Introduction
38.1. The terms in which any expert witness expresses his opinion are bound to have an impact on the court, as the National Academy of Sciences has said: "Many terms are used by forensic scientists in scientific reports and in court testimony that describe findings, conclusions, and degrees of association between evidentiary material (e.g. hairs, fingerprint, fibres) and particular people or objects. Such terms include, but are not limited to 'match', 'consistent with', 'identical', 'similar in all respects tested', and 'cannot be excluded as the source of'. The use of such terms can and does have a profound effect on how the trier of fact in a criminal or civil matter perceives and evaluates scientific evidence."1
38.2. The National Academy of Sciences has also highlighted that "subjectivity is intrinsic to friction ridge analysis"2 and the findings of the Inquiry have shown that a complex range of variables interact in the formation of the opinion of a fingerprint examiner.3 These inescapable propositions have significant implications for the terms in which the opinion of a fingerprint examiner can be properly presented to the court.
The expression of opinion
Context
38.3. The context in which expert evidence can legitimately be led in court is relevant to the terms in which it is expressed.
38.4. The circumstances in which it is legitimate to lead expert evidence in a Scottish court have been discussed in chapter 30. It must relate to some matter that is beyond the ordinary knowledge and experience of a judge or juror. A recent case illustrating the point is Gage v HMA4 in 2011. Mr Gage was convicted of murder in a shooting that occurred at about 22:00 on a night in March. The evidence against him included eyewitness evidence. The widow of the deceased identified him for the first time in court as resembling the gunman, the resemblance being in his "scary eyes". She also identified clothing that the gunman was wearing. The clothing had been found in a Saab car that a second witness said was similar to the getaway car and Mr Gage's DNA was found on that clothing. On appeal the defence sought to lead evidence from a professor of psychology who had expertise in the reliability of eyewitness evidence but the court held that it was not appropriate to lead evidence from an expert when the issue was the credibility and reliability of eyewitness evidence. The potential for eyewitness evidence to be unreliable is notorious but ordinary experience enables judges and jurors to assess the reliability of such evidence by considering factors such as the limited opportunity that the witness had to observe the person or the object in question, the time of day and the quality of the light. It was not necessary to supplement ordinary experience with expert evidence to enable the jury to assess the reliability of eyewitness evidence and therefore it was not legitimate to lead expert evidence on this issue.
38.5. Eyewitnesses will regularly be asked to express an opinion on the chances of error in their identification but, again, readily familiar factors (such as familiarity with the accused, the brevity of the opportunity to observe him and the prevailing lighting) will enable the judge or jurors to form an independent assessment of the reliability of the evidence irrespective of the witness's self-declared level of certainty. In Jenkins v HMA an eyewitness identification that the witness said was "100% sure" was rejected as unreliable for the common sense reason that the witness had on various occasions identified three different people as the assailant: "One might suggest that a person who is 100% sure that three different persons were witnessed by him doing something for which only one person was responsible is one upon whom little or no trust or confidence can be placed."5
38.6. The complexities associated with fingerprint pattern recognition are beyond ordinary experience and therefore it is legitimate to lead evidence from a fingerprint expert. By the same token, the very fact that such evidence is beyond ordinary experience restricts the ability of the judge or the jury independently to assess the reliability of such evidence. Without a careful exposition of the relevant factors by the expert neither judge nor jury may be able to assess the reliability of fingerprint evidence, and in particular to check whether the witness is justified in any claimed level of certainty. Even then, there is always the innate tendency to defer to expertise, so the expert has to be accurate in the terms in which evidence is given and, in particular, accurate in any claim to any degree of certainty attaching to the conclusion.
100% certainty
38.7. Fingerprint examiners give evidence that marks can be 'matched' to the level of 'individualisation', by which they mean that the mark can be attributed to one single person in the whole world throughout human history. That is a conclusion that they profess to express to the standard of '100% certainty'.
38.8. The term '100% certainty' is open to interpretation. It could mean:
(i) fingerprint evidence is 100% reliable or infallible;
(ii) the examiner is himself 100% sure that his conclusion is correct;
(iii) the examiner is 100% sure that another examiner would agree with his opinion, whether or not that includes seeing the same similarities.6
38.9. There have been claims to infallibility and even in the trial in HMA v McKie Ms McBride advanced a variant that it was not possible for a combination of five examiners to make a mistake.7 The claim to infallibility is wrong because fingerprint evidence is opinion evidence and there have been instances of erroneous identifications. Even where multiple examiners have been involved in verification an error can be made. The first interpretation of '100% certainty' is not one that can be justified.
38.10. Fingerprint examiners working for SPSA continue to express themselves in terms of 100% certainty and, as understood by them, that denotes a combination of the second and third interpretations. That is what they continue to be taught in training.8
38.11. There is evidence of significant variation of opinion among examiners and, therefore, one examiner cannot possibly know for certain what all other examiners might say. Consequently, the third proposition is one that cannot necessarily be substantiated.
38.12. As for the second, the requirements for scientific validation of any forensic science methodology have to be borne in mind:
"The determination of uniqueness requires measurements of object attributes, data collected on the population frequency of variation in these attributes, testing of attribute independence, and calculations of the probability that different objects share a common set of observable attributes. Importantly, the results of research must be made public so that they can be reviewed, checked by others, criticized, and then revised, and this has not been done for some of the forensic science disciplines."9
38.13. The ability of any examiner to 'individualise' without the potential for any error at the claimed level of one person in the whole of human history is not scientifically validated. Fingerprint examiners do not presently base their conclusions on validated statistics of the incidence of variation in friction ridge details in the population. Their opinions on 'sufficiency' are derived from personal assessments founded in training and personal experience.10 The second proposition is, accordingly, one that cannot be substantiated.
38.14. In short, no one of the three interpretations is justifiable and, hence, the claim to '100% certainty' is not one that should be advanced. That is consistent with the evidence of Mr Nelson, Mr Chamberlain and Professor Champod.
38.15. Mr Nelson explained that in forensic science the only evidence about which there can be 100% certainty is physical fit evidence, for example where one can physically fit two pieces of wood together. Other evidence is opinion evidence. One may be confident in the result but never 100% certain.11
38.16. Mr Chamberlain explained that individualisation is not achievable on a scientific basis. Examiners give a personal opinion that a mark can only have come from one particular individual. What the examiner is doing is dismissing the likelihood that the source of the mark could be another person because the likelihood is so diminishingly small but that conclusion is a matter of personal opinion and not something founded on scientific principle.12
38.17. Professor Champod said that when an examiner claims certainty he is making what has been described as a 'leap of faith'. At some point, the examiner has identified so many corresponding features that he becomes subjectively convinced that the chance of duplication is zero. He went on: "This little jump from a probabilistic position to certainty escapes the science. It has been accepted by courts for years but it escapes a logical argumentation."13
Commentary
38.18. It is important that the limitations inherent in the subjective nature of fingerprint evidence are clearly understood and that it is appreciated that it is not infallible. It can also no longer be claimed to be a matter of 'practical infallibility' because, as understood by Lord Justice General Clyde in HMA v Hamilton,14 even that claim was predicated on long experience showing that the presumption in the truth of this type of evidence has never been successfully rebutted. The claim to '100% certainty' cannot be substantiated and should not be made.
38.19. Fingerprint trainees should not be taught '100% certainty'. Their training should proceed on the premise that fingerprint evidence is opinion evidence. As discussed in chapter 35, recognition that this is 'opinion' evidence is not an end in itself. It opens up a series of consequential questions disclosing the complex nature of the series of subjective judgments that are involved in the process and the variability of the materials with which they are required to work. Trainees and qualified examiners alike require to be conscious at all times of the variables that are involved and the potential for differences of opinion and, indeed, error.
Standard applied in arriving at conclusion
38.20. Removing the claim to '100% certainty' should not be understood to imply that the fingerprint examiners should lower the standard to which they work. They should not. The distinction can be seen in the 2011 OIG review:
"Historically, latent fingerprint examiners expressed identification conclusions in terms of '100% certainty', with a zero likelihood that the latent fingerprint was made by a different person. Although the FBI Laboratory has not lowered the standard required to make an identification, examiners no longer testify that they are '100% certain'. Instead, examiners testify that they are confident in the conclusion, would not expect to see the same amount of information repeated if the fingerprints originated from different people, and find no physical evidence causing them to doubt that the fingerprints are from the same source."15
'No doubt'
38.21. The style of joint report that has been in use since 2006 expresses the conclusion in these terms: "Each result has been subject to a verification process and we have no doubt our conclusions are as stated."16 That is a conclusion expressed by examiners who have been taught to expect that they can have 100% certainty in their conclusion and when expressing 'no doubt' that is the standard that they imply; they mean absolutely no doubt. Accordingly, as currently used by an examiner with that training it is truly a synonym for '100% certainty'.
38.22. The formulation used in the SPSA style report and the terms in which FBI witnesses testify both use variants of the expression 'no doubt'. It is not my intent that this Inquiry be prescriptive as to the language that a witness may use. The point that is being made is of more substantive significance. The objection to '100% certainty' and any synonymous expression is an objection to the claim to scientific accuracy, which cannot be substantiated.
38.23. The legal background has been surveyed in chapter 30. '100% certainty' is not a precondition for the admissibility of expert evidence. Proof to the criminal standard does not require mathematical certainty.17 In testifying on the basis of '100% certainty' fingerprint examiners are, at one and the same time, addressing a standard not set by law and overstating their evidence in order to meet it. An expert witness can give evidence where there is room for doubt about the conclusion and, indeed, if there is room for doubt the expert is obliged to disclose that fact. An expert witness may express an evaluative opinion, that is to say the witness may explain the strength of the opinion and may do so even where (as in the case of fingerprint evidence) there is no statistical basis for the evaluation. But the facts that there is no supporting statistical database and that the assessment is subjective and based on experience must be clearly disclosed enabling the evidence to be assessed accordingly. As the Court of Appeal put it in R v Reed and Reed, "care must be taken to guard against the dangers of that evaluation being tainted with the verisimilitude of scientific certainty."18
38.24. A fingerprint witness has to start from the premise that fingerprint evidence is opinion evidence. If the strength of the personal opinion is not stated in the report the witness may be asked about it in court and will answer as that witness sees fit. What matters more than the choice of language (whether the witness says that he is 'confident', 'sure', 'certain' or 'in no doubt') is the transparency of the opinion.
Evidence on a scale
38.25. The question arises whether fingerprint examiners could adopt the practice of certain forensic scientists, such as the expert in facial comparison in R v Atkins and Atkins and the footwear mark expert in R v T,19 and express the strength of their conclusion by using a sliding scale ranging from, for example, 'consistent with' to 'lends powerful support' to the particular proposition.
38.26. There are two practical objections to the adoption of a sliding scale in relation to fingerprint evidence in current practice.
38.27. The first is that fingerprint examiners do not currently reason in terms of any gradation in the probability of their conclusions. The two positive findings that they can make are either (1) an identification or (2) an exclusion and in each case the decision is categoric. Where the examination results in anything less than a categoric decision the finding is 'inconclusive'.20 To adopt any language suggestive of a gradation in the probability of a match or exclusion would be misleading.
38.28. Secondly, as it currently stands, there is no statistical base that would provide a reliable basis on which a fingerprint examiner could assign a quantifiable value to his conclusion on a scale of probability.21 Statistical models are under development and are discussed in chapter 41. Until such time as a statistical model is validated there is no basis upon which the chance of a match between mark and print can be expressed (or assessed) on a scale of probability, whether in percentage terms or by use of some conventional or descriptive language. There is no alternative but to present fingerprint evidence as being based on subjective opinion informed by training and personal experience.22
Training for court
38.29. Subjectivity is manifest throughout the comparison process, no more so than at the final inferential step when the examiner is considering whether there is 'sufficient' matching detail for a conclusion of identity. Examiners operate to a personal threshold.23 This can, therefore, be anticipated to be a fertile area for cross-examination and this has arisen in training at SPSA and the NPIA. In his evidence Mr McGinnies approached it from two different perspectives.
38.30. The first is what is understood by 'sufficiency' and that was explained as a quantitative and a qualitative analysis:
"You are looking at the whole mark. So it would be when you are satisfied... [you] have enough, sufficient information of quality and quantity within the two to effect an identification. Depending on the mark, it may take longer or may take more to convince them but it could be very early in the comparison or, depending on the complexity of the mark, it may take longer to work to get right across the mark and make sure that everything is in sequence and agreement. So, depending on the clarity of the mark and the tolerance level, will depend on how soon they came to their conclusions.
... When are you satisfied? What is satisfied?
The answer to those would be by looking at each mark on its own merit, by studying all the details, all the features that are present, by taking into account all your information, all your analysis, and it is when you are satisfied to that conclusion of identification that anyone else coming behind you with the same training, qualifications and experience will come to the same conclusion you have effected an identification. So it sounds [inaudible] to say it depends on the mark but that is what they are trained when they are satisfied."24
38.31. The second perspective was approaching it as a question aimed at ascertaining the minimum number of points required for an identification:
"It is something that they are taught at NPIA as well very early on is to - it is something that in court and in mock courts especially we try to hypothetically take people down the line of 'could you identify on'. So they have got to be aware that the answer has to be 'depends on the mark' and they can only talk about the mark that they have identified in this case. They should not really be drawn into going down a way, a line. They have got to talk about what they are there in court for that day to.
Q. Just following that theme, if a defence agent does ask, 'Well, would you identify on four points?' is there any reason why he is not entitled to a straight answer?
A. No, certainly not but the answer would be I would need to see the four points. Hypothetically could I identify four points? Yes, if they were clear, if there were other features in sequence and agreement, yes, but it would need to be qualified by, 'Yes but I would need to see the mark'. You would need to see the four points. But there would never be an evasion. There would be no evasive answer of, 'No, I could not answer that'. They would give their opinion obviously."25
38.32. The answer that 'sufficiency' depends on the mark is circular and unilluminating. The same circularity is evident in textbooks.26 The natural consequence of the abandonment of the numerical standard is that 'sufficiency' cannot be defined in terms of a prescriptive number of characteristics (level 2 or level 3) and will depend on the clarity of the mark and the individual examiner's perception of the rarity of the observed pattern. In the absence of validated statistics for the incidence of level 2 detail and proof of the reproducibility of level 3 detail, examiners can truly only speak in terms of their personal perception of rarity.27 The judgment is an empirical one that may defy precise definition.
38.33. The inability to give a definitive explanation of 'sufficiency' does not render fingerprint evidence inadmissible. The legal analysis in paragraph 23 is relevant: the evidence remains admissible but fingerprint examiners must be more transparent and, to that end, more descriptive of not only the methodology that they apply in general (emphasising its subjective base) but also the particular judgments that they have made when identifying the mark which is the subject of their evidence and the lack of a statistical base for the conclusion of 'sufficiency'.
38.34. The 'sufficiency' of the observed common characteristics to establish unique identity is mentioned only as one example. Consideration of the five questions in chapter 35 shows that a fingerprint examiner's conclusion is not based on a single subjective judgment but a series of judgments each of which may be a matter of personal opinion. It is recognition of the empirical nature of the series of judgments that necessitates (1) conspicuous care when each examiner is conducting a comparison; (2) appropriate verification and (3) clear exposition when evidence is being given in court.
38.35. Each examiner, when carrying out the 'ACE' components of ACE-V, has to be alert to the subjective variables that may affect their conclusion, including the quality of the mark, the scope for variability in the interpretation of incomplete detail, the degree of tolerance being applied and the robustness or cogency of the explanation for any differences and must carry out a critical evaluation of the reliability of the conclusion having regard to these variables and mindful of the potential for error.
38.36. The empirical nature of the series of judgments also provides the rationale for a requirement for independent verification as an appropriate check against unjustified subjective judgments by a single individual.
38.37. Finally, these same considerations support the need for a more transparent and expansive narrative in reports prepared for court, with each examiner explaining the judgments that he has made in arriving at his conclusion.
38.38. Mention has been made in chapter 3528 of the potential for different interpretations of the ridge detail that SCRO identified as SCRO 4-6 in Y7 and SCRO 1, 10, 16, 11 and 12 in QI2 Ross. Where ridge detail in a mark (or for that matter in the print: for example, SCRO 4 in Ms McKie's print) lacks clarity and is open to a number of interpretations, the examiner should disclose in the report that the ridge detail is open to interpretation, should explain the reasons for the preferred interpretation and draw attention to the implications if any of the alternative interpretations.
38.39. Opinions should be expressed in a way that accurately reflects the limitations of fingerprint methodology and not in terms that could be misleading. There is a danger in examiners being taught stereotyped answers to cross-examination, such as responding to questions about the general approach to 'sufficiency' by saying that they can only talk about the mark that they have identified in the particular case. Such answers appear defensive, designed to avoid discussion of the limitations of the underlying methodology, in particular the fact that a personal threshold has been applied by the witness. These are facts which may be material to the reliability of the conclusion.
38.40. The legal profession (including the judiciary) has contributed to the complacency concerning fingerprint evidence in court. This is part of the wider issue of deference to expert witnesses. As Ashbaugh has commented, in the past challenges to fingerprint evidence may have been "haphazard at best, usually ill-prepared".29 I have recommended that those involved in the criminal justice system should recognise fingerprint evidence as opinion evidence that needs to be assessed on its merits.30 Where the mark is very clear there may be little scope for difference of opinion and fingerprint examiners may not face questioning but where the mark lacks clarity31 they should expect in the future that they may be cross-examined by lawyers with more insight into the methodology of fingerprint comparison work as a result of this Report and fingerprint examiners need to be prepared to answer more searching questions.
38.41. Fingerprint examiners need to be aware of their responsibilities to the court as an expert witness. COPFS has, since the Inquiry hearings, produced a guidance booklet for SPSA expert witnesses, covering the role of the expert witness and disclosure.32 The expert's duties to the court have been discussed in general terms in chapter 30 but at a practical level the problem lies in the application of those general principles to the subject in hand: how should fingerprint examiners approach the presentation of their evidence in a written report and evidence in court?
38.42. The short response that a fingerprint examiner should disclose the limits to the discipline of fingerprint comparison, and hence the limits of the opinion being expressed, leads to the challenging statement by Judge Edwards:
"My concern is that some forensic practitioners may not know what they do not know about the limits of their discipline; they will have to be taught this so they can be circumspect in their testimony."33
38.43. Professor Champod has made the point that it is unfair to criticise fingerprint examiners for the lack of transparency in the explanations that they have provided hitherto. Fingerprint comparison is "a field that practitioners learned as straightforward and unquestionable because of its inherent quality."34 Moving away from a premise of '100% certainty' is more than a stylistic change. The move is intended to reflect the facts that fingerprint comparison is not straightforward and the conclusions not necessarily unquestionable. Exposing those facts represents a fundamental shift in the basis on which fingerprint evidence is presented. Further research is required to ascertain the limits of the discipline or technique but meantime practitioners require to adjust to working "under conditions of uncertainty".35 The teaching that Judge Edwards mentions as needed is most likely to come from academics working in the field.
38.44. An emphasis needs to be placed on the importance of not only learning the methodology of fingerprint work, but also of engaging with members of the academic community working in the field. That need begins at the point of training36 but is of continuing relevance throughout professional development.
38.45. As has been said on a number of occasions in this Report, there is no reason to believe that fingerprint evidence lacks reliability. Nonetheless, unjustified claims should not be made for the discipline and transparency is required.
Consequential implications of fingerprint evidence being recognised to be opinion evidence
38.46. Recognition of the fact that fingerprint evidence is opinion evidence has a number of practical implications. The first concerns the range of individuals who may be considered to be suitably qualified to testify as an 'expert' witness in relation to fingerprints. That is addressed in chapter 40. Other practical consequences are considered here.
Open discussion of rival opinions
38.47. One of the side-effects of the belief in 100% certainty (particularly when expressed as a belief that any other competent examiner would agree) is that it leaves no room for an alternative view:
"In the past the friction ridge identification science has been akin to a divine following. Challenges were considered heresy and challengers frequently were accused of chipping at the foundation of the science unnecessarily. The cultish demeanour was fostered by a general deficiency of scientific knowledge, understanding, and self-confidence within the ranks of identification specialists. A pervading fear developed in which any negative aspect voiced that did not support the concept of an exact and infallible science could lead to its destruction and the destruction of the credibility of those supporting it."37
38.48. Tenable differences of opinion come to be seen in terms of a 'dispute' and examiners can believe that those who take a different view are in 'the opposite camp' and can only be wrong, incompetent or even dishonest. Y7 came to be a regrettably extreme example of this tendency. This can lead to unjustified personalised attacks being made in various media that can discredit the discipline itself.
38.49. Belief in '100% certainty' should no longer be taught not only because it overstates the degree of certainty that can be attached to the evidence in the eyes of the court but also because it is inimical to open discussion between practitioners. Fingerprint examiners address and assess numerous matters with a subjective dimension in forming their opinions about comparisons, including image selection, the observation and interpretation of characteristics in the mark, the treatment of differences between the mark and print and the sufficiency of information for an identification. It is possible for competent examiners to formulate differing opinions about the same mark and print38 and the fingerprint community requires to be receptive to differences of opinion.
38.50. The relevance of this within a bureau, where there requires to be an open environment in which practitioners are encouraged to air doubts and to challenge the views of others as necessary during the ACE-V process, is discussed in chapter 36. However, the point is broader. The fingerprint community generally has to be more transparent and challenging and to move away from what Ashbaugh termed the 'cultish demeanour'. Practitioners have a significant contribution to make to the adjustments required to present fingerprint evidence realistically in a world no longer simplified by '100% certainty' and in order to do that they must begin by recognising that opinions can differ and progress from that to investigate the sources of variability and hence identify the factors that should be taken into account by a fingerprint examiner in reaching a conclusion on a comparison and by the court (judge or jury) when assessing the evidence. There requires to be a healthy debate across the profession free from any imputation that the expression of differences of opinion gives rise to issues relating to personal professional competence or undermines the reliability of the discipline.
Erroneous identifications
38.51. The September 2000 HMICS report recommended that a national policy be established to deal with all 'erroneous' fingerprint identifications.39
38.52. I endorse that recommendation subject to two observations.
38.53. The first is that care has to be taken in relation to the definition of 'erroneous'. The mere fact that a difference of opinion occurs does not imply that any 'error' has been made. A clear distinction has to be drawn between cases where there is scope for a difference of opinion and cases where there may have been a mistaken fingerprint analysis or a breach of accepted procedures. Excessive investigation of differences of opinion among examiners could prejudice the objective of promoting an open culture where practitioners are encouraged to air doubts and differences and is to be avoided. There is no need for investigation where opposing views are tenable and an examiner who identifies when others are not persuaded of the match should not be branded as having made an 'erroneous' identification merely by virtue of the fact that others disagreed. The need is to investigate cases where a wrong finding of identification (or for that matter exclusion) displays a mistake in the application of accepted procedure or proper fingerprint analysis or evaluation.
38.54. The second observation relates to the objectives of such an investigation. The article by Simon A. Cole, 'More than zero: accounting for error in latent fingerprint identification',40 shows that little is known worldwide about instances of error in fingerprint comparison work. One of the examples mentioned by Cole was the case of Lee and the Inquiry has found that prior to SCRO handling Y7, Nottinghamshire Police had made a misidentification that was investigated by the Metropolitan Police. While the report, written in 1992, does not say what the cause of the error was, it did comment adversely on the practice of verifiers being presented with charted comparisons on a comparator machine or photographic enlargement and recommended that verifiers reach conclusions independently of the findings of any other examiners.41 Inappropriate use of the comparator machine was a weakness in the working practices at SCRO five years later that may have contributed to the error in relation to Y7.42 Other factors implicated in the erroneous identification of Y7 and QI2 Ross (failing to pay due attention to the quality of points in similarity, failing to give a satisfactory explanation for differences and reverse reasoning) were to recur later in the FBI's misidentification in the Brandon Mayfield case. Errors made in the handling of ordinary casework can provide valuable lessons not only for the fingerprint examiners directly involved in the immediate case but also for the fingerprint community as a whole. Erroneous identifications (or exclusions) need to be investigated to ensure that any lessons to be learned are identified and, as appropriate, drawn to the attention of the wider fingerprint community. SPSA will require to consider whether procedures require to be revised in the light of the findings of an investigation within any of its own bureaux and, for that matter, in the light of any report from any other country such as the OIG Mayfield reports.43
The range of conclusions and 'unable to exclude'
38.55. By convention44 fingerprint examiners are understood to restrict their findings to categoric statements of identity or exclusion, with anything less being said to be inconclusive.45 The conclusions were constrained by resolutions of the IAI from 1979 and 198046 but those resolutions were rescinded by IAI Resolution 2010-18 of 16 July 2010, passed after the Inquiry hearings.47
38.56. That statement of the conventional position itself requires explanation. Reference to the 2011 OIG report reveals some uncertainty about the significance of the 'inconclusive' category as used by the FBI.48 Two refinements are required in relation to Scotland. The first is that fingerprint examiners begin by determining whether a mark is of comparable quality. If it is not it is set to the side as 'fragmentary and insufficient' and is not subject to comparison. The categories of finding as (a) an identification, (b) an exclusion or (c) inconclusive would be applied to marks that have been compared. The second refinement is that in the past Scottish practitioners (in common with their UK colleagues) were prepared to report to police where they have a 'strong suspicion' as to identity, though they would not speak to that in court.49
38.57. The practical issue is whether evidence ought to be given in court where there are reasons for suspicion of either identification or exclusion.
38.58. It might be thought that there is a continuum between exclusion and identification. On one side of the continuum there is a varying degree of possibility of exclusion and, on the other side, an increasing possibility of a match.50 At present in Scotland fingerprint examiners do not testify in court to any finding short of certainty as to unique identity or exclusion, for example where there is ridge detail consistent with a person but not sufficient to individualise. The same practice applies in Northern Ireland. PSNI would inform the police for intelligence purposes if they found ridge characteristics in agreement with an individual but not enough to identify, but Mr Logan had not given evidence in court on this basis.51
38.59. The position is different in England and Wales. Mr Chamberlain had twice given evidence that a mark was possibly made by a person although the examination was inconclusive. He described the judgment as to the likelihood of a match in those cases as "intuitive". In one case there was a limited number of persons who could have entered the relevant locus with the result that he did not require to establish uniqueness to the normal level of individualisation, only to discriminate among members of a limited class.52
38.60. Mr Zeelenberg said that he regularly gives evidence in the Dutch courts where his findings fall below the standard required for identification but he is unable to exclude a relevant person.53 His situation is not directly analogous to the UK because in the Netherlands fingerprint examiners operate to a numerical standard.
38.61. Mr Nelson was of the view that it would be possible to give evidence that was not about identification, for example that a mark was consistent with an individual. One would need to be very careful that it did not mislead the court in any way. Care would have to be taken to ensure that it was not presented so as to appear more conclusive than it was.54
38.62. Discussion of this matter with Mr Chamberlain suggested that the response that a fingerprint examiner might be expected to give might vary depending on the terms of the question. Fingerprint examiners might be uncomfortable expressing an opinion on the likelihood of a match short of unique identity, but might be more comfortable answering a question whether or not they could exclude a particular person.55
38.63. Mrs Tierney told the Inquiry that she had become aware that in England and Wales reports did include a category of 'unable to exclude' and that was a development that SPSA's Scientific Advisory Group required to discuss.56
38.64. Mr Pattison explained that 'unable to exclude' evidence might assist the prosecution or the defence. Such evidence may provide corroboration. It was possible to foresee a situation where the Crown would seek to use 'unable to exclude' evidence to corroborate a high quality eyewitness identification. 'Unable to exclude' evidence could also be relevant to the defence. It might, for example, link a mark to an incriminee (i.e. a person the defence alleges may have committed the crime). The Crown would want to be made aware of and to have access to such evidence.57
38.65. Mr Chamberlain's evidence was that by not reporting such evidence fingerprint examiners may be denying the courts a valuable source of evidence.58 However, Mr Pugh was not convinced that there was a significant amount of additional evidence that was being missed as a result of current practice.59
38.66. Mr Pugh also highlighted the difficulty in interpreting such evidence. The statement that an individual cannot be excluded may inevitably lead to the question what the probability of a match is and a fingerprint examiner would be uncomfortable answering that question because there is no scale enabling the examiner to express an opinion as to the degree of similarity.60 All that the examiner could do would be to place the raw evidence (i.e. the points in agreement) before the court, leaving the court to assess its value with the risk that the court may over or under-estimate its value.61
38.67. That echoes the reservation expressed by Ashbaugh in 'Quantitative-Qualitative Friction Ridge Analysis':
"The so-called probability identifications of friction ridge prints is extremely dangerous, especially in the hands of the unknowing. The insufficiently detailed print still has only one possible source of origin, but a reliable means of determining the probability as to whether the examiner would be correct or incorrect is, as yet, unavailable."62
38.68. That, though, goes to the core of any fingerprint evidence because, as matters currently stand, examiners do not have data to support the probability of a unique identification and the gap between a finding of 'unable to exclude' and 'identity' could be narrow and bridged only by the 'leap of faith' of which Professor Champod spoke.63
Commentary
38.69. The possibility of using statistical models to quantify the likelihood of a match is discussed in chapter 41. The question is whether 'unable to exclude' evidence should be led in court before a statistical model has been validated.
38.70. A number of issues do arise.
38.71. It is a matter for the prosecution, defence and, ultimately, the court whether such evidence may have relevance in any particular case. I was advised that SPSA's Scientific Advisory Group was to discuss the matter.64 I would not preclude evidence of this type being introduced, particularly in exceptional circumstances of the kind mentioned by Mr Chamberlain, but more general deployment of 'unable to exclude' evidence requires careful consideration of the first two issues that I have mentioned; and certainly if such evidence were to be led the examiner would have to give a full explanation of its limitations.
Recommendations
Differences of opinion between examiners
38.72. Differences of opinion between examiners should not be referred to as disputes.
Erroneous identifications or exclusions
38.73. The SPSA should investigate all 'erroneous' fingerprint identifications or exclusions.65
38.74. Cases where there is scope for a difference of opinion should not be classified as 'erroneous'. The cases that merit investigation are where there may have been a mistaken fingerprint analysis or a breach of accepted procedures.
38.75. The SPSA should consider whether their procedures require to be revised in the light of the findings of an investigation of an erroneous fingerprint identification or exclusion.
38.76. Reference is also made to the recommendation in chapter 41, paragraph 45, on the need for SPSA to monitor developments in other countries and to review its procedures accordingly.
Expression of opinion by examiners
38.77. Examiners should discontinue reporting conclusions on identification or exclusion with a claim to 100% certainty or on any other basis suggesting that fingerprint evidence is infallible.
38.78. In chapter 35, paragraph 136, I have recommended training to enable examiners to articulate their reasoning.
38.79. In order to allow the court to assess the strength of their evidence, fingerprint examiners should highlight the variables relevant to their assessment and how they have formed their conclusions in the light of those variables. The conclusion should state if it has been reached through training and personal experience or on any other basis such as statistical analysis.
38.80. A more specific list of matters germane to the opinion is given in the recommendation in chapter 37, paragraph 123.
Training for court
Fingerprint examiners
38.81. All fingerprint examiners at SPSA should receive court skills training at suitable intervals. The training should emphasise the role of the expert witness.
38.82. Examiners should be discouraged from using stock phrases or responses to questions.
COPFS
38.83. COPFS should ensure that appropriate written guidance as to fingerprint evidence is available to its staff. COPFS should also ensure that a sufficient number of lawyers fully conversant with fingerprint evidence are available to deal with any issues that may arise.
Unable to exclude
38.84. Before a finding of 'unable to exclude' is led in evidence, careful consideration will require to be given to (a) the types of mark for which such a finding is meaningful and (b) the proper interpretation of the finding. An examiner led in evidence to support such a finding will require to give a careful explanation of its limitations.
1. Committee on Identifying the Needs of the Forensic Sciences Community, Committee on Science, Technology and Law Policy and Global Affairs, Committee on Applied and Theoretical Statistics Division on Engineering and Physical Sciences, National Research Council. Strengthening Forensic Science in the United States: A Path Forward, Washington, D.C.: National Academies Press, 2009, page 21
2. NAS, Strengthening Forensic Science in the United States: A Path Forward, 2009, page 139
3. See chapter 35
4. 2011 SCL 645 and [2011] HCJAC 40 URL:
http://www.scotcourts.gov.uk/opinions/2011HCJAC40.html
5. [2011] HCJAC 86, para [48] URL:
http://www.scotcourts.gov.uk/opinions/2011HCJAC86.html
6. Mr McGinnies 4 November pages 46-47 and FI_0152 paras 86-94 Inquiry Witness Statement of Mrs Tierney
7. See chapter 12 para 37
8. Mr McGinnies 4 November page 46ff
9. NAS, Strengthening Forensic Science in the United States: A Path Forward, 2009, page 44
10. See chapter 33 para 43ff
11. Mr Nelson 13 November pages 8-9
12. See Mr Chamberlain 18 November page 44 and Professor Champod 25 November page 81
13. Professor Champod 25 November page 83, Stoney, D.A. What made us ever think we could individualize using statistics? Journal of the Forensic Science Society, 1991, 31(2): 197-199
14. 1934 JC 1 at page 4
15. US Department of Justice, Office of the Inspector General (2011) A Review of the FBI's Progress in Responding to the Recommendations in the Office of the Inspector General Report on the Fingerprint Misidentification in the Brandon Mayfield Case, URL: http://www.justice.gov/oig/special/s1105.pdf, pdf page 10
16. See chapter 37 para 24
17. See, for example, the New Zealand Court of Appeal in R v Buisson [1990] 2 NZLR 542 at pages 546-549
18. R v Reed and Reed, R v Garmson [2009] EWCA Crim 2698, [2010] Cr App R 23 at para 121
19. See chapter 30 para 29ff
20. Professor Champod 25 November page 77
21. Mr Pugh 24 November pages 13, 20-21
22. Ashbaugh D. Quantitative-Qualitative Friction Ridge Analysis: An Introduction to Basic and Advanced Ridgeology. Boca Raton, Florida: CRC Press, 1999, pages 146-148
23. See chapter 33 para 43ff
24. Mr McGinnies 4 November pages 77-78
25. Mr McGinnies 4 November pages 78-79
26. See chapter 33 para 44ff
27. See chapter 35 paras 99-100; and chapter 33 para 44ff
28. Para 34ff
29. Ashbaugh D., Quantitative-Qualitative Friction Ridge Analysis, 1999, page 4
30. See chapter 35 para 132
31. See, more generally, chapter 39
32. CO_4524
33. Judge Edwards, "Solving the Problems That Plague the Forensic Science Community", keynote address at the conference on Forensic Science for the 21st Century: The National Academy of Sciences Report and Beyond, Arizona State University, 3 April 2009 URL: http://lst.law.asu.edu/FS09/pdfs/H.T.%20Edwards,%20Solving%20the%20Problems%20That%20Plague%20Forensic%20Science.pdf page 9
34. Champod C. Fingerprint examination: towards more transparency. Law, Probability & Risk, 2008, 7:111-118 at page 113
35. NAS Strengthening Forensic Science in the United States: A Path Forward, 2009, page 217; and see chapter 35 para 127ff
36. See chapter 35 para 129
37. Ashbaugh D. Quantitative-Qualitative Friction Ridge Analysis, 1999, page 4
38. Mr Chamberlain 18 November page 87
39. SG_0375 para 5.20.4 and recommendation 13
40. Cole S.A. The Journal of Criminal Law & Criminology, 2005, 95: 985
41. MP_0007 pdf page 84, MP_0006 and Mr Pugh 24 November page 51ff correcting FI_0082 para 49 Inquiry Witness Statement of Mr Sheppard and Mr Sheppard 7 July page 182ff
42. See chapter 28 para 68ff
43. See chapter 41 para 35ff
44. Mr Pugh 24 November page 9
45. Professor Champod 25 November page 63
46. Professor Champod 25 November page 77
47. IAI Resolution 2010-18 URL:
http://www.swgfast.org/Resources/100716_IAI_Resolution_2010-18.pdf
48. OIG (2011) pdf pages 37-40
49. See chapter 32 paras 8, 9, 15 and 17
50. Professor Champod 25 November page 77
51. Mr Logan 16 November page 37
52. Mr Chamberlain 18 November pages 46-49
53. Mr Zeelenberg 8 October pages 36-37
54. Mr Nelson 13 November pages 20-21
55. Mr Chamberlain 18 November pages 46-49
56. Mrs Tierney 12 November pages 48-51
57. Mr Pattison 13 November pages 145-147
58. Mr Chamberlain 18 November pages 46-48
59. Mr Pugh 24 November pages 9-10, 114-115
60. Mr Pugh 24 November pages 9-10
61. Mr Pugh 24 November pages 13-14
62. Ashbaugh D. Quantitative-Qualitative Friction Ridge Analysis, 1999, page 147
63. See chapter 33 para 49
64. See para 63
65. SG_0375 para 5.20.4 and recommendation 13