Gage L J in R v Harris said that ‘… developments in scientific thinking should not be kept from court, simply because they remain at the stage of hypothesis’.[16]

Though English law has no ‘gate keeper’ test of reliability for admissibility of scientific evidence, the courts have inconsistently applied various reliability test by ‘drawing vague analogies with approaches taken in other jurisdictions and adopting statements of principles provided by foreign courts’.[17]In R v Gilfoyle[18], Rose L J rejected Professor Canter’s ‘psychological autopsy’ of an alleged murder victim, whom the defendant claimed had committed suicide, although he was ‘clearly an expert in his field’, because the evidence ‘tendered was not expert evidence of a kind to be properly placed before the court’.[19] The Court in rejecting Professor Cantor’s testimony referred to United States v Frye [20] which requires expert testimony to be based on evidence ‘which must be sufficiently established to have gained general acceptance in the particular field in which it belongs’.[21]

The Courts of Appeal in R v Luttrell[22] and R v Ciantar[23] did place emphasis on the need for reliability of expert evidence as criteria of admissibility. Rose L J in Lutterll, while admitting expert evidence on lip reading, spoke of the authority which would make the expert witness’s opinion admissible.  He quoted King C J in The Queen v Bonython[24], that the subject matter of the expert’s opinion should form ‘…part of a body of knowledge or experience which is sufficiently organised or recognised to be accepted as a reliable body of knowledge or experience…’.[25] In Ciantar, Moses L J in admitting evidence on facial mapping reiterated that ‘the subject matter ….  must be part of a body of knowledge or experience recognised to be a reliable body of knowledge’ and ‘[s]econdly, the witness must be qualified to express an opinion by reason of his special acquaintance with that body of knowledge’.[26] Bonython though an Australian case has been widely cited in English law with regards to admissibility of expert evidence.[27]The Bonython approach though widely cited in England and Wales, it is not binding in English law and neither is it universally accepted, even in Australia. In R v Parenzee[28] the court explained that the principles enunciated in Bonython incorporated the Frye acceptance test. So has the Australian Law Reform Commission in its review of Uniform Evidence Acts affirmed that the law in South Australia incorporated Frye general acceptance test.[29]In Gilfoyle the court did refer to the Frye test in rejecting admission of expert testimony. However the court in Dallagher[30] rejected the Frye test stating that the Frye test has been superseded by the Federal Rules of Evidence.[31] There seems to be no uniformity in English courts regarding admissibility of expert evidence. The prevailing practise in English courts regarding reception of expert evidence appears to be one of pragmatism and flexibility to allow the court to ‘enjoy the advantages to be gained from new techniques and new advances in science’.[32]The Court of Appeal in R v Reed and Reed[33] has indicated the current common law stand on admissibility of expert evidence in the following words:

 There is …no enhanced test of admissibility …. If the reliability of the scientific basis for the evidence is challenged, the court will consider whether there is a sufficiently reliable scientific basis for that evidence to be admitted, but, if satisfied that there is a sufficiently reliable scientific basis for the evidence to be admitted, then it will leave the opposing views to be tested in the trial.[34]

In England and Wales there is no reliability test for admissibility of expert testimony. The courts use the ‘relevancy test’ for admissibility of such evidence, where the courts decide whether the evidence is relevant, the witness is a qualified expert (Silverlock test) and the evidence will assist the trier of fact (Turner test). Once the evidence has passed the ‘relevancy test’ and has been admitted, than it is up to the jury to determine the reliability of the evidence through adversarial cross examination. This can create problems, when the opposing party does not have sufficient understanding to ask the relevant questions during cross examination[35] or when the expert testimony is too complex and overwhelming for the jury and this may potentially lead the jury to place too much reliance on such evidence.[36]

What is sufficiently reliable scientific evidence remains undefined and is left to the judge to decide. The courts have placed much trust in the adversarial system to weed out unreliable evidence notwithstanding the fact that the NAS report has expressed doubts that such a system is ‘suited to the task of finding “scientific truth”’.[37]  

          (B) Standards In USA

A common law reliability test for admissibility of novel scientific evidence was first introduced in the US in 1923. In Frye v United States[38]the court ruled that a defendant in murder trial cannot use a precursor of the lie detector test as exculpatory evidence. The court defined the test for admissibility of scientific evidence:

Just when a scientific principle or discovery crosses the line between experimental and demonstrable stage is difficult to define. Somewhere in this twilight zone the evidentiary force of the principle must be recognised, and while the courts will go a long way to admitting expert testimony deduced from a well-recognised scientific principle or discovery, the thing from which the deduction is made must be sufficiently established to have gained general acceptance in the particular field to which it belongs.[39]

Frye came to be widely known as the ‘general acceptance’ rule which prohibited admission of expert evidence unless the scientific evidence was generally accepted by the scientific community in the relevant scientific discipline. The courts were left to decide who the relevant scientific community was and also to decide whether the scientific community accepted the scientific principle.[40]Historically, according to Robert  Kohar, in some jurisdictions such as in Illinois where Frye is strictly followed, ‘it is only  necessary that the expert make the statement or offer evidence that the methodology employed is accepted in his or her discipline’ for the court to admit such evidence in most  cases.[41]The Frye rule was often applied in criminal cases and less often in civil matters and in testimony involving ‘soft or social’ sciences because it was more difficult apply in these areas.[42]The problem with Frye acceptance rule was that new novel scientific principles although sound were kept out of the courts especially in criminal cases.[43]

In 1975 when the United States Federal Rules of Evidence were introduced the issues relating to Frye were not addressed. The Federal Rule 702 (1975) regarding testimony of expert states that,

[i]f scientific, technical, or other specialised knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training or education may testify thereto in a form of opinion or otherwise.[44]

After 1975 majority of the courts continued to use the Frye general acceptance rule, but soon a split in the federal and state courts began to appear with some using the Frye test and others different versions of the Federal Rule of Evidence.[45]The lack of clarity and confusion over the relationship of the Frye test and Federal Rule of Evidence, prompted the Supreme Court in Daubert v Merrell Dow Pharmaceuticals Inc.[46], to declare that Frye test was no longer a federal rule, in the area of expert evidence. The court also ruled that Rule 702 applied to all scientific evidence and not only to novel scientific evidence.[47]Rule 702 is essentially a relevancy rule which allows admission of ‘scientific, technical or other specialised knowledge’ if it is provided by an expert and it will assist the trier of fact. The court in Daubert however interpreted ‘scientific knowledge’ as a ‘standard evidentiary test’ based on ‘scientific validity’.[48]The court ruled that the basis of evaluating scientific evidence be based on four considerations:

·         Whether the reasoning or methodology has been tested (falsibility).

·         Whether the reasoning or methodology has been subjected to peer review and publication.

·         Potential rate of error of methodology or reasoning.

·         Whether the methodology or reasoning has been generally accepted or respected in the relevant scientific community.[49]

Chief Justice Rehnquist and Justice Steven concurred in part and dissented in part with the judgement in Daubert. Rehnquist CJ had doubts that judges will be able to evaluate evidentiary reliability based on scientific validity as ‘it imposes on them either the obligation or the authority to become amateur scientist in order to perform the role’ and his opinion was to’ leave further development of this important area of law to future cases’.[50]Rehnquist CJ’s pessimism has been borne out by subsequent research in US which shows that judges there have difficulty in applying the Daubert criteria effectively.[51] In the U.S, some states have adopted the Daubert standard but others have rejected it and continue to retain Frye. In fact the state of Wisconsin in US still uses the relevancy test for admission of scientific or other expert evidence. It uses a three part test to determine relevance; (1) the court determines whether the evidence is relevant, (2) whether the witness is qualified as an expert and (3) whether the evidence will assist the trier of fact.[52] Studies have also found that adoption of the Daubert test at state or federal level has had no statistical significant effect on admission rate of scientific evidence.[53]

The Daubert test remains the current law in the United States Federal Courts. Two subsequent cases, General Electric Co. v Joiner[54] and Kumho Tire Co. v Carmichael[55] extended the gatekeeper role of the courts to include non-scientific expert evidence. In 2000 the Federal rules of evidence were amended to incorporate the Daubert criteria.[56]A case law survey of the standards governing admissibility of scientific expert testimony in the US, carried out by Lustre in 2009, showed that 25 states were governed by Daubert standards, 15 states and District of Columbia by Frye standards and 6 states had a combination of Frye and Daubert standards and 4 states have developed their own standards.[57] The State of Wisconsin has rejected all reliability tests for admissibility of scientific evidence and continues to use the relevancy test.

The majority of criminal cases in the US are heard at state courts and there is yet no uniformity of the standards governing admissibility of scientific evidence in criminal cases in these courts. This may be due to a need to admit evidence, to answer questions of fact, where the evidence does not meet the requirements of a standard test governing admissibility of expert testimony. Even Federal Courts that apply Daubert standards regularly admit evidence with unproven methodologies.[58]  

       

        (C) Standards in Canada

Commonwealth countries generally apply the relevancy test for admission of expert evidence, however the admission criteria varies from case to case. Canadian jurisprudence for admission of expert evidence has generally shifted towards adopting  reliability test with some of it being adopted from US jurisprudence. In modern Canadian jurisprudence the most important decision regarding admissibility of expert evidence was made in R v Mohan.[59]This was an appeal from the Ontario Court of Appeal, where a paediatrician was charged with sexual assault of his patients and an expert witness testified that the defendant did not fit into the psychological profile of a perpetrator of such a crime. The question for the Supreme Court was whether the expert evidence was admissible which the Court of Appeal had allowed. The Supreme Court excluded the evidence and restored the conviction.

Sopinka J, giving judgement in the case outlined four criteria for admission of expert opinion evidence:

(1)Relevance

(2)Necessity in assisting the trier of fact

(3) Absence of exclusionary rule

(4) A properly qualified expert[60]

Relevance- a question of law to be decided by the judge is governed by logical relevance of the evidence and it has to be considered with other factors in mind, such as, time consumed, resources consumed and the probative value of the evidence as well as its prejudicial effect.[61]

Necessity- the information must be necessary to ‘enable the trier of fact to appreciate the matters in issue due to their technical nature’. This precondition standard of necessity should not be set too high or too low (helpful). Too low a standard would result in the trial becoming ‘nothing more than a contest of experts with the trier of fact acting as a referee…’.[62]

The third factor – absence of exclusionary rule, should take into consideration the rules of evidence (exclusionary rules) governing the trial.[63]The fourth factor-( A properly qualified expert)- the evidence must be given ‘by a witness who is shown to have acquired special or peculiar knowledge through the study or experience in respect of matters on which he or she undertakes to testify’.[64]

Sopinka J summarised the four criteria in the following words:

it appears from the foregoing that expert evidence which advances a novel scientific theory or technique is subjected to special scrutiny to determine whether it meets a basic threshold of reliability and whether it is essential in the sense that the trier of fact will be unable to come to a satisfactory conclusion without the assistance of the expert.  The closer the evidence approaches an opinion on an ultimate issue, the stricter the application of this principle.

The above four criteria mirrors the English admissibility criteria with an additional caveat that expert evidence involving ‘novel scientific theory or technique’ be subjected to special scrutiny before it is deemed reliable enough to be admissible.

The court in R v J-L.J[65]subsequently consolidated the concept of reliability as a threshold for admission of expert evidence. Binnie J in delivering the judgement said that ‘Mohan kept the door open to novel science, rejecting “general acceptance” test formulated in “Frye” and moving in parallel with its replacement, the “reliable foundation” test laid down in Daubert’[66], although Mohan had not referred to Daubert. Mohan and J-L.J reiterated the need to protect the judge and jury against ‘junk science’.[67]Both Mohan and J-L.J placed emphasis on the need to satisfy the trail judge that the evidence offered has ‘underlying principles’ and ‘methodology’ which was ‘reliable and more importantly applicable’.[68]The judgement also drew attention to the influential case of Davie v Magistrates of Edinburgh (1953)[69] which called for the experts to furnish the ‘necessary scientific criteria for testing the accuracy of their conclusion’.[70]

In R v Trochym[71], the J –L.J approach to admissibility of expert evidence was affirmed in a divided court. Justice Deschamps, providing the decision for the majority, referred to a number of high profile wrongful convictions and reiterated the ‘need to carefully scrutinise evidence presented against the accused for reliability and prejudicial effect to ensure basic fairness of the criminal process’.[72]

Deschamps J stressed that,

[r]eliability is an essential component of admissibility. Whereas the degree of reliability required by the courts may vary depending on the circumstances, evidence that is not sufficiently reliable is likely to undermine the fundamental fairness of the criminal process.[73]

The court revisited the Daubert criteria of reliability that was taken into consideration in J-L.J.[74]

The Canadian jurisprudence has in recent years moved towards the requirement of reliability as a standard for admissibility of expert evidence but the courts have not ‘explained how indicia of reliability… should be weighted or applied’.[75] Gary Edmond sums up the current state of law on admissibility of scientific evidence in Canada by saying that ‘as things stand a vague concept of reliability erratically impacts upon Canadian expert evidence jurisprudence and legal practice’.[76]The Canadian Supreme Court tends to decide on admissibility of scientific evidence on a case to case basis,[77] and lacks uniform standards for admissibility of expert testimony.   

         (D) Standards in Australia

In Australia too, the rules for admissibility of expert evidence remains unsettled. The law here has not resolved the test of ‘field of expertise’.[78]Some courts have adopted the Frye test and others the Daubert reliability test or a combination of both.[79]Incidentally the general acceptance test existed in Australian law long before Frye in the US. In R v Parker[80]the court held that fingerprint evidence could be admitted as evidence if the theory supporting the evidence was ‘generally recognised by scientific men’.

Following a high profile miscarriage of justice in R v Chamberlain[81], the Australian Law Reform Commission (ALRC), in 1985, was tasked with reviewing the existing standards of admissibility of expert evidence. ALRC in its interim report rejected Frye test as well as the need for a reliability test, stating that reliability is not an admission criteria but is of probative value.[82] In 1995 the Australian Parliament passed the Evidence Act 1995. Federal courts and courts in Australian Capital Territory apply the law in this Act while New South Wales, Tasmania and Norfolk Island have passed mirror legislation. Section 79 of the Evidence Act[83] states that, ‘[i]f a person has specialised knowledge based on person’s training, study or experience, the opinion rule does not apply to evidence of an opinion of that person that is wholly or substantially based on that knowledge’.

The rule is quite similar to the US Federal Rule of Evidence 702. Section 79 refers to ‘specialised knowledge’ while Rule 702 refers to ‘scientific, technical or other specialised knowledge’. Both Rule 702 and s79 are vague and this gives the courts some flexibility and opportunity to interpret the rules as the situation demands. The court in Daubert however has interpreted Rule 702 and established a set of criteria for admissibility and subsequently Rule 702 was amended in 2000 to accommodate the reliability criteria of Daubert.[84]However the interpretation of s79 has given rise to a lot of controversy. The definition of ‘scientific knowledge’ remains undefined.

At common law in Australia, Clark v Ryan[85] and Bonython v R[86]have confirmed that the ‘field of expertise’, requirement is based on the requirement that opinion should be derived from ‘body of knowledge which is both ‘organised’ and accepted’.[87]This is the threshold requirement of evidentiary reliability in most states in Australia.[88]This is similar to the requirement of general acceptance test in Frye. Einstein J in Idoport Pty Ltd v National Australian Bank Ltd[89]was of the opinion that s79 represents a rejection of Frye test. In R v Tang[90], Spigelman CJ held that the meaning of ‘knowledge’ in s79 is the same as in Daubert. Spigelman CJ in Tang also held that evidentiary reliability is not a consideration under s79.[91]Interpretation of statutory rule s79 remains unsettled. The statutory rule for admissibility of expert evidence in Australia has left the legal community more confused than ever before. No wonder Gary Edmond has been critical of the expert evidence admissibility law in Australia. He says that in New South Wales and Australia in general, ‘judges have not exhibited much interest in reliability of expert opinion evidence’[92] which has allowed incriminating expert testimony of doubtful reliability to be admitted despite the existence of statutory rules governing admissibility.

We need to be cognisant of the fact that Australian law has had a test of general acceptance eleven years before Frye came into existence. In R v Parker, Justice Malden held that fingerprint evidence can be admitted if the individuality of fingerprint ‘were generally recognised by scientific men’ or was ‘sufficiently studied to enable these propositions to be laid down on scientific fact’.[93]

The law in UK too has had a reliability test since 1953. The Court in Davie v Magistrates of Edinburgh[94]held that the duty of expert witness is ‘to furnish the judge or jury with the necessary scientific criteria for testing the accuracy of their conclusions, so as to enable the judge or jury to form their own independent judgment by the application of these criteria to the facts proved in evidence’.[95]Despite the fact that the law in Australia had a reliability test since 1912 and the US since 1923, yet rules of admissibility of expert scientific evidence remain unsettled in the common law jurisdiction. In the US, state courts still have no single uniform test for admissibility of expert evidence. In the US federal courts, judges have difficulty in applying the Daubert criteria as we have seen earlier. The only rule that is settled is the ‘relevancy test’ for admissibility of expert evidence.

 The reason for this unsettled state of affairs with the law appears to be borne out of a need to admit expert evidence to answer questions of fact, despite reliability issues with certain types of scientific expert testimony. These reliability issues prevent the formulation of standard reliability tests. Flexible admission standards appear to be the way forward until these reliability issues can be ironed out. What are these reliability issues with scientific expert testimony?

     (3) Reliability of Scientific Evidence

The Law Commission report on expert evidence in England and Wales (Law Com. No 325)[96] has defined reliable expert opinion as one that is ‘soundly based’ and that which is not sufficiently reliable as:

·         Opinion based on a hypothesis which has not been subjected to sufficient scrutiny (including, where appropriate, experimental or other testing), or which has failed to stand up to scrutiny.

·         Opinion based on flawed data

·         Opinion which relies on an examination, technique, method or process which was not properly carried out or applied….

·         The opinion relies on an inference or conclusion which has not been properly reached.[97]

The schedule to the draft bill provides a list of ‘generic factors’[98] for scientific, technical and other type of expertise, and some of them were;

·         Validity of methods by which data was obtained

·         Has the ‘accuracy or reliability of the results’ taken into consideration the ‘degree of precision or margins of uncertainty’

·         Whether the ‘material upon which the expert’s opinion is based has been reviewed by others with relevant expertise (for instance, in peer review publications), and the views of those others on the material’[99]

The definition of reliability in the Law Commission report is not as precise as the criteria laid down in Daubert. The reliability criteria as laid down in Daubert offer trial judges five criteria to consider for admissibility of expert evidence;

·         Whether the theory or technique ‘can be (and has been) tested’ (falsibility)

·         ‘Whether the theory or technique has been subjected to peer review and publication’

·         Whether the ‘known or potential error rate’ is acceptable

·         Whether standards exist and are maintained to control the techniques’ operation

·         Whether the technique or theory is generally accepted[100]

The question remains whether all medical and forensic expert evidence currently proffered in criminal trials can stand up to such rigorous scrutiny. The National Academic of Sciences (NAS) report, which was generated by a highly qualified and respected committee in USA, noted that ‘[t]he forensic science system, encompassing both research and practise, has serious problems …’.[101]  The report acknowledged that advances in forensic disciplines (esp. DNA) have helped solve many crimes but at the same time wrongful conviction of innocent people may have taken place due to faulty forensic analysis. What are these serious problems?

Some areas of medical and forensic expert evidence which has generated much academic debate is reviewed below, to see whether such evidence meets the reliability requirements stipulated by the Law Commission and to assess weaknesses and strengths of the evidence.

       

         (A)DNA- ‘God’s Signature’?

The most common biological evidence obtained in criminal investigation is samples of blood, semen, and saliva, besides many others which contain nuclear DNA. DNA typing is now universally accepted as a gold standard because of its reliability.[102]DNA profiling was first developed in the 1980’s and by 1990’s had become established enough to be called a gold standard in forensic science.[103]

          (1)DNA Profiling

The DNA revolution in forensic science has come a long way since it was discovered by Alec Jefferys and his team in 1984. They discovered hypervariable loci termed ‘minisatellites’ and developed a technique called ‘multilocus probe’ (MLP) which used enzymes to dissolve the DNA into fragments, and with the help of markers a pattern of bands was created which could be matched.[104]In the mid 1990’s Multiple Short Random Repeat (STR) technique was introduced which uses sequences of shorter length. This was made possible by a polymerase chain reaction (PCR) which permitted smaller amount of genetic material to be amplified.[105]Initially STR profiling was done with a single locus probe but subsequently two highly variable complex STR’s were added to reduce random match probabilities to about 1 in 50 million. These were termed as ‘second generation multiplex’ (SGM).In year 2000 four more loci were added to this multiplex and it came to be known as SGM plus (SGM+) which further reduced the match probability to less than 10^-13.[106]

According to Bentley and Lownds[107], there are two types of commercial kits commonly used in the UK; the SGM+ and the Identifiler. The SMG+ targets 10 loci and the Identifiler 15 loci and both kits target an additional locus (amelogenin) which identifies the sex of the contributor. The result of the analysis is represented by a graph called the electropherotogram (EPG) which is then interpreted by the scientist. Interpretation is usually not a problem when there is good quality and quantity of DNA material from one source. However the problem arises when the quantity of DNA is less than 0.5 nanogram. These samples are called low template DNA (LTDNA). Existing techniques for analysis of such samples were uninformative, which led to the development of a new technique- the Low Copy Number (LCN) analysis where the amplification cycle is increased and the samples are chemically enhanced to give better results.  The graphs produced by this method can produce ‘stochastic random effects’ which include a peak obtained from ‘background noise’, contamination (from other source) or ‘stutter’ (false peak obtained by the analytic process).[108]These random effects can generate ‘partial or incomplete’ peaks on the graph. This is where the interpretation of results can become very subjective. With LTDNA it is often not possible to provide reliable statistical random match probabilities. Computer programmes have been developed to overcome subjectivity in the analysis but such science is not yet established.[109]

DNA profiling raises both scientific and legal issues. For DNA profiling, the print has first to be created, then it has to be established if the prints match and finally the frequency of random match probabilities in the reference population have to be calculated.[110]

           (2)Creating a DNA print

The creating of the print is usually a straight forward process in diagnostics and research where ample amounts of DNA material are available and the tests can be repeated to produce good results. However in forensic science where the samples are often minute the tests cannot be repeated. Furthermore the samples are often mixtures of more than one person and to complicate matters the samples are often contaminated. The situation with SGM+ and Identifiler testing is not as bad as with LTDNA testing due to the quantity of DNA material available. The NRC report[111]concluded that ‘[t]he current laboratory procedures for detecting DNA variation…is fundamentally sound’,[112]but it stressed that there is ‘a need for standardization of laboratories procedures, proficiency testing and accreditation of laboratories in order to assure quality of forensic analysis’.[113]It also recommended that laboratory error rates as determined by proficiency testing should be disclosed to juries.[114]

           (3)Matching of the DNA Prints

After the print is obtained, the current procedure in laboratories, is to declare whether the pair of samples match or don’t match. Some laboratories have a third category of an inconclusive match. Some experts however favour a likelihood ratio of a match rather than a rigid match or no match approach.[115]Matching can be made difficult by the variations in the DNA prints which is caused by the quality of the sample (degradation can reduce the intensity of the bands) and the testing conditions.[116]This variation can produce observer errors. Having mixed samples from multiple sources can make the matching more complicated.

           (4)Estimation of Frequency of Match Obtained

The most controversial aspect of DNA testing is statistical estimation of the frequency of match in the reference population. It is necessary to know the statistical estimate so that a related person in the reference population does not match by chance.^[117]Frequency estimates of each band or allele are obtained from the national database. Based on these frequency estimates of the individual allele, the overall frequency of the match is obtained ‘using formulae that assure that the alleles are statistically independent’.^[118]The problem with these estimates is the assumption that the frequency of alleles is independent, since there are variabilities in the frequency of alleles in population subgroups from different racial background. The frequency of heterogeneous genotype are calculated by a formula, 2 pq (p and q are the two alleles in the genotype). If the frequency of band A is 0.03 and the frequency of band D is 0.05 then the frequency of genotype would be 0.003(0.03 x 0.05 x 2), that is to say the frequency of the AD genotype would be 3 in 1000. This calculation holds good if the alleles are statistically independent. For the alleles to be statistically independent the given population has to be in Hardy-Weinberg equilibrium.^[119]Most critics, however, believe that most populations are not in Hardy-Weinberg equilibrium because of non-random breeding in the population.^[120]

Having calculated the frequency of genotype, the frequency of the entire DNA profile has to be calculated (multilocus genotype). This is done by multiplying the frequencies of the individual genotypes (called the precedent rule). For this again the frequency of the genotypes has to be statistically independent. If the genotypes at different loci are statistically independent, the population is said to be in linkage equilibrium. Here again most critics believe that populations are not in linkage equilibrium because of non-random breeding.[121]

The 1992 NRC report [122]on DNA technology acknowledged that there are controversies regarding population structure and proposed using the ‘ceiling principle’ to estimate match frequencies. It recommended ‘random samples of 100 persons should be drawn from each of 15-20 populations, each represent a group relatively homogenous genetically; the largest frequency in any of these population or 5%, which is larger should be taken as the ceiling frequency’.[123] Soon this approach came under criticism and in 1996 the National Research Council, acknowledging the criticism, proposed that ‘profile frequencies be assigned a “confidence interval” formed by multiplying and dividing the frequency by 10’.[124]However the controversies surrounding estimation of frequency of match remain far from settled.

A stringent inquiry into fallibility of DNA profiling was first seen in the US in People v Castro.[125]Castro was accused of two counts of murder and the investigators obtained blood samples from Castro’s watch which he claimed was his own blood. Judge Sheindlin ruled that the DNA evidence was inadmissible for proving that the stain on the watch was that of the victim but was admissible if it was to show that the DNA profile on the watch was not that of the victim. The court held that ‘methods for determining exclusion’ were ‘less complex and more reliable than those used to show inclusion’.[126]Though the court ruled that DNA tests when properly performed were reliable and admissible under Frye rule, the evidence was not admissible to determine inclusion because of various deficiencies in the testing carried out by Lifecode laboratory. The court was critical of its ‘use of contaminated probes, the absence of laboratory controls and for the inconsistency between its methods for declaring match between samples and declaring a measured match in population data base’.[127]

In England and Wales the Forensic Science Regulator commissioned a review on the interpretation of DNA evidence and published the review in December 2012.[128]The aim of the report was to come up with principles that can be used in future to help generate a ‘unified interpretation and reporting policy’.[129]The report advocated the removal of the ‘artificial divide’ between conventional DNA and low template DNA. The report agreed with the ‘Caddy Review’[130]that the commercial term Low Copy Number (LCN) which is confusing, should be replaced by a more generic term- Low Template DNA analysis (LT-DNA). It proposed that the threshold of 100-200 picogram of genetic material available should not be used to differentiate conventional DNA and low level target profiles because DNA levels above this threshold levels may also have stochastic effects that are seen in low concentration DNA profiles.

The report also highlighted the lack of uniformity of validation methods among the police, government and private laboratories that provide DNA profiling. There is also no framework for monitoring and assessing the techniques used and the results produced by these laboratories. A study by John Butler in the US showed that ‘random match probability estimates (from the same electropherogram) varied by ten orders of magnitude between different suppliers’.[131]The problem of contamination of consumables and the lack of standard DNA database is also highlighted in the report. In conclusion the review states that at present there is no standard interpretation method for complex DNA analysis.

(5)Admissibility of DNA Evidence

The court in R v Hoey[132]in rejecting admissibility of DNA profiling evidence was critical of the poor handling of samples to protect the ‘ integrity and freedom from possible contamination’ as well as poor bagging, labelling and recording of the item.[133]The court was of the opinion that without a proper system in place for such protection, the evidence generated by forensic testing will be of no probative value to the tribunal.[134]The court was also of the view that LCN analysis ‘has not been “validated” by the international scientific community’.[135]Hot on the heel of this judgement was a review of the science of Low Template DNA analysis by Caddy et al. which concluded that the science supporting LTDNA was sound.[136]

In R v Reed and Reed[137]the court concluded that DNA profiling using LCN technique for samples above the stochastic threshold was reliable and is admissible unless and until further new discovery shows otherwise. The court was not certain what the stochastic level was but scientific evidence points to a level somewhere between 100-200 picogram.[138]As for quantity between 100-200 picogram, disagreement would be expected but such a situation would be rare[139]and the scientific disagreement will be ‘resolved as science of DNA profiling develops’. [140]The court however made no observation on the admissibility of evidence when the evidence was generated with quantities of DNA below 100 picogram.[141]The court in Reed and Reed also did not specify whether the stochastic threshold level referred to the total DNA available in a mixture sample or that of the minor profile.[142]In R v C [2010] EWCA Crim 2578 the court was of the opinion that reliability is the issue not the quantity.[143]However, the quantity of DNA is not the only factor that affects the stochastic threshold level because this threshold level depends on other factors such as sample degradation and the ‘presences of other materials that can affect profiling chemistry (termed inhibitors)’.[144]

The court in R v Dlugosz[145]acknowledged, quoting a paper by Dror and Hampikian[146], that there is no objective standard for the interpretation of mixed DNA profiling. The court admitted DNA evidence in the absence of statistical random match probability of the profiles. The court held that an evaluative opinion of the analysis of a mixed profile based on experience of the expert is admissible as long as the jury is informed of the limitation of the evidence that is provided.

The validity of the theory underlying the science of DNA analysis is universally accepted as reliable. However the validity of the techniques applying this theory has given rise to concern, especially its introduction into the courts. The problem of ‘poorly defined rules for declaring a match, experiments without controls; contaminated probes and samples; and sloppy interpretation of autoradiograms’ continues to give rise to doubts about reliability of such evidence.[147]There have been reports in the US of abuse of scientific evidence such as over stating frequency of genetic match, reporting inconclusive results as conclusive, failing to report conflicting results and reporting improbable results, among others, which has added to these concerns.[148]The courts in England and Wales have not laid down any uniform principles for admissibility of DNA evidence, except for saying that DNA profiling is reliable and evaluative opinion of DNA analysis based on experience of the expert is admissible as long as the jury is informed of the limitation of the evidence that is provided.

DNA profiling and its application in criminal investigations is therefore not fool proof. The courts have to take heed of these flaws in DNA evidence when admitting such evidence in criminal trials.

 

(B)Fingerprinting – A Gold Standard or A Junk Science?

Fingerprinting is based on the presences of friction ridges on the volar surface of the fingers and hand. These ridges form patterns in the form of arches, whorls and loops which persist throughout life unless there is skin loss or scarring. The arrangement of ridges or patterns varies from finger to finger and between individuals. It is believed that there are no two individuals with the same ridge patterns in the world. This uniqueness is the basis of fingerprinting. The analysis of these patterns is variously described as ‘finger ridge analysis, fingerprint comparison, fingerprint identification or individualisation’.[149]

Fingerprints taken from the crime scene are usually latent and various techniques are used to make these latent prints visible by fingerprint experts, who by training and experience are skilled to perform this. The prints are then compared with those obtained from the suspect or from a database. The finger ridge pattern may be unique to an individual but the fingerprint examiners work with impressions which at times are partial impressions. The NAS report 2009 argues that,

   

‘Uniqueness does not guarantee that prints from two different people are always sufficiently different that they cannot be confused or that the two impressions made by the same finger will be sufficiently similar to be discerned as coming from the same source. The impression left by a given finger will be differ every time, because of inevitable variations in pressure, which change the degree of contact between each part of the ridge structure and the impression medium. None of these variabilities – of feature across a population of fingers or of repeated impressions left by the same finger- has been characterised, quantified, or compared’.[150]

The friction skin ridges are three dimensional whereas the prints or impressions obtained are two dimensional and many details of the unique friction skin ridges do not survive this transition, where pressure with which the impression is made and the surfaces from which the impression are taken can produce distortions which further affects the quality of the print.[151]

Fingerprinting identification has in the past been regarded as a ‘gold standard’[152]and has been accepted without challenge in the US courts for more than 100 years, which has led everyone, including the courts to believe that it is infallible.[153]The fingerprinting community believes that the technique when properly applied has an error rate that approaches zero.[154]However the NAS report says that ‘there is limited information about the accuracy and reliability of friction ridge analyses, claims that these analyses have zero error rates are not scientifically plausible’.[155]Jennifer Mnookin, while highlighting the weaknesses of fingerprinting evidence, sums it up by saying that;

     … [g]iven the general lack of validity testing for fingerprinting; the relative dearth of difficult proficiency tests; the lack of a statistically valid model of fingerprinting; and the lack validated standards for declaring a match, such claims of absolute, certain confidence in identification are unjustified….Therefore, in order to pass scrutiny under Daubert, fingerprinting experts should exhibit a greater degree of epistemological humility. Claims of “absolute” and “positive” identification should be replaced by more modest claims about the meaning and significance of a match.[156]

   

Simon Cole has pinpointed some of the problems with fingerprinting evidence;        

·         Fingerprinting has never been scientifically tested

·         Fingerprinting community has not yet articulated what constitutes a fingerprinting forensic match

·         Fingerprinting experts are self-regulated, but weakly self-regulated

·         Forensic fingerprint identification claims an exaggerated degree of scientific certainty

·         Fingerprinting identification has enjoyed enormous freedom from scrutiny[157]

He concludes by saying that ‘fingerprinting has constructed a perfect rhetorical system, in which the actual accuracy of the technique is irrelevant’ and that the ‘conceptual framework in which fingerprint examiners operate is “junk science”’. This framework he contends is not suitable for the post Daubert era.[158]Cole is right when he says that it is ‘becoming increasing difficult to find a scholar who will argue that latent print individualization is valid’.[159]

In the UK too there has been a ‘blanket reception’ of fingerprinting evidence.[160] The legal community in the UK woke up to the fallibility of fingerprinting evidence in 2007, when the Scottish Government, acceded to calls for a public inquiry, following the evidence of two defence fingerprinting experts in McKie’s trial[161] that the Scottish Criminal Record Office had misidentified fingerprints. The Fingerprint Report 2011[162]acknowledged that there is ‘no evidence … to suggest that fingerprinting evidence as a class is inherently reliable’,  there is also ‘no basis for a claim of infallibility ’ and ‘[i]t is opinion evidence and where appropriate, it should be subject to robust scrutiny and challenge’.[163] The report goes on to say that ‘ [t]he legal profession, judges and juries need to be alert to the subjective nature of fingerprinting evidence and to the other factors of relevance to the assessment of the opinion of a fingerprinting examiner in order to consider this evidence on merit’.[164]

However as the NAS report suggest, this scrutiny by itself is not going to ‘cure the infirmities of the forensic science community’ because of the ‘limitations of the judicial system’ and the problems with the forensics sciences itself.[165]Despite growing academic claims of lack of evidence of reliability of fingerprinting evidence, the courts in UK continue to admit fingerprinting evidence without any serious challenge. In R v Buckley [1999] 163 JP 561 an unsuccessful challenge to admissibility of fingerprinting evidence was made. The Court of Appeal laid down grounds on which the court uses its discretion to admit or exclude fingerprinting evidence. The court held that, admissibility of such evidence will depend on all circumstances of the case, in particular;

 (i) the experience and expertise of the witness;

 (ii) the number of similar ridge characteristics;

 (iii) whether there are dissimilar characteristics;

(iv) the size of the print relied on, in that the same number of similar ridge characteristics   may be more compelling in a fragment of print than in an entire print; and

 (v) the quality and clarity of the print on the item relied on, which may involve, for example, consideration of possible injury to the person who left the print, as well as factors such as smearing or contamination.[166]

The most recent and only other challenge to fingerprint evidence was in R v Smith [2011] 2 Cr App R 16. The Court of Appeal quashed the appellant’s conviction of murder after hearing conflicting fingerprint evidence from experienced experts. At trial, the prosecution’s fingerprint evidence could not be effectively challenged because the Crown had indicated that it would challenge the qualification of the defence fingerprint expert, because her name was not on Home Office register of fingerprint experts. In UK only the names of UK police personnel who have completed the requisite training appear on the register. The court held that it is not for police but for the court to decide who is a competent witness.[167]The court was also critical of the unprecedented monopoly held by the police force and lamented the lack of defence access to independent experts in this field. The court also highlighted the poor quality fingerprint reports prepared by the prosecution. There were no detailed notes kept of the examination and no reasons stipulated for the conclusion reached.[168]There court was of the opinion that fingerprint evidence should be presented to the court and the jury using modern presentation methods so that the jury can make an informed decision.[169]In conclusion the court recommended that fingerprinting be the subject of further, wider investigations.[170]

In response to the court’s recommendation in R v Smith, the Forensic Science Regulator, Andrew Rennison and Gary Pugh, the chair of the Fingerprint Quality Standards Specialist Group, released a paper, ‘Developing a Quality Standard for Fingerprint Examination’, in December 2011.[171]The paper reiterated that: ‘ …it is the accuracy of fingerprinting examiners rather than the uniqueness or persistence of fingerprints that is the foundational issue in the reliable provision of fingerprinting evidence’.[172]The report also stressed that:

·         Fingerprinting is not a science

·         There is a need to recognise the risk of human error

·         There is a need for high level of individual competence

·         That fingerprinting evidence must be objective and impartial

·         That the methods must be valid[173]

Such quality frameworks are to some extent available in other areas of forensic science such as DNA profiling but are still missing in the fingerprinting discipline, which raises questions of reliability of the fingerprinting evidence presented to the courts. With the recent challenge to fingerprinting evidence in Smith, the release of the fingerprinting report (Scotland 2011) and the Forensic Science Regulator report (December 2011), more challenges to such evidence can be expected in future. The days of blanket reception of such evidence should be over but it is left to be seen if the courts, judges and lawyers are prepared to take up the challenge.

 

(C)Bite-Mark Evidence

Bite-mark analysis involves the ‘detection, recognition, description, and comparison’ of bite-marks.[174]Forensic odontology was originally used for identification of victims of mass disasters. However in the 1970’s forensic odontologist extended their expertise to provide testimony in criminal proceedings. Bite-marks found on the victim or the suspect and sometimes on other objects are compared or matched to the dentition of the person who inflicted the bite. A ‘match’ would imply that human dentition and the imprint that it creates is unique and that it excludes the possibility of any other individual’s dentition could have created the same imprint. However to date there is no universal agreement that human dentition is unique.[175] Let alone the uniqueness of human dentition, bite-marks are usually made by only a limited number teeth. Often the bite mark leaves a very unsatisfactory impression on the skin or food material and the impression is often characterised by ‘shrinkage’ and distortion’.[176]

In fact scholars argue that ‘no data that could permit forensic scientist to offer identification “to the exclusion of all others in the world” exists and they are unlikely to come into being in the foreseeable future. Such testimony is speculative and improper, both scientifically and legally’.[177]According to Adam ‘[b]ite-mark testimony…fails each of the five prongs of the Daubert test’.[178]Giannelli further dents the validity of bite-mark testimony by saying that given its background ‘it is critical that bite-mark evidence be challenged’.[179]Testimony to the weakness of bite-mark evidence is the ‘increasing numbers of wrongful convictions that are associated at least in part with bite-mark analysis’, which have come to light after DNA testing.[180]The strength of bite-mark evidence probably lies in exclusion i.e. a person’s dentition and bite-mark do not match than for finding a match. [181]Despite the lack of scientific basis of bite-mark evidence there has been consistent judicial acceptance of such evidence under Daubert, Frye and the Federal Rules of Evidence.[182]

(D)Foot Wear Mark Evidence (Shoe print Evidence)

Impression evidence of footwear at crime scene which may be latent (not visible to naked eye) or patent (visible) has to be collected, preserved and enhanced. The quality of such impression obtained will depend on the ‘experience, training and the scientific knowledge of the scene investigator as well as the agency’s resources’.[183]After the impression is analysed, it is compared for individual characteristics with the characteristics of the suspected source.[184]The criteria for positive identification depend on the individual laboratory as well as on the experience of the examiner and the clarity and uniqueness of the characteristics. There are two types of characteristics that are compared, the class characteristic which results from manufacture of the shoe (design/size) and identifying characteristic which results from objects attached to the sole and damage caused by cuts. The outcome of the comparison is usually declared as a ‘match or not a match’. However the Scientific Working Group for Shoeprint and Tire Thread Evidence (SWGTREAD), USA, recommends the following terminology:

·                     Identification—a definite conclusion identity;

·                     Probably made—a very high degree of association;

·                     Could have made—a significant association of multiple class characteristics;

·                     Inconclusive—limited association of some characteristics;

·                     Probably did not make—a very high degree of non-association;

·                     Elimination—definite exclusion;

·                     Unsuitable—lack of sufficient characteristics for a meaningful comparison.[185]

Foot wear impression comparison is usually not assigned any ‘probabilistic or statistical significance.[186]

Footwear print identification is a subjective analysis, as with other impression/pattern evidence, and it gives rise to observer errors in the absence of rigorous laboratory standards. Majamaa and Anja[187] did a survey in which six sets of shoeprint photographs with fictitious crime scenes were distributed to 34 crime laboratories for analysis. The results showed that there were considerable differences ‘in the conclusion of identical cases in the reports from different laboratories’.[188]

According to NAS report, for footwear mark evidence ‘there is no consensus regarding the number of individual characteristic needed to make a positive identification’ and the committee was not aware of ‘any data about the variability of class or individual characteristics or about the validity or reliability of the method’.[189]

In England and Wales the Court of Appeal in R v T[190]made it clear that experts can give footwear mark evidence in court but they cannot use likelihood ratios in forming an evaluative opinion. Thomas L J held that; ‘there is not a sufficiently reliable basis for an expert to be able to express an opinion based on the use of a mathematical formula ’and that the court is ‘satisfied that in the area of footwear evidence, no attempt can realistically be made in the generality of cases to use a formula to calculate probabilities’. The practise apparently has no ‘sound basis’.[191]

The court disagreed with the Forensic Science Regulator which practised such an approach of calculating likelihood ratios for footwear evidence.[192]The court preferred the use of words such as, ‘could have made’(the mark),which would help the jury better understand the nature of evidence, than to use ‘more “opaque phrases” such as moderate (scientific) support’ which would be more misleading to the jury, especially use of the word “scientific”.[193]

The court rightly subjected the expert evidence to strict scrutiny and also admonished the expert for adopting a reporting protocol which failed to provide a transparent basis for reaching the conclusions that were made.[194]At the appeal, the court ruled that in the absence of sufficient and accurate data, probabilistic calculations are ‘inherently unreliable’ except in the ‘field of DNA (and possibly other areas where the practice has a firm statistical base)…’.[195] This is a timely reminder that, with the exception of DNA evidence which is supported by established scientific data, forensic evidence such as footwear mark evidence has reliability issues. Such evidence may have some probative value but it cannot be used as a basis for conviction in criminal trials till its reliability can gain more credence.

(E)Ear Print Evidence

Ear-print evidence has been in use in Europe since 1965 but in UK it came into use in 1996. In some jurisdictions, it has in the past been successfully used for prosecution in crimes without much challenge.[196] The use of ear-prints has come under criticism because of lack of evidence that human ears are unique or the ear print impressions obtained are unique. Ears are malleable three dimensional structures and the two dimensional prints obtained from the crime scene are affected by several variables, the main being pressure distortion. The reliability of these prints is further affected by the lack of standardised protocols for retrieving and analysing these impressions. [197] The process of retrieval and individualisation is subjective and there is a dearth of publications and peer review on the subject. Research in this area of forensic science is on-going. The EU funded Forensic Ear Identification Research Project (FearID) completed its first project in 2005, which looked into ways to improve retrieval of ear print impressions, overcoming difficulties associated with pressure distortions and developing a more reliable system of matching and classifying ear prints. FearID intends to continue its research into finding ways to present its data using likelihood ratios which can be used in the courts.[198]

In the US Courts, ear print evidence does not meet the admissibility standards. In State v Kunze[199], the defendant was convicted of murder based on partial latent ear print evidence. At the Court of Appeal, Morgan J ruled that ear print identification evidence has yet to gain general acceptance in the relevant scientific, technical or specialist community and hence fails to meet the requirements of the Frye test of admissibility.[200] Morgan J however held that the court does not bar testimony, of lifting and preparing latent prints and also testimony of the similarities and differences between the prints obtained, which could easily be evaluated by the jury. A testimony of non-inclusion i.e. the defendant cannot be excluded as the person who made the latent prints can be accepted but an opinion of inclusion i.e. the defendant made or probably made the prints cannot be accepted.[201]

Ear print identification evidence is however admissible in English law. In R v Dallagher[202]the defendant was convicted of murder based in part on ear print evidence. The trail judge in fact directed the jury that they could convict the defendant on the basis of ear print evidence alone in the absence of other supporting evidence.[203] At Appeal the defence counsel introduced evidence from three expert witnesses who challenged the reliability of ear print evidence. They cited the absence of empirical research on the uniqueness of individual ears or that of the ear prints. In fact the prosecution expert agreed that high variability or uniqueness of ear print was an assumption based on limited experience.[204]Other issues, such as pliability of the ears and pressure distortion of the impression, which affect reliability of the prints, were also raised.[205]The lack of standardised methodology, absence of objective universally accepted criteria for analysis and comparison of the prints and the subjective nature of the analysis were also highlighted. The appeal was allowed and a retrial ordered. The case was however dropped by the Crown in 2004.[206]

In R v Kempster[207]the defendant had been convicted in2001 for one count of attempted burglary and three counts of burglary. His first appeal in 2003 was dismissed. An appeal was heard again in2008 at the recommendation of the Criminal Cases Review Commission. The court allowed the appeal against conviction on the first count of burglary and upheld conviction on other three counts. The court held that ‘ [w]e have no doubt that evidence of those experienced in ear print is capable of being relevant and admissible’ but ‘ [t]he question in each case will be whether it is probative’.[208]The court having examined the ear prints found that they ‘do not provide a precise match’.[209]The appeal was allowed and the court found the conviction to be unsafe. The experts from both sides were in agreement that ear prints pose a ‘ different and difficult problem than fingerprints’ and that the ear can deform under pressure and the fact that the person pressing the ear on a surface does not remain motionless leads to distortions of the ear and of the impression left on the surface.[210]

Although ear print evidence is admissible in English Courts, both the appeals have cast serious doubts about the reliability of such evidence. Ear print evidence is admitted in English law since it fulfils the requirements of the ‘relevancy test’ but obviously the probative value of such evidence is too low for it to be placed before the jury and should be filtered out at the beginning of the trial.

A strict application of the Law Commission’s and or the Daubert criteria for reliability of expert evidence would show that, except for DNA evidence (which is also not fool proof), other evidence such as fingerprinting, bite mark, ear print and footwear mark evidence do not meet most of the criteria.  There are obviously many reliability issues with such expert evidence which is proffered in criminal proceedings. Some evidence has more probative value while others have very little such value. The problem is where to draw the line between what is and is not reliable. Reliability test require some such distinction to be made.

 

(F)Publication and Peer Review

The Court in Daubert said that,

publication (which is but one element of peer review) … does not necessarily correlate with reliability…[b]ut submission to the scrutiny of the scientific community is a component of “good science”, in part because it increases the likelihood that substantive flaws in methodology will be detected’ and that ‘publication (or a lack thereof) in a peer reviewed journal will be a relevant… consideration in assessing scientific validity….[211]

On the other hand Richard Smith, the then editor of British Medical Journal (BMJ) had this to say:

   …there is something rotten in the state of scientific publishing and … we need radical reform. The problem with peer review is that we have good evidence on its deficiencies and poor evidence on its benefits. We know that it is expensive, slow, open to abuse, possibly anti-innovatory, and unable to detect fraud. We also know that the published papers that emerge from the process are often grossly deficient.[212]

Standards to assure quality of research produced have long existed. There are published standards and checklists, in addition there is peer review for funding and there are supervisory advisory groups and committees which review the whole process. At the end of it all this, research goes through publication peer review.[213]

As with all standards and procedures, there often exists the question of whether these standards are effectively applied to the processes to ensure quality. It is often assumed that publication peer review is a guarantee that the research fulfils the criterion of a quality product. It is supposed to guard against ‘mediocrity, bias, and deception – both conscious and unconscious- in research’.[214]

However a review of literature on peer review by Grayson revealed startling inadequacies in the system. It is unable to ‘detect deliberate deception or bias when, as is probable, the author has gone to great length to massage or fabricate plausible data and interpretation’.[215]The relationship between research and industry has allowed the industry to dictate and guide research which is in their favour. The industry sponsors control not only the research process but also the ultimate publication of results that favour them.[216]In summary Grayson concludes that biomedical peer review is:

·         Slow - because of sheer volume of paper that unpaid editors and reviewers have to handle leading to years of delay

·         Expensive – In terms of academic and editorial time consumed

·         Biased – including factors such as Intellectual bias, seniority of authors, institutional affiliations, nationality, language, geographical location, conservatism and discrimination against dissenting opinion.

·         Abused – including factors such as unconscious bias merging into conscious bias, special pleadings, witch hunting, promotion favouring colleges and protégés, anonymity of reviewers, favouring of articles on ‘hot topics’ and limited pool of reviewers especially in sub-specialities

·         Incompetent – sloppy practices - senior members of editorial board who act as reviewers not having training in epidemiology and statistics - and the poor quality of some publications suggests incompetence of peer reviewers who know less than authors due to highly specialised modern science

·         Unable to detect fraud – Inability of peer review to detect gross scientific fraud with potentially dangerous consequences -  fraud generated by authors to improve personal standing in a very competitive scientific arena -  senior honorary authorship favours the junior researchers chance of publication and it makes judging the true quality of the research more difficulty, the outstanding example was that of Malcolm Pearce in the 1990’s[217]

Armstrong highlights other problems with peer review. His research has found that, peer review is, unreliable, uninformative to readers, weak on quality, biased against replication, ignores usefulness and rejects surprising finding. His observations are that to get ones research published one has to avoid ‘examining important problems, challenging existing beliefs, obtaining surprising results, using simple method, providing disclosure or writing clearly’.[218]

The role of peer review is to evaluate manuscripts to either accept for publication or to reject them and not to ascertain their authenticity. The process does not ‘ensure scientific authenticity, accountability or authority’.[219] There appears to be ‘no universal, objective and infallible procedures, standards and goals’ in the review process.[220]However attempts are being made to improve the quality of research that is published by standardising the reporting of research. In the past (even today) most publications used the IMRAD (Introduction, Methods, Results, and Discussion/Conclusion) approach of reporting. Now several new formats have been proposed depending on the methodology of the research or the specific research designs. These include:

·         Consort- Consolidated Standards for Reporting Trials (22 item checklist)

·         Quorom – Quality of Reporting of Meta-Analysis (17 item checklist)

·         Moose – Meta-Analysis of Observational Studies in Epidemiology (35 item checklist)

·         Trend – Transparent Reporting of Evaluation with Non-randomised Designs (22 item checklist)

·         Stard – Standards for Reporting of Diagnostic Accuracy ( 25 item checklist)[221]

Checklists in research publication can improve the quality of the research but these publication standards need to be widely accepted by the publication fraternity.

The ICMJE (International Committee of Medical Journal Editors), a group of editors of leading biomedical journals has since 1978 (the Vancouver Group), been working to develop editorial polices and guidelines. Despite having such policies and guidelines, these leading medical journals still have problems ‘identifying instances of gross errors and fraud, let alone systematic bias, omissions and exaggerations’.[222]

Given these inherent weakness in the publication and the peer review process, shouldn’t  the courts stop ‘idealizing images of science’ and ask whether they should admit evidence based on publication which are redundant,  has no full disclosure of conflict, and those which are corporate sponsored or corporate influenced.[223]  Without doubt there is a need for a more intense scrutiny of the published and peer reviewed evidence that is presented to the courts. What is reliable remains the perennial question?

The criminal justice system is heavily reliant on forensic and medical expert testimony to resolve conflicts expeditiously and with finality. However, can expert testimony based on science; forensic science, forensic techniques and or technical expertise fulfil such legal needs. Does such testimony meet the Law Commission’s reliability criteria? The theory underlying the science of DNA obviously does but validity of the techniques applying this theory has given rise to concerns. Reliability of such evidence has been tainted by lack of laboratory controls, poorly defined rules, contamination and sloppy interpretations besides others. The reliability of fingerprinting evidence, which was once considered infallible, is now been questioned. Even the court in R v Smith recommended that fingerprinting be the subject of further, wider investigations. Currently other evidence such as that of bite-marks, footwear prints and ear prints, appears to have little or no quality framework in place for it to be of much probative value. However efforts are in place to build such quality frameworks. In recent years even evidence based on publication and peer review has come under criticism.

Considering these reliability issues it becomes difficulty to draw a fine line to distinguish reliable from not so reliable evidence which is often necessary to formulate standard reliability tests. The adversarial process and judicial review are not sufficient to overcome these problems. Further reforms in these areas are necessary and tightening the rules of admissibility of expert evidence with introduction of statutory reliability tests is unlikely to resolve these problems as has been seen in the US. Despite the existence of such reliability issues, medical and forensic expert testimony has a tendency to be given undue weight in the courts and evidence that lacks scientific basis or that based on errors, fraud, bias, omissions and exaggerations can lead to miscarriages of justice. 

                                  (4) Miscarriages of Justice

Miscarriages of justice are often blamed on the laissez-faire approach to admissibility of expert testimony. Are miscarriages of justice the result of unreliable expert testimony alone? An analysis of some recent miscarriages of justice may provide an insight.

Miscarriage of justice has been said to have occurred when someone is treated by the state in ‘breach of their right’ as a result of a ‘deficient process’ or ‘misapplication of law’ and when ‘factual justification’ for the punishment does not exist.[224]The causes of wrongful convictions leading to miscarriages of justice have been quite extensively studied and reported in the western literature. Some of these include:

·         Mistaken eye witness identification ( about 75% of the cases)

·         False confessions ( about 14-25 % of the cases)

·         Tunnel vision- Selective ‘filtering’ of evidence by investigating officers, scientists and lawyers to ‘build’ a case.     

·         Unreliable informant testimony/ falsification of evidence

·         Prosecutorial misconduct – nondisclosure of relevant evidence

·         Inadequate defence representation – incompetence of defence lawyers – lack of funds

·         Unreliable expert testimony and imperfect forensic science[225]

The exact extent to which unreliable expert evidence contributes to wrongful convictions is difficult to gauge but based on DNA exonerations, forensic errors have been implicated in about 66% of the cases and fraud or tainted evidence in about 31% of the cases.[226] Evidently it plays an important role in the miscarriages of justice. It attracts significant media and political attention which is often followed by a perceived need to reform the existing criminal justice system.

(A)The Guildford Four and The Maguires

In the late 1980’s and early 1990’s a number of high profile miscarriages of justice came to light including the Birmingham Six, the Guildford Four, the Maguires, the Tottenham Three , the Cardiff Three and the Taylor Sisters. Two public inquiries were established to improve the English justice system and restore public confidence.[227] The May Inquiry (1994) into the circumstances surrounding the miscarriage of justice in relation to the Guildford Four and Maguire Seven raised considerable doubts about the forensic science evidence.

The Guildford Four had spent 14 years in prison before their convictions for Guildford and Woolwich bombings were set aside by the Court of Appeal in 1989. The Maguires had their convictions for possession of explosives set aside in 1990 having served their full sentence.

The May Inquiry[228] concluded that the convictions were unsafe because of several reasons, one of the main being the prosecution’s contention that traces of nitro-glycerine found on the hands of the defendants and on the gloves in the house had come from handling of nitro-glycerine (NG) compound and could not have been from innocent contamination.[229] However tests conducted by experts especially for the inquiry showed that innocent contamination by touching or a use of a towel could occur.[230]Furthermore late in the trial fresh evidence became available that the TLC method used for identification of NG could not distinguish NG from pentaerythritol tetranitrate(PETN).The prosecution assured the defence that if the experts were recalled they would be able to exclude PETN by the same test[231]which was in fact not so. There were tests available to distinguish between NG and PETN.[232] The defence counsels were ‘seriously misled’, leading them to make ‘unnecessary concessions which the judge then accepted’.[233]A second test on the Maguire samples was carried out using different solvents but this was not revealed at the trail.[234]

The May report concluded that there was a failure to disclose relevant information and that the ‘scientists … imperfectly understood their duties as forensic scientists and witnesses’. The report was critical of, the failure of the prosecution to disclose expert notebooks[235]and weaknesses in the conduct of the trial by Donaldson J.[236] . The report ‘apportioned most of the responsibility to individuals, particularly the individual scientist’.[237]This individual failing according to the report was not due to weakness or fault of the criminal justice system and no rules in the system could provide complete protection from these failings.[238]A reliability test for admission of expert evidence could not have prevented these wrongful convictions, when there is lack of judicial vigilance and failure to adhere to proper investigative and prosecutorial procedures.

(B)Sally Clark and Angela Cannings

In R v Sally Clark[239], Sally Clark had been convicted of murdering her two infant sons, Harry and Christopher. Her first appeal against the conviction was dismissed. A subsequent appeal was successful on two grounds. The first, on grounds that the Crown pathologist, Dr Williams had not revealed crucial microbiological test on Harry which suggested he might have died from natural causes and secondly the statistical evidence provided by Professor Meadow on the chances of two infants deaths in the same family was flawed.

The Court of Appeal admitted that, taking all medical evidence into consideration, this was a difficult case. There was considerable disagreement between experts as to whether to classify the deaths as due to unnatural causes or uncertain causes.[240]The court put into perspective, the failings on the part of the prosecution witness, Dr Williams, by agreeing with a report by Professor Byard, a forensic science specialist from Australia who had written to the appellants solicitors:

  Standard protocols were not followed and essential steps such as routine dissection and histology were omitted which prevented verification of alleged autopsy findings. As well, a number of potentially important diagnoses and conclusions were altered over time. For example, Christopher's initial cause of death of lower respiratory tract infection was withdrawn; observations of no significant haemorrhage within his lungs were changed to marked haemorrhage, …. The finding of retinal haemorrhages in Harry which was vital to sustain the diagnosis of shaken-impact syndrome was altered to no haemorrhage, brain lacerations were found to represent post-mortem artefact, swelling of the spinal cord was not present and bruising of paraspinal tissues was also not able to be substantiated. This is not a unique situation with statements in the literature in recent years that “investigations into the pathology and circumstances of sudden infant death are often scanty and inexpert” with significant omissions being documented when cases were audited. The Clark brothers demonstrate difficulties that may arise if cases are not fully investigated with all of the results being clearly summarised and discussed in the autopsy report. Trying to clarify findings, diagnoses and circumstances of death at a later stage may simply not be feasible due to a wide variety of possibilities other than inflicted injury.[241]

Furthermore Dr Williams had failed to reveal vital information regarding microbiological and virology reports of Harry which were all along with him but not revealed at trial or at the first appeal. Various samples collected at post-mortem tested positive for Staphylococcus which could have accounted for death from natural causes and could have compelled the jury to reach a different verdict, had all the information been available to them. [242]This led the Court to view the verdict as unsafe. The court also held the view that if ‘Harry’s death may have been from natural causes’ then ‘no safe conclusion could be reached that Christopher was killed unnaturally’.[243]

The court was also very critical of Professor Meadow’s statistical evidence which probably had an impact on the jury. He was asked of the risk of sudden infant deaths (SIDS) in a family. He quoted, a figure of 1 in 8543 for chance of a single SIDS in a family, from the Confidential Enquiry into Sudden Death in Infancy (CESDI). He quoted the risk of two deaths in the same family from SIDS as 1 in 73 million (a figure he obtained by multiplying 1 in 8543 by 1 in 8543).[244] He went on say that the chances of it happening in England and Wales was once every 100 years.[245]The evidence was later proven to be wrong. The court was surprised that there was no objection from the defence and that the evidence was put before the jury.

The case highlights misconduct and serious non-disclosure on the part of Dr Williams, misinformation by Professor Meadow, poor scientific and forensic methodology overall and lack of vigilance on the part of the defence and the court. Arguably standard common law test should have served to exclude Prof. Meadow’s evidence which was outside his field of expertise. It is doubtful that a reliability test for admission of expert evidence would have prevented such failings. Science cannot provide simple answers to complex scientific questions posed by deaths related to SIDS.

In R v Cannings[246], Angela Cannings was convicted for the murder of her two infants, Mathew and Jason, by smothering. Angela had four children, three of them in died in infancy. Three of the four children suffered from Acute Life-Threating Events (ATLEs) of who two died and one other child survived.

The Crown’s case was that Angela had smothered her two infants to death and the death of another child and three ALTEs in the family formed an ‘overall pattern’ of harm to the infants by smothering. The appellant who had no personality or psychotic disorder, and was described as a good and loving mother, denied hurting her children. Her contention was that the infants died from natural causes, which though unexplained, fell in the category of deaths known as Sudden Infant Death Syndrome (SIDS).

The Court of Appeal’s analysis of the evidence before it showed that there was not a ‘single piece of evidence conclusive of guilt’.[247]The Crowns case was exclusively based on ‘specialist evidence about the conclusion to be drawn from the history of three infant deaths and further ALTEs in the same family’.[248] 

The Crowns first witness Professor Meadow was particularly concerned by three infant deaths in the same family. His testimony was that where no natural cause of death has been found, the fact that the child was well before he died, and  the fact that three deaths occurred under similar circumstances in a family are features of a clinical diagnosis of smothering, although it may be a condition that is yet to be ‘understood or described’ by doctors.[249]Dr Ward Platt when asked about the Mathew’s and Jason’s death opined that earlier admission of them both to hospital ‘was highly suggestive of a baby who had been killed’.[250]The other evidence relied upon by the prosecution was the ‘pattern of events’ of six instances where all four children suffered serious ALTEs or death when in the sole care of the mother.[251]

The defence argued that there was evidence that unexpected natural death within the same family is not a rare event and studies have shown that three infant deaths from natural causes can take place in the same family.[252]Evidence of multiple infant deaths in the extended Cannings family suggestive of a genetic link was adduced by the defence.[253]Furthermore there was no physical evidence to support the allegations of inflicting harm to cause violent death in the infants.[254]

The Court of Appeal was critical of ‘dogma’ on the part of the prosecution witnesses for introducing evidence which has no scientific basis.  The court held that in view of the fact that ‘the fundamental basis of the Crown’s case, based on extreme rarity of three separate infant deaths in the same family and the pattern of events in this particular  family …’ has been undermined and there is ‘persuasive fresh evidence, which was not before the jury’, the conviction cannot be safe.[255]The court went on to say that in a criminal case murder has to be proved and that a high probability of guilt will not suffice.[256]

The Court of Appeal in Cannings acknowledged that the burden of proof for criminal conviction was not met. It was apparent that fallible scientific evidence by eminent experts tends to be readily accepted. It raises the question of why such evidence is accepted in spite of the inherent weaknesses that were exposed in Clark and Cannings. The answer according to Adam Wilson is social utility.[257]Despite any reliability test, social utility would require admission of testimony which may be potentially unreliable. The bulk of wrongful convictions are attributed to mistaken eye witness identification (75%) and false confessions (14% to 25%).[258]Unreliable expert testimony and imperfect forensic science forms a small segment of this spectrum, though it attracts the most attention especially in high profile cases. From the social context such evidence remains more favourable for securing a conviction in the absence of a guilty plea.[259]Failure to prosecute in the face of potentially flawed scientific evidence may lead to failure to ‘protect vulnerable members of society’.[260]

An analysis of the recent miscarriages of justice reveals that the failures were not due to inherent weaknesses of the forensics or medical sciences alone but also due to individual failings. Some of these failings that stand out are;

·         Prosecutorial impropriety

·         Lack of judicial vigilance

·         Unskilled defence counsel (who are often poorly funded)

·         Failure of expert witness to adhere to established protocols, improper documentation and dishonesty on the part of the expert witness

·         Undue weight been given to flawed expert testimony

 A reliability test in such circumstances could not have prevented miscarriages of justice but would in fact reduce pragmatism which is in the interest of social utility. Notwithstanding such weaknesses in the criminal justice system, problems of reliability of expert scientific/forensic evidence on which the system is heavily reliant, need to be addressed.

(5) Addressing the Problems of Reliability of Expert Scientific Evidence 

There is a fundamental difference between science and law. Though both seek the truth, the fundamental objective of law is justice which requires a ‘clear decision be made in a reasonable and limited amount of time’.[261]Daubert’s attempt to resolve the conflict between ‘legal truth’ and ‘scientific truth’ by only admitting scientific truth which will assist the trier of fact to make a just decision(legal truth) has not lived up to its expectation. ‘Criticism of Daubert and its progeny in state courts has been rampant…’ and ‘[a]cademic practitioners and trial judges all have complained that the standard is unclear, is difficult to interpret, leads to inconsistent results… and is confusing to the jury’.[262]Tightening the rules on the admissibility of scientific/forensic evidence is unlikely to resolve the perennial problem of unreliability of such evidence where Daubert appears to have failed. The weaknesses have to be addressed in some other way.

(A)Forensic Science Reforms

Though the contribution, of unreliable expert testimony and imperfect forensic science, to the overall causes of wrongful convictions is small, DNA based exonerations show that forensic errors play a significant role in miscarriages of justice.[263] Improving the reliability of scientific/forensic testimony by reforming the forensic institutional structure is needed.[264]Forensic science has been plagued by accusation of ‘sloppy, biased fraudulent work’, ‘pro-prosecution bias’, ‘police investigators and forensic scientist bias’ (both conscious and unconscious), ‘error and false interpretation of legitimate results’ among others.[265]In fact there have been accusations that ‘some forensic science expert witnesses are in a position where they can manufacture evidence merely by wishing it into existence, and evidence suggests that some of them have done precisely that’.[266]

The NAS report reiterated that one of the problem with admission of forensic evidence in criminal trials was the ‘degree to which methodology relies on human interpretation that is susceptible to bias, human error, or a lack of operational procedures and quality control of standards’.[267]The report suggested setting up of a cross-disciplinary, independent federal agency- National Institute of forensic Science- to addresses these weaknesses in forensic sciences.[268]It emphasized the need for standard reporting[269], standardizing operating procedures to reduce bias and human errors,[270]laboratory accreditation and professional certification[271]and quality control procedures.[272]

Conventional forensic ‘individualization’ sciences such as fingerprinting, dentition, writing, firearm, footwear, ear prints and bite marks, lack scientific basis that other science such as DNA printing offers. In each of these disciplines ‘… little rigorous, systemic research has been done to validate the discipline’s basic premises and technique, and in each area there is no evident reason why research would not be feasible’.[273]

In England and Wales, the forensic service providers included the Police Laboratories, the Forensic Science Service (FSS, which had virtual monopoly in the 1980’s) and some private forensic science providers (FSPs), the largest of which is LGC Forensics. The FSS was government owned contract-operated and provided services to the police force and the Crown Prosecution Service besides others.[274]The Forensic Science Regulator (FSR), a public appointee, ensures the quality standards. It has introduced accreditation to international standards (ISO 17025) to ensure quality of the services. The standards assess the organisational competence, individual competence as well as validity of methods, objectivity and impartiality.[275]The standards however do not guarantee quality and it does not cover ‘complex interpretation of results and presentation of evidence in court’.[276] The private providers that provide service to the government are subjected to these standards but the police laboratories are not. The impartiality of the police laboratories and the private service providers who did work for the government remains questionable. To overcome this problem the FSR has provided draft Codes of Practice and Conduct for forensic science providers and practitioners in the criminal justice system, in 2010.[277]The FSR however has no statutory powers to enforce compliance of its regulatory framework.

Besides reforms in forensic science, another important area which will benefit from reforms is forensic science research and education. However in England and Wales the state of affairs in this area remains ‘lamentable’.[278] Few universities have significant research output in forensic science, partly due to lack of government funding. The US House and Senate Democrats have proposed legislation, ‘The Forensic Science and Standards Act of 2012, S. 3378’ which will provide 300 million dollars over the next five years for research and to develop standards in forensic science.[279]In UK, the government could pass some form of legislation to make funding available for such purposes in the interest of criminal justice. Not only research but education in forensic science at the university level needs to be promoted aggressively.

The FSS has been closed down since 31^st March 2012 but the archives of files and samples have been retained to allow for review of old cases. There have been concerns that the closure of the FSS would affect forensic research which was previously carried out with public money and that commercialization of forensics may undermine the criminal justice system. [280]On the other hand the closure of the FSS may eliminate monopoly and decoupling of crime laboratories from the government agencies may improve the standards of forensic sciences. What is in store in the future is difficult to predict but without doubt, there is a definite need for robust reforms to improve forensic sciences for the good of criminal justice. There is a need to institutionalise reforms to produce valid results, eliminate bias and incompetence, as well as put in place adequate internal controls to produce trustworthy evidence, to improve the reliability of expert testimony. This has successfully been done in clinical laboratories and surely it can be done in forensic laboratories (unless science generated for litigation purposes intentionally lacks objectivity). To date the government has made admirable progress in this area of reforms. It is also apparent that such reforms alone will not be enough to overcome the reliability issue without riding the system of problems of partisan testimony.

(B)Court Appointed Experts

The Law Commission has recommended that the Crown Judges be given the power to appoint independent experts’… ‘in exceptional cases’,[281] where this may be of assistance to the court.[282] This is to be facilitated by the formation of ‘an independent panel of experienced legal professionals’ chaired by an experienced Circuit Judge.[283]. Criminal Procedure Rule 33.7 allows for the appointment of court-appointed experts. In the US this proposal has been in existence for over 100 years and has finally been codified by Rule 706 of the Federal Rules of Evidence.[284] The proposal is attractive because a court appointed neutral expert, who is compensated by the court, would reduce the bias that comes with partisan expert witnesses who are compensated and instructed by other parties. Court appointed experts can clarify issues for the jury which the partisan experts don’t do when they tailor their testimony to meet the needs of their clients who pay them. They can also narrow issues for the court which will save a lot of time.[285] However, experience in the US has shown that there are inherent problems in appointing such experts. There are difficulties with, locating these experts, complexities of communication and matters of compensation, besides the problem of distrust of the expert by the judges and difficulties in finding a neutral expert.[286]According to Saks, ‘court appointment of non-party experts’…’has been a resounding failure everywhere it has been tried’ since ‘either it is rarely used’ or ‘where it had been used more aggressively, it promptly fell into disuse’.[287]

Some of the other arguments against appointment of neutral experts include:

·         These experts would wield excessive power and it would be difficult to contradict their testimony which would become dispositive.

·         They may compromise the impartiality of the judges and the jury and undermine the adversarial litigation system.

·         They can mislead, be partial, fallible, and do as bad a job as anyone else.[288]

Samuel Gross disagrees with these critics and believes that appointment of neutral expert is ‘the most appealing solution to the problem of partisan expert evidence …’. He believes that the reason why it has persistently failed is that, the judges do not want to do so due to lack of time and resources, while the lawyers fear that they will not be able to have control over the witnesses. Conceptually this is an excellent reform, which can only be implemented if appointment of neutral witnesses is made mandatory by ‘restricting expert testimony to party chosen court appointed experts’ and compelling ‘the parties to secure the appointment of neutral experts’.[289]

In mainland Europe, the inquisitorial system using court appointed experts is widely used.[290]Problems in the US such as difficulty with finding or locating experts, difficulty of communication and matters of compensation have been surmounted in these countries. In France for example, there is a list of highly qualified experts in all fields which is prepared by strict scrutiny by a committee. Admission to this list is considered an honour and these experts provide their services for a fixed fee which is considerably lower than the legal fee in England.[291]The defence and the prosecution can call their own expert witnesses but it is rarely ever done.[292] The problem of lack of time and resources is not an issue since the system is already in place in mainland Europe. The possible stumbling block for such a reform appears to be the resistance from trial lawyers, in part due to personal or financial interest.[293]

(C)Multi-Disciplinary Advisory Panel (MAP)/ Validation Committee

Gary Edmond has proposed setting up of a multi-disciplinary advisory panel to overcome the problems of incriminatory scientific evidence being admitted, often without challenge, in criminal proceedings.[294]The composition of the advisory panel would include eminent experts from various disciplines, including ‘chemistry, the biosciences, epidemiology or medicine, engineering, statistics or mathematics, experimental psychology, along with forensic sciences, forensic pathology…, legal practice or the judiciary and an academic lawyer’; akin to the NAS committee. The eminent committee members would be selected but not remunerated to contain cost.[295]

The committee would produce written advice in areas where scientific evidence is problematic. Such areas would be identified by the judiciary, lawyers or scientific institutions and the committee will produce written reports based on the state of existing knowledge from relevant research literature. It would be in the form of a consensus statement on the reliability or unknown reliability of the technique or methodology or the science. The report would be in a simple non-technical language. It will provide useful information to the judiciary, lawyers and the relevant scientists and will be particularly useful for the defence who may not have access to such scientific and technical information.[296]

Resistance from key institutions and organisations in the criminal justice system, the composition and selection of the panel and the costs involved may be some of the issues to be addressed before such a reform can be instituted.[297]

Peter Alldridge has proposed ‘an extra-forensic validation committee for scientific methods to be used in courts’.[298]Such a committee would decide which evidence based on novel scientific procedure is sufficiently reliable to be admitted in legal proceeding and which has outlived its usefulness and is not sufficiently reliable to be admitted as evidence. Such certification and decertification of scientific procedures should preferably be done at an international level.[299]

Scholars have been arguing for a long time that many of the impression identification procedures, such as fingerprinting, bite-mark, footmark and ear printing, lack necessary quality framework and scientific reliability and yet evidence based on such techniques is regularly admitted as expert evidence. It is apparent that a line has to be drawn somewhere between which scientific/forensic evidence is suitable for legal purposes and which is not. The judges are often not qualified to make such decisions and adversarial system has not been able determine what is and is not reliable evidence. The advice of expert panels or committees would be helpful in such circumstances

The principle of the criminal justice system where the defendant is innocent until proven guilty is undermined when scientific evidence of doubtful accuracy and reliability is put forward to the jury by partisan witnesses. In the existing unbalanced adversarial system where the defence is disadvantaged by lack of resources, the defendant is ‘effectively guilty until proven innocent’.[300]Reforms are needed to prevent abuse of scientific evidence in this adversarial process. It can only come in the form of reforms to improve the generation of reliable scientific evidence, developing a system where partisanship can be eliminated through court appointed experts and having expert advisory committees to advice on reliability of the evidence presented to the courts. Reliability test alone cannot stop dubious scientific evidence from being admitted in the courts.

                  (6) Conclusion

Sporadically miscarriages of justice come to light creating a lot of media publicity and the policy makers respond by calling for reforms. The latest being the recommendations by the Law Commission for the introduction a statutory reliability test for admissibility of scientific evidence in criminal proceeding.[301] This is despite evidence that reliability test in other Common law jurisdictions has failed to prevent incriminating evidence being adduced in criminal trials. A review of admissibility standards in other Common law jurisdictions shows that none of the courts have been able to establish a satisfactory uniform practical standard which can determine the validity and reliability of scientific/forensic evidence that is proffered in criminal proceedings.

The nature of expert witness testimony is such that there can be no cut and dry binary distinction between reliable -unreliable or admissible - non-admissible evidence. There are no gold standards in science which is evolving all the time. DNA evidence, latent impression identification evidence such as fingerprinting, as well as evidence from peer review publications, is now viewed with scepticism. The greater the inherent weakness in the science, greater is the chance of expert witnesses over-claiming the reliability of such evidence. Strengthening the sciences through robust institutional reforms appears to be the way forward when there appears to be no solution to the thorny issue of reliability of expert evidence in criminal proceedings.

The call for a statutory reliability test for admissibility of expert testimony followed a spate of miscarriages of justice in recent years. However a review of recent miscarriages of justice reveals that there were many factors responsible for the wrongful convictions and the major one being individual failings. It is quite obvious that a reliability test would not have prevented these injustices. There were elements of prosecutorial improprieties, lack of judicial vigilance, weak defence, failure of individual expert witnesses, and partisan dishonest expert testimony. These weaknesses need to be addressed by improving vigilance on the part of the judiciary and elimination of partisan expert testimony. Court appointed or neutral non-partisan expert witnesses and expert panels will go a long way in achieving these aims.

Tightening the rules of admissibility with statutory reliability test alone is not going achieve the aim of preventing incriminating scientific evidence from creeping into the courtroom. Reliability test have been in existence for over a century but they have not resolved the problems associated with incriminating evidence being admitted in criminal trials. The current standard common law rules if properly applied should suffice but it would require more judicial vigilance, as well as consistency and uniformity in the application of these rules. The further drawbacks of strict reliability tests are that they reduce judicial pragmatism, flexibility and discretion which are necessary for the law to develop in tandem with science which is constantly evolving. Such flexibility, discretion and pragmatism is in the interest of social utility and justice. What is most needed, however, is trustworthy evidence which even when weak can be of substantive probative value. The quality of evidence generated has to be improved by necessary reforms. Ultimately a balance has to be struck between allowing juries access to expert testimony to help them reach a conclusion and admitting expert testimony of sufficient reliability to reduce the likelihood of miscarriages of justice. Reliability test alone is unlikely to solve all the problems.

Maybe, what Chief Justice Rehnquist said 20 years ago, when delivering his dissenting judgement in Daubert, would hold true today, that an evidentiary reliability test based on scientific validity is not necessary and that ‘further developments in this important area of law should be left to future cases’. 

Bibliography

Cases

R v Atkins (Dean) [2009] EWCA Crim 1876

Attorney-General’s Reference (No. 2 of 2002) [2003] 1 Cr App R 21

The Queen v Bonython (1984) 38 SASR 45

R v Briddick [2001] EWCA Crim 984

R v Broughton [2010] EWCA Crim 549

R v Buckley [1999] 163 JP 561

R v C [2010] EWCA Crim 2578

R v Cannings [2004] EWCA Crim 1; WLR 2607(CA (Crim Div) )

People v Castro 144 Misc. 2d 956, 545 N.Y.S 2d, 985 (Sup. Ct. 1989)

R v Chamberlain [1983] 46 A.L.R 493

R v Ciantar [2005] EWCA Crim 3559

Clark v Ryan (1960) 103 CLR 486

R v Clarke [1995] 2 Cr App R 425

R v Dallagher [2003] 1 Cr. App. R. 12

Daubert v Merrell Dow Pharmaceuticals Inc. 509 U.S. 579,(1993)

Davie v Magistrates of Edinburgh [1953] S.C. 34

R v Dlugosz [2013] EWCA Crim 2 (CA Crim Div)

Folkess, Bart v Chadd and others (1782)3 Douglas 157, 99 E.R. 589

United States v Frye 293 F 1013 (DC Cir., 1923)

R v G [2004] EWCA Crim 1240

R v Gardner [2004] EWCA Crim 1639

General Electric Co. v Joiner 522 U.S. 136 (1997)

R v Gilfoyle (Norman Edward) (Appeal against conviction) [2001] 2 Cr. App. R 5 (CA (Crm. Div.))

R v Harris [2005] EWCA Crim 1990

R v Harris (Lorraine) [2005] EWCA Crim 1980 at 270

HM Advocate v McKie 1999

R v Hodges [2003] EWCA Crim 290

R v Hoey [2007] NICC 49

R v Hookway [1999] Crim LR 750

R v Ibrahima [2005] EWCA Crim 1436

Idoport Pty Ltd v National Australian Bank Ltd [1999] NSWSC 828

R v J-L.J [2000] 2 SCR 600

R v Kempster [2008] EWCA Crim 975

Kumho Tire Co. v Carmichael 526 U.S. 137 (1999)

State v Kunze 97 Wash App 832, 988 P2d 977 (1999)

R v Luttrell [2004] 2 Cr App R 31

R v Mohan [1994] 2 SCR 9

R v Parenzee [2007] SASC 143

R v Parker [1912] V.L.R 152

R v Reed and Reed [2009] EWCA Crim 2698, [2010] 1 Cr App R 23

R v Robb [1991] 93 Cr. App. R 161

R v Sally Clark (No 2) [2003] EWCA Crim 1020

R v Silverlock [1894] 2 Q.B. 766

R v Smith [2011] 2 Cr App R 16

R v Stubbs [2006] EWCA Crim 2312

R v T [2010] EWCA Crim 2439, [2011] 1 Cr. App. R. 9

R v Tang [2006] NSWCCA 167

R v Trochym [2007] 1 SCR 239

R v Turner [1975] QB 834

Legislations

Criminal Procedure Rules 2005

Criminal Procedure Rules 2012

Criminal Procedure Rule 33

Criminal procedure Rule 33.7

Evidence Act 1995 (Australia)

Evidence Act 1995 section 70 and 79

United States Federal Rules of Evidence 1975

United States Federal Rules of Evidence 2000

United States Federal Rule of Evidence 702

United States Federal Rule of Evidence 706

Books

Goodstein D, ‘How science works in Reference Manual on Scientific Evidence’ 67, 80-82 (Federal Judicial Centre Ed. 2^nd ed. 2000)

Articles

Alldridge P, ‘Scientific expertise and comparative criminal procedure’ (1999) 3 3 Int’l J. Evidence & Proof 141

Armstrong J S, ‘Discovery and communication of important marketing findings: evidence and proposals’ (2003) Wharton School, University of Pennsylvania: PA, oct 31 pp at <repository.upen.edu/cgi/viewcontent.cgi?article=1021>

Bentley D and Lownds P, ‘Low template DNA’ (2011) 1 Archbold Review 5

Bernstein D, Junk science in the United States and the Commonwealth’ (1996) 21 Yale J. Int’l L. 123

Boaz A and Ashby D, ‘Fit for purpose? Assessing research quality for evidence based policy and practice’ (2003) ECRC UK Centre for Evidence Based Policy and Practice: Working paper11 at http://www.kcl.ac.uk/sspp/departments/politicaleconomy/research/cep/pubs/papers/assets/wp11.pdf

Caddy B, Taylor G and Linacre A, ‘A Review of the Science of Low Template DNA Analysis’ (2008) at http://www.bioforensics.com/articles/Caddy_Report.pdf

Caleb J, ‘Beyond People v Castro: A new standard of admissibility for DNA fingerprinting’ (1991) 7 J. Contemp. Health & Pol’y 269

Cheng E, and Yoon A, ‘Does Frye or Daubert matter? A study on scientific admissibility standards’ (2005) 91 Va. L. Rev. 471

Cole S, ‘Fingerprinting: The first junk science?’(2003) 28 Okla. City U. L. Rev.73

Cole S, ‘Comment on ‘scientific validation of fingerprint evidence under Daubert’ (2007) 7 Law, Probability and Risk 119

Cole S and Roberts A, ‘Certainty, individualisation and subjective nature of expert fingerprinting evidence’ (2012) 11 Crim. L. R. 824

Committee on DNA Tech. In Forensic Science, Nat’l Res. Council, DNA Technology in Forensic Science vii (1992)

Deitch A, ‘An inconvenient tooth: Forensic odontology is an inadmissible junk science when it is used to “match” teeth to bite-marks in skin’ (2009) Wis. L. Rev. 1205

Dror I and Hampikian G, ‘Subjectivity and bias in forensic DNA mixture interpretation’ (2011) 51 Science and Justice 2004

Edmond G, ‘Constructing miscarriages of justice: Misunderstanding scientific evidence in high profile criminal appeals’ (2002)22 Oxford J. Legal stud. 53

Edmond G, ‘Pathological science? Demonstrable reliability and expert forensic pathological evidence’ at www.attorneygeneral.jus.gov.on.ca accessed on 10 march 2013

Edmond G, ‘Specialised knowledge, the exclusionary discretion and reliability: Reassessing incriminating expert opinion evidence’ (2008) 31 NSWLJ 1

Edmond G, ‘Judging the scientific and medical literature: Some legal implication of changes to biomedical research and publication’ (2008) 28(3) Oxford J. Legal Studies 523

Edmond G, ‘Constructing miscarriages of justice: Misunderstanding scientific evidence in high profile criminal appeals’ (2002)22 Oxford journal of legal studies 53

Edmond G, ‘Advice for the courts? Sufficiently reliable assistance with forensic science and medicine (Part 2), (2012) 16 Int’l J. Evidence & Proof 263

Focus: ‘What are the standards for quality research’ (2005) NCDRD Technical Brief number 9 at www.acddr.org/kt/products/focus/focus9/

Gatowski S et al., ‘Asking the gatekeepers: A national survey of judges on judging expert evidence in a post-Daubert world’ (2001) 25(5) Law and Human Behaviour 433

Giannelli P and Imwinkelried, ‘Scientific evidence: the fallout from the Supreme Court’s decision in Kumho Tire’ (2000)14 Crim. Just. 12

Giannelli P, ‘Forensic science, Frye, Daubert and the Federal rules’ (1993) 29 Criminal Law Bulletin 428 at 432

Giannelli P, ‘Forensic symposium: The use and misuse of forensic evidence – Admissibility of scientific evidence’ (2003) 28 Okla. City U. L. Rev. 1

Giannelli P, ‘Bite-mark evidence’ (2007-2008) 22 Crim. Just. 42

Gould J and Leo R, ‘One hundred years later: Wrongful convictions after a century of research’ (2010) 100 (3) The Journal of Criminal Law & Criminology 825

Grayson L, ‘Evidence based policy and quality of evidence: Rethinking peer review’ (2002) ESRC UK Centre for Evidence Based Policy and Practice (working Paper 7) at www.kcl.ac.uk.

Gross S, ‘Expert Evidence’ (1991) Wis. L. Rev. 1113

Haber L and Haber R, ‘Scientific validation of fingerprinting evidence under Daubert’ (2008) 7 Law, Probability and Risk 87

Halpin S, ‘What have we got ear then? Developments in forensic science: Ear Prints as identification evidence at criminal trials’ (2008) 8 U. C. Dublin L. Rev. 65

Hon. Hammond G, ‘The new miscarriages of justice’ (2006) 14 Waikato L. Rev. 1

Howard M.N., ‘The neutral expert: a plausible threat to justice’ (1991) Feb Crim. L. R. 98

Jamieson A, ‘Case note - LCN DNA analysis and opinion on transfer: R v Reed and Reed’ (2011 Int’l J. Evidence & Proof 161

Jobling M and Gill P, “Encoded evidence: DNA in forensic analysis’ (2004) 5 Nature Reviews: Genetics 739

Justice McClellan P, ‘Admissibility of expert evidence under the Uniform Evidence Act’, Judicial College of Victoria, Emerging issues in expert evidence workshop, Melbourne, 2009

Keiser-Nielsen S, ‘Forensic odontology’ (1969) 1 U. Toledo L. Rev. 633

Kohar R,’FDCC Quarterly/Spring 2007

Koppl R, ‘How to improve forensic science’ (2005) E.J.L. & E. 256

Kovera M et al, ‘Assessment of the common sense psychology underlying Daubert’ (2002) 8 Psychology Public Policy and Law 180

Lustre A, ‘Annotations, Post-Daubert standards for admissibility of scientific and other evidence in state courts’ (2001)90 A.L.R. 453

Lynch M, ‘God’s signature: DNA profiling, the new gold standard in forensic science’ (2003) 27 Endeavour 93

Majamaa H and Ytti A, ‘Survey of the conclusion drawn of similar footwear cases in various crime laboratories’ (1996) 82 Forensic Science International 109

Mason M, ‘The scientific evidence problem: A philosophical approach’ (2001) 33 Ariz .St. L.J. 887

Meijer L, Thean A and Maat G, ‘Ear prints in forensic investigations’ (2005) 1:4 Forensic Sci. Med. Pathol. 247

Mnookin J, ‘The validity of fingerprinting identification: Confessions of a moderate’ (2008) 7 Law, Probability and Risk, 127

Note, ‘Admitting doubt: A new standard for scientific evidence’ (2009-2010) 123 Harv. L. Rev. 2021

O’Brian W, ‘Court scrutiny of expert evidence: recent decisions highlight the tensions’ (2003) 7 (3) International Journal of Evidence & Proof 172

Odger S and Richardson J, ‘Keeping bad science out of the courtroom’ (1995) 18 U.N.S.W.L.J 108

Pretty I and Sweet D, ‘A paradigm shift in the analysis of bite-marks’ (2010) 201 Forensic Science International 38

Pretty A and Sweet D, ‘The judicial view of bitemarks within the United States criminal justice system’ (2006) 24 J. Forensic Odonto-Stomatology 1

Reisinger K, ‘Court- appointed expert panels: A comparison of two models’ (1998) 32 Ind. L. Rew. 225

Rennison A and Pugh G, ‘Developing a quality standard for fingerprinting examination’ at www.homeoffice.gov.uk/agencies-public-bodies/fsr/

Riffe B. ‘Comment, The aftermath of Melendez: Highlighting the need for accreditation-based rules of admissibility for forensic evidence’ (2010) 27 T. M. Cooley L. Rev. 165

Roberts A, ‘Drawing on expertise: legal decision- making and the reception of expert evidence’ (2008) 6 Crim. L.R 443

Roberts P, ‘Forensic evidence after Runciman’ (1994) Crim. L. R. 780

Robins J, ‘The fall-out of closing the Forensic Science Service’ (2012) 176 Criminal Law & Justice Weekly 107

Rodrigues P, ‘Towards a new standard for admission of expert evidence in Illinois: A critique of the Frye general acceptance test and an argument for adoption of Daubert’ (2009-2010) 34 Ill. U. L. J. 289

Saks M, ‘The Phantom of the Courthouse’ (1995) 35 Jurimetrics 233

Saks M et al, ‘Model prevention and remedy of erroneous convictions Act’ (2001) 33 Ariz. St. L. J. 669

Smith R, ‘Peer review: Reform or revolution?’(1997) 315 British Medical Journal 759

Spencer J. R., ‘The neutral expert – an implausible bogey’ (1991) Feb Crim. L. R. 106

Thompson W, ‘Evaluating the admissibility of new genetic identification tests: Lessons from the “DNA wars”’ (1993/94) 84 J. Crim. L. & Criminology 22

Weir B, ‘The second National Research Council report on forensic DNA evidence’ (1996)59 Am. L. Hum. Genet. 500

Wilson A,’ Expert opinion evidence: the middle way’ (2009) 73 (5) J. Crim. L. 430

Wilson A, ‘Expert testimony in the dock’ (2005) 69 Journal of Criminal Law 330

Young R and Sanders A, ‘The Royal commission on criminal justice: A confidence trick?’(1994) 14 Oxford J. Legal Stud. 435

Other Sources

Guidance- Interpretation of DNA evidence www.gov.uk/goverment/publication/the interpretation-of-dna-evidence

House of Commons Science and Technology Committee; The Forensic Science Service- Seventh Report of session 2010-2012 at http://www.publications.parliament.uk/pa/cm201012/cmselect/cmsctech/855/855.pdf, accessed on 24/6/2013

Justice May, Second Report: Return to an address of the Honourable the House of Commons dated 3^rd December 1992 for a report of the Inquiry into the circumstances surrounding the convictions arising out of the bomb attacks in Guilford and Woolwich in 1974 (1992).

Justice May, Final Report: Return to an address of the Honourable the House of Commons dated 3^rd December 1992 for a report of the Inquiry into the circumstances surrounding the convictions arising out of the bomb attacks in Guilford and Woolwich in 1974 (1994)

Law Commission, ‘Expert Evidence in Criminal proceeding in England and Wales’ (Law Com no. 325), march 2011

National Research Council, ‘Strengthening forensic science in the United States: A part forward’ at http://ag.ca.gov/meetings/tf/pdf/2009_NAS_report.pdf

The Fingerprint Report / Scotland 2011 at www.thefingerprintreportscotland.org.uk

The fingerprint Inquiry 2011, at www.thefingerprintinquiryscotland.org.uk

---

[1] Richard Young and Andrew Sanders ‘The Royal Commission on Criminal Justice: A confidence trick?’(1994) 14 Oxford J. Legal Stud. 435.

[2]Royal Commission on Criminal Justice (Runciman Commission), 1991-1993 http://discovery.nationalarchives.gov.uk/SearchUI/Details?uri=C3042 and Justice May, Final Report: Return to an address of the Honourable the House of Commons dated 3^rd December 1992 for a report of the Inquiry into the circumstances surrounding the convictions arising out of the bomb attacks in Guilford and Woolwich in 1974 (1994).

[3] Paul Roberts, ‘Forensic evidence after Runciman’ (1994) Crim. L. R. 780.

[4] Gary Edmond, ‘Constructing miscarriages of justice: Misunderstanding scientific evidence in high profile criminal appeals’ (2002)22 Oxford J. Legal stud. 53.

[5] House of Commons Science and Technology Committee, Forensic Science on Trial, Seventh Report of Session 2004-5, HC 96-1.

[6] Law Commission Consultation Paper No 190 (2009) on admissibility of expert evidence in criminal proceedings in England and Wales at http://lawcommission.justice.gov.uk/docs/cp190_Expert_Evidence_Consultation.pdf .

[7] Law Commission, Expert Evidence in Criminal proceeding in England and Wales (Law Com no. 325), march 2011 at http://lawcommission.justice.gov.uk/docs/lc325_Expert_Evidence_Report.pdf.

[8] Gary Edmond’ ‘Is reliability sufficient? The Law Commission and expert evidence in international and interdisciplinary perspective: Part 1’ (2012) 16 (1) Int’l J. Evidence & Proof 30 at 42 and 47.

[9] Folkess, Bart v Chadd and others (1782)3 Douglas 157, 99 E.R. 589.

[10] R v Silverlock [1894] 2 QB 766 at 771.

[11] R v Robb [1991] 93 Cr. App. R 161.

[12] ibid at 161.

[13] Ibid at 164.

[14] See, William O’Brian, ‘Court scrutiny of expert evidence: recent decisions highlight the tensions’ (2003) 7 (3) International Journal of Evidence & Proof 172 and Adam Wilson,’ Expert opinion evidence: the middle way’ (2009) 73 (5) J. Crim. L. 430.

[15] R v Atkins (Dean) [2009] EWCA Crim 1876; R v Clarke [1995] 2 Cr App R 425; R v Hookway [1999] Crim LR 750; R v Briddick [2001] EWCA Crim 984; Attorney-General’s Reference (No. 2 of 2002) [2003] 1 Cr App R 21 and R v Gardner [2004] EWCA Crim 1639.

[16] R v Harris (Lorraine) [2005] EWCA Crim 1980 at 270.

[17] Andrew Roberts, ‘Drawing on expertise: legal decision- making and the reception of expert evidence’ (2008) 6 Crim. L.R 443.

[18] R v Gilfoyle (Norman Edward) (Appeal against conviction) [2001] 2 Cr. App. R 5 (CA (Crm. Div.)).

[19] Ibid at 25.

[20] United States v Frye 293 F 1013 (DC Cir., 1923).

[21] Ibid at 1014.

[22] [2004] 2 Cr App R 31.

[23] [2005] EWCA Crim 3559.

[24] (1984) 38 SASR 45(Australia).

[25] R v Luttrell at 32.

[26] R v Ciantar at 21.

[27] R v Harris [2005] EWCA Crim 1990; R v Clarke [1995] 2 Cr App R 425; R v Hodges [2003] EWCA Crim 290; R v Ibrahima [2005] EWCA Crim 1436; R v Stubbs [2006] EWCA Crim 2312; R v Luttrell [2004] EWCA Crim 1344; R v G [2004] EWCA Crim 1240.

[28] R v Parenzee [2007] SASC 143(Australia).

[29] Australian Law Reform Commission, Review of the Uniform Evidence Acts, Discussion Paper 69, 2005 at www.alrc.gov.au .

[30] R v Dallagher [2003] 1 Cr. App. R. 12.

[31] Ibid at 29.

[32] R v Clarke [1995] 2 Cr App R 425 per Steyn L J at 430.

[33] [2009] EWCA Crim 2698, [2010] 1 Cr App R 23.

[34] Ibid at 111.

[35] See R v George [2002] EWCA Crim 1923.

[36] See R v Sally Clark (No 2) [2003] EWCA Crim 1020 and R v Cannings [2004] EWCA Crim 1; WLR 2607(CA (Crim Div) ).

[37] National Research Council, ‘Strengthening forensic science in the United States: A part forward’ at http://ag.ca.gov/meetings/tf/pdf/2009_NAS_report.pdf.

[38] 293 F 1013 D.C. Cir. 1923.

[39] Ibid at 1014.

[40] Paul Rodrigues, ‘Towards a new standard for admission of expert evidence in Illinois: A critique of the Frye general acceptance test and an argument for adoption of Daubert’ (2009-2010) 34 Ill. U. L. J. 289.

[41] Robert Kohar, FDCC Quarterly/Spring 2007.

[42] Stephen Odger and James Richardson, ‘Keeping bad science out of the courtroom’ (1995) 18 U.N.S.W.L.J 108.

[43] P Giannelli, ‘Forensic science, Frye, Daubert and the Federal rules’ (1993) 29 Criminal Law Bulletin 428 at 432.

[44] Federal Rules of Evidence 702 at http://www.uscourts.gov/uscourts/RulesAndPolicies/rules/2010%20Rules/Evidence.pdf.

[45] Above at n40.

[46] 509 U.S. 579,(1993).

[47] Ibid at 592.

[48] Ibid at 590 and footnote 9.

[49] Ibid at 593 and 594.

[50] Ibid at 601.

[51] Sophia Gatowski et al., ‘Asking the gatekeepers: A national survey of judges on judging expert evidence in a post-Daubert world’ (2001) 25(5) Law and Human Behaviour 433 and Margaret Kovera et al, ‘Assessment of the common sense psychology underlying Daubert’ (2002) 8 Psychology Public Policy and Law 180.

[52] Alice Lustre, ‘Annotations, Post-Daubert standards for admissibility of scientific and other evidence in state courts’ (2001)90 A.L.R. 453.

[53] Edward K Cheng, and Albert H Yoon, ‘Does Frye or Daubert matter? A study on scientific admissibility standards’ (2005) 91 Va. L. Rev. 471.

[54] 522 U.S. 136 (1997).

[55] 526 U.S. 137 (1999).

[56] Above at n40.

[57] Above at n52.

[58] Note, ‘Admitting doubt: A new standard for scientific evidence’ (2009-2010) 123 Harv. L. Rev. 2021.  Also see Section 3 below under ‘Reliability of scientific evidence’.

[59] [1994] 2 SCR 9.

[60] Ibid at 13.

[61] Ibid at 14.

[62] Ibid at 17.

[63] Ibid at 18.

[64] Ibid at 18.

[65] [2000] 2 SCR 600.

[66] Ibid at33.

[67] Ibid at 25.

[68] Ibid at 50.

[69] [1953] S.C. 34.

[70] Ibid at 40.

[71] [2007] 1 SCR 239.

[72] Ibid at 1.                                                                     

[73] Ibid at 27.

[74] Ibid at 36.

[75] Gary Edmond, ‘Pathological science? Demonstrable reliability and expert forensic pathological evidence’ at www.attorneygeneral.jus.gov.on.ca accessed on 10 march 2013.

[76] Ibid.

[77] David Bernstein, Junk science in the United States and the Commonwealth’ (1996) 21 Yale J. Int’l L. 123.

[78] Stephen Odgers and James Richardson, ‘Keeping bad science out of the courtroom- changes in American and Australian expert evidence law’(1995) 18 U.N.S.W.L.J 108.

[79] Ibid.

[80] [1912] V.L.R 152.

[81] [1983] 46 A.L.R 493.

[82] Austl. Law Reform Commission Interim Report No. 26. At www.austlii.edu.au accessed on 11 march 2013.

[83] Evidence Act 1995 (Australia) at www.austlii.edu.au.

[84] Above at n40.

[85] (1960) 103 CLR 486.

[86] (1984) 38 SASR 45.

[87] Justice Peter McClellan, ‘Admissibility of expert evidence under the Uniform Evidence Act’, Judicial College of Victoria, Emerging issues in expert evidence workshop, Melbourne, 2009.

[88] Ibid.

[89] [1999] NSWSC 828.

[90] [2006] NSWCCA 167.

[91] Above at n 87.

[92] Gary Edmond, ‘Specialised knowledge, the exclusionary discretion and reliability: Reassessing incriminating expert opinion evidence’ (2008) 31 NSWLJ 1.

[93] R v Parker [1912] VLR 152 at 154.

[94] [1953] S.C. 34.

[95] Ibid at 40.

[96] The Law Commission (Law Com No 325), Expert evidence in criminal proceedings in England and Wales, (21 March 2011) at http://lawcommission.justice.gov.uk/docs/lc325_Expert_Evidence_Report.pdf.

[97] Ibid 5.17.

[98] Ibid at Page 32, 3.62.

[99] Ibid at page 65, 5.35.

[100] 509 U.S. 579,(1993) at 593-594.

[101] National Research Council, ‘Strengthening forensic science in the United States: A part forward’ at http://ag.ca.gov/meetings/tf/pdf/2009_NAS_report.pdf.

[102] Michael Lynch, ‘God’s signature: DNA profiling, the new gold standard in forensic science’ (2003) 27 Endeavour 93.

[103] Ibid.

[104] `Mark Jobling and Peter Gill, “Encoded evidence: DNA in forensic analysis’ (2004) 5 Nature Reviews: Genetics 739.

[105] Ibid.

[106] David Bentley and Peter Lownds, ‘Low template DNA’ (2011) 1 Archbold Review, 5.

[107] Ibid

[108] Ibid.

[109] Ibid.

[110] William Thompson, ‘Evaluating the admissibility of new genetic identification tests: Lessons from the “DNA wars”’ (1993/94) 84 J. Crim. L. & Criminology 22.

[111] Above at n101.

[112] Ibid at 149.

[113] Ibid at 16,98,108-109.

[114] Ibid at 88-89.

[115] Above at n110.

[116] Ibid.

[117] Ibid.

[118] Ibid.

[119] Ibid.

[120] Ibid.

[121] Ibid.

[122] Committee on DNA Tech. In Forensic Science, Nat’l Res. Council, DNA Technology in Forensic Science vii (1992).

[123] Ibid at 83.

[124] Bruce Weir, ‘The second National Research Council report on forensic DNA evidence’ (1996)59 Am. L. Hum. Genet. 500.

[125] 144 Misc. 2d 956, 545 N.Y.S 2d, 985 (Sup. Ct. 1989).

[126] John Caleb, ‘Beyond People v Castro: A new standard of admissibility for DNA fingerprinting’ (1991) 7 J. Contemp. Health & Pol’y 269.

[127] Ibid.

[128] Guidance- Interpretation of DNA evidence www.gov.uk/goverment/publication/the interpretation-of-dna-evidence.

[129] Ibid at 1.

[130] Brain Caddy, Graham Taylor and Adrian Linacre, ‘A Review of the Science of Low Template DNA Analysis’ (2008) at http://www.bioforensics.com/articles/Caddy_Report.pdf.

[131] Above at n128.

[132] [2007] NICC 49.

[133] Ibid at 46.

[134] Ibid at 46.

[135] Ibid at 62.

[136] Above at n130.

[137] [2009] EWCA Crim 2698, [2010] 1 Cr APP R 23.

[138] Ibid at 74.

[139] Ibid at 74.

[140] Ibid at 74.

[141] R v Broughton [2010] EWCA Crim 549 at 30.

[142] R v C [2010] EWCA Crim 2578 at 27.

[143] Ibid at 27.

[144] Alan Jamieson, ‘Case note - LCN DNA analysis and opinion on transfer: R v Reed and Reed’ (2011) Int’l J. Evidence & Proof 161.

[145] [2013] EWCA Crim 2 (CA Crim Div).

[146] Itiel Dror and Greg Hampikian, ‘Subjectivity and bias in forensic DNA mixture interpretation’ (2011) 51 Science and Justice 2004.

[147] Paul Giannelli, ‘Forensic symposium: The use and misuse of forensic evidence – Admissibility of scientific evidence’ (2003) 28 Okla. City U. L. Rev. 1.

[148] Ibid.

[149] The fingerprint Inquiry 2011, at www.thefingerprintinquiryscotland.org.uk chapter 2, page 44.

[150] Above at n101, page 144.

[151] Above at n149 page 48.

[152] Simon Cole, ‘Fingerprinting: The first junk science?’(2003) 28 Okla. City U. L. Rev.73.

[153] Lyn Haber and Ralph Haber, ‘Scientific validation of fingerprinting evidence under Daubert’ (2008) 7 Law, Probability and Risk 87.

[154] Jennifer Mnookin, ‘The validity of fingerprinting identification: Confessions of a moderate’ (2008) 7 Law, Probability and Risk, 127.

[155] Above at 101, Page 142.

[156] Above at n154.

[157] Above at n152.

[158] Ibid.

[159] Simon Cole, ‘Comment on ‘scientific validation of fingerprint evidence under Daubert’’. (2007) 7 Law, Probability and Risk 119.  Jennifer Mnookin, ‘The validity of fingerprinting identification: Confessions of a moderate’ (2008) 7 Law, Probability and Risk, 127.

[160] Simon Cole and Andrew Roberts, ‘Certainty, individualisation and subjective nature of expert fingerprinting evidence’ (2012) 11 Crim. L. R. 824.

[161] HM Advocate v McKie 1999.

[162] The Fingerprint Report / Scotland 2011 at www.thefingerprintreportscotland.org.uk.

[163] Ibid at 34.20.

[164] Ibid at 34.21.

[165] Above at n101, page 12 and 13.

[166] R v Buckley [1999] 163 JP 561 at 7.

[167] [2011] 2 Cr App R 16 at 61.

[168] Ibid at 61.

[169] Ibid at 61.

[170] Ibid at 63.

[171] Andrew Rennison and Gary Pugh, ‘Developing a quality standard for fingerprinting examination’ at www.homeoffice.gov.uk/agencies-public-bodies/fsr/.

[172] Ibid at 2.2.

[173] Ibid at 2.3.

[174] Iain Pretty and David Sweet, ‘A paradigm shift in the analysis of bite-marks’ (2010) 201 Forensic Science International 38.

[175] Adam Deitch, ‘An inconvenient tooth: Forensic odontology is an inadmissible junk science when it is used to “match” teeth to bite-marks in skin’ (2009) Wis. L. Rev. 1205.

[176] S Keiser-Nielsen, ‘Forensic odontology’ (1969) 1 U. Toledo L. Rev. 633.

[177] Paul Giannelli, ‘Bite-mark evidence’ (2007-2008) 22 Crim. Just. 42.

[178] Above at n175.

[179] Above at n177.

[180] Above at n174.

[181] Above at n177.

[182] I.A. Pretty and D.J. Sweet, ‘The judicial view of bitemarks within the United States criminal justice system ’(2006) 24 J. Forensic odonto-Stomatology 1.

[183] Above at n101, page 146.

[184] Ibid.

[185] Ibid at page 148.

[186] Ibid.

[187] Heikki Majamaa and Anja Ytti, ‘Survey of the conclusion drawn of similar footwear cases in various crime laboratories’ (1996) 82 Forensic Science International 109.

[188] Ibid at 119.

[189] Above at n101 page 149.

[190] [2010] EWCA Crim 2439, [2011] 1 Cr. App. R. 9.

[191] Ibid at 86.

[192] Ibid at 76.

[193] Ibid at 73 and 96.

[194] Ibid at 108.

[195] Ibid at 90.

[196] Lynn Meijer, Andrew Thean and George Maat, ‘Ear prints in forensic investigations’ (2005) 1:4 Forensic Sci. Med. Pathol. 247.

[197] Ibid.

[198] Simon Halpin, ‘What have we got ear then? Developments in forensic science: Ear Prints as identification evidence at criminal trials’ (2008) 8 U. C. Dublin L. Rev. 65.

[199] 97 Wash App 832, 988 P2d 977 (1999).

[200] Ibid at page 9.

[201] Ibid.

[202] [2002] EWCA Crim 1903, [2003] 1 Cr App R 12.

[203] Ibid at 199.

[204] Ibid at 201.

[205] Ibid at 200 and 202.

[206] Subsequent DNA testing showed that Dallagher could not have been the person responsible for the ear prints.

[207] [2008] EWCA Crim 975.

[208] Ibid at 28.

[209] Ibid.

[210] Ibid at 23.

[211] Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 US 579 - Supreme Court 1993 at 593.

[212] Richard Smith, ‘Peer review: Reform or revolution?’(1997) 315 British Medical Journal 759.

[213] Annette Boaz and Deborah Ashby, ‘Fit for purpose? Assessing research quality for evidence based policy and practice’ (2003) ECRC UK Centre for Evidence Based Policy and Practice: Working paper11 at http://www.kcl.ac.uk/sspp/departments/politicaleconomy/research/cep/pubs/papers/assets/wp11.pdf.

[214] Lesley Grayson, ‘Evidence based policy and quality of evidence: Rethinking peer review’ (2002) ESRC UK Centre for Evidence Based Policy and Practice (working Paper 7) at www.kcl.ac.uk.

[215] Ibid.

[216] Ibid.

[217] Ibid; Pearce published an article in the British Journal of Obstetrics and Gynecology in 1994 in which he claimed to have successfully transplanted an embryo from the fallopian tube to the uterus leading to delivery of a baby. The article was coauthored by Pearce’s head of department who was also the president of the Royal College of Obstetrics and Gynecology. Later the fraud was detected.

[218] Armstrong J S, ‘Discovery and communication of important marketing findings: evidence and proposals’ (2003) Wharton School, University of Pennsylvania: PA, oct 31 pp at <repository.upen.edu/cgi/viewcontent.cgi?article=1021> accessed on 10^th June 2013.

[219] Gary Edmond, ‘Judging the scientific and medical literature: Some legal implication of changes to biomedical research and publication’ (2008) 28(3) Oxford J. Legal Studies 523.

[220] Ibid.

[221] Focus: ‘what are the standards for quality research’ (2005) NCDRD Technical Brief number 9 at www.acddr.org/kt/products/focus/focus9/. Accessed on 15^th March 2013.

[222] Above at n219.

[223] Ibid.

[224] Hon. Grant Hammond, ‘The new miscarriages of justice’ (2006) 14 Waikato L. Rev. 1.

[225] Jon b. Gould and Richard A. Leo, ‘One hundred years later: Wrongful convictions after a century of research’ (2010) 100 (3) The Journal of Criminal Law & Criminology 825.

[226] Michael Saks et al, ‘Model prevention and remedy of erroneous convictions Act’ (2001) 33 Ariz. St. L. J. 669.

[227] Above at n2.

[228] Justice May, Second Report: Return to an address of the Honourable the House of Commons dated 3^rd December 1992 for a report of the Inquiry into the circumstances surrounding the convictions arising out of the bomb attacks in Guilford and Woolwich in 1974 (1992).

[229] Ibid at 8.1 and 8.9.

[230] Ibid at 9.5.

[231] Ibid at 10.11.

[232] Ibid at 11.16.

[233] Editorial, ‘Miscarriages of justice again’ (1990) Crim. L. R. 613.

[234] Above at n228 at 11.14.

[235] Ibid at 14.2.

[236] Ibid at 14.6.

[237] Gary Edmond, ‘Constructing miscarriages of justice: Misunderstanding scientific evidence in high profile criminal appeals’(2002)22 Oxford journal of legal studies 53.

[238] Justice May, Final Report: Return to an address of the Honourable the House of Commons dated 3^rd December 1992 for a report of the Inquiry into the circumstances surrounding the convictions arising out of the bomb attacks in Guilford and Woolwich in 1974 (1994).

[239] R v Sally Clark (No 2) [2003] EWCA Crim 1020.

[240] Ibid at 93.

[241] Ibid at 169.

[242] Ibid at 134.

[243] Ibid at 135.

[244] Ibid at 96.

[245] Ibid at 97.

[246] [2004] EWCA Crim 1; WLR 2607(CA (Crim Div) ).

[247] Ibid at 129.

[248] Ibid at 129.

[249] Ibid at 133.

[250] Ibid at 137.

[251] Ibid at 137.

[252] Ibid at 141 and 145.

[253] Ibid at 147.

[254] Ibid 160.

[255] Ibid 178.

[256] Ibid 179.

[257] Adam Wilson, ‘Expert testimony in the dock’ (2005) 69 Journal of Criminal Law 330.

[258] Above at n225.

[259] Above at n257.

[260] Ibid.

[261] David Goodstein, ‘How science works in Reference Manual on Scientific Evidence 67, 80-82 (Federal Judicial Centre Ed. 2^nd ed. (2000).

[262] Michael C. Mason, ‘The scientific evidence problem: A philosophical approach’ (2001) 33 Ariz .St. L.J. 887.

[263] Above at n226 and above at 225.

[264] Roger Koppl, “How to improve forensic science’ (2005) E.J.L. & E. 256.

[265] Ibid.

[266] Above at n226.

[267] Above at n101.

[268] Ibid at 19.

[269] Ibid at 22.

[270] Ibid at 24.

[271] Ibid at 25.

[272] Ibid at 26.

[273] P Giannelli and E Imwinkelried, ‘Scientific evidence: the fallout from the Supreme Court’s decision in Kumho Tire’ (2000)14 Crim.Just.12.

[274] House of Commons Science and Technology Committee; The Forensic Science Service- Seventh Report of session 2010-2012 at http://www.publications.parliament.uk/pa/cm201012/cmselect/cmsctech/855/855.pdf, accessed on 24/6/2013. The FSS was closed down on31st March 2012.

[275] Ibid.

[276] Ibid.

[277] Ibid

[278] Ibid

[279] Proposed bill ‘The Forensic Science and Standards Act of 2012, S. 3378’ http://thehill.com/images/stories/blogs/flooraction/jan2012/s3378.pdf accessed on 25 march 2013.

[280] Jon Robins, ‘The fall-out of closing the Forensic Science Service’ (2012) 176 Criminal Law & Justice Weekly 107.

[281]Law Commission, Expert Evidence in Criminal proceeding in England and Wales (Law Com no. 325), March 2011 at 6.41; 6.78 and 6.79.

[282] Ibid. (Criminal Procedure Rules 2012, Part 33 already provide Judges with this power).

[283] Ibid at 6.45- 6.46.

[284] Samuel Gross, ‘Expert Evidence’ (1991) Wis. L. Rev. 1113.

[285] Karen Reisinger, ‘Court- appointed expert panels: A comparison of two models’ (1998) 32 Ind. L. Rew. 225.

[286] Michael Saks, ‘The Phantom of the Courthouse’ (1995) 35 Jurimetrics 233.

[287] Ibid. In the US most often, such expert witnesses when used were for civil cases rather than for criminal cases.

[288] Above at n284.

[289]Ibid

[290] M.N. Howard, ‘The neutral expert: a plausible threat to justice’ (1991) Feb Crim. L. R. 98.

[291] J.R.Spencer, ‘The neutral expert – an implausible bogey’ (1991) Feb Crim. L. R. 106.

[292] Ibid.

[293] Peter Allridge, ‘Scientific expertise and comparative criminal procedure’ (1999) 3 3 Int’l J. Evidence & Proof 141.

[294] Gary Edmond, ‘Advice for the courts? Sufficiently reliable assistance with forensic science and medicine (Part 2), (2012) 16 Int’l J. Evidence & Proof 263.

[295] Ibid

[296] Ibid

[297] Ibid

[298] Above at n293.

[299] Ibid.

[300] Beth Riffe, ‘Comment, The aftermath of Melendez: Highlighting the need for accreditation-based rules of admissibility for forensic evidence’ (2010) 27 T. M. Cooley L. Rev. 165.

[301] Above at n7.