31.1. Witnesses to the Inquiry spoke of the way in which they regarded fingerprint evidence prior to 'the McKie case' referring to it as 'infallible' and save for those rare occasions on which it was challenged as 'routine evidence'. It is necessary to see these comments in an historical context.
'Infallibility'
31.2. An accused may be convicted on fingerprint evidence alone in Scotland1 and in England and Wales.2 The mark must be found in incriminating circumstances. For example, an accused's mark on a false number plate on a getaway car does not justify conviction in the absence of other incriminating evidence, as the mark could have been made at a time other than during preparation for the crime.3
31.3. The foundation authority relating to fingerprint evidence in Scotland is the case, in 1933, of Hamilton which concerned a theft committed by breaking in to a shop. The accused's mark was found on a bottle in the shop. The bottle, which was wrapped before the crime, was found after it, unwrapped and with a fingermark on it. The opinion of Lord Justice General Clyde contains a summary of the evidence of the fingerprint officers and it included a claim to infallibility:
"Evidence was given by two Scotland Yard experts, who deponed that, in the first sixteen of the recognised points of comparison, there was complete identity between the finger-mark on the bottle and the print of one of the appellant's fingers. The experts further deponed that identity to the extent of the first sixteen points was, in their opinion, more than sufficient to warrant the inference of identity between the finger which made the mark on the bottle and the finger from which the print was taken. Lastly, they deponed that this method of identification has been widely followed not only in this country but elsewhere for many years; and that-so far, at any rate-no case of identity in the ridges of the skin of the fingers of two different persons has been discovered. They went as far as to claim for the finger-print method of identification the quality of infallibility. This may well be thought (in words at least) to put the case too high, but the substance of their evidence was that, over an area of experience of great extent, and over a tract of time of highly significant duration, the finger-print method of identification has never once proved to be unreliable. There was no counter-evidence."4
31.4. Though the Lord Justice General proceeded to disapprove of the claim to infallibility he used the scarcely weaker alternative of 'practical infallibility':
"I deprecate the use of the word 'infallibility' in this connection at all. What the experts obviously mean is, not absolute, but practical infallibility-that is to say, a presumption of truth, the reliability of which may be accepted, not because it is irrebuttable in its own nature, but because long and extensive experience is shown to provide no instance in which it has ever been successfully rebutted. All proof depends at bottom on presumption; even the evidence of two credible and uncontradicted witnesses who speak to the same occurrence is probatio probata5 not because it is impossible that they should both be mistaken, but because of the high presumption that what two credible witnesses say happened in their presence did actually happen. Accordingly, the strength of the link provided by the finger-print depends on the degree of reliability which-on the evidence presented to them-the jury thought should be attributed to the finger-print method as applied by the police and the experts in the present case. They evidently thought the presumptions in favour of its reliability so high as to warrant the circumstantial inference that the appellant was the criminal, or one of the criminals, who broke into the shop."6
31.5. Lord Sands was perhaps more cautious:
"...as regards finger-marks, if they are to be relied upon and treated as conclusive, it must be clearly proved that they are the finger-marks of the accused person. That is done by calling in persons expert in regard to finger-marks.... This is a somewhat novel mode of criminal investigation. It will not do for the Crown simply to lead evidence of two experts who say these were the man's finger-marks. As I say, the day may come when that may be enough, and then it will be for the accused to endeavour to challenge or shake that evidence. But, as matters stand, the Crown must proceed to justify the opinion of the experts by eliciting from them upon what their opinion is based. That was done in the present case in the manner which your Lordship in the chair has indicated, and the jury were satisfied with the explanation. Now, it might very well be that such evidence might not satisfy every man, that one might find persons who would say, 'It may be that that evidence is true as regards past experience, but we will not take it that there cannot be two people with identical finger-marks, or that this is such an improbability as amounts practically to an impossibility.' A or B might say that; but it is a question for the jury. It is for them to judge of the evidence, and, if on the evidence they are satisfied with the explanations of the experts and the reasons they give, I do not think we can interfere. In the present case the jury were satisfied. I do not for a moment say that a jury are bound to accept finger-print evidence, and expert evidence connected with it, in every case as conclusive. It is a question for the jury in the particular case..."7
31.6. The next authority of note is HMA v Rolley8 (in 1944), which was also a prosecution for theft. After the crime a palm print was found on the sideboard of the house that had been broken in to. Mr Rolley gave evidence denying his guilt. The only evidence against him was that of four fingerprint officers who identified the palm print as having been made by him by reference to two sets of charted comparisons, one small and showing 16 points of identity and the other enlarged and showing 25 points in identity. The law report reproduces the directions that the trial judge (Lord Justice Clerk Cooper) gave to the jury. Having read to the jury Lord Justice General Clyde's reference to 'practical infallibility' in Hamilton he gave the jury this explanation of it:
"¿ in this imperfect world absolute certainty is unobtainable by any method known to the law or to science, and that what we are looking for is practical infallibility. The real question for you will be whether you are satisfied, from the explanations given by these four gentlemen, that their experience has been gathered over a sufficiently long number of years-I think twenty-six years in one case-and has extended to a sufficiently large number of cases to justify the confidence with which their conclusion was expressed."9
In other words, fingerprint evidence was as certain as that provided by any method known to the law or to science. It may come as no surprise that, despite Mr Rolley testifying as to his own innocence, the law report ends: "THE JURY returned a verdict of guilty."
31.7. The first edition of the Scottish textbook on evidence 'Walkers on Evidence'10 was written in 1964 and notes that fingerprint evidence was not regarded as infallible "but as providing evidence of a degree of probability which is ample for judicial proof".11 The text goes on to note that in Britain experts would not be prepared to commit themselves to the view that two marks were made by the same person unless there were at least 16 identical characteristics, adding that there is no reason why evidence of a lower number of characteristics in common should not be admissible, though perhaps not sufficient in itself to justify a conviction.
31.8. That concluding comment was repeated as part of the submission by the Solicitor General (Rodger) in the first Scottish case concerning DNA evidence, Welsh v HMA,12 in 1991, where parallels were drawn with fingerprint evidence. Lord Justice Clerk Ross records the Solicitor General's submission:
"The Solicitor-General explained that this was the first occasion upon which the appeal court had had to consider evidence of DNA profiling. He maintained, however, that that did not create any difficulty in the present case. The trial judge had correctly reminded the jury that they did not require to be satisfied to the extent of mathematical certainty. Although in practice in fingerprint cases one looked for sixteen points of similarity, there was no reason why evidence should not be given that a smaller number of similarities would suffice."13
31.9. The opinion of the Lord Justice Clerk in Welsh is also worthy of mention for the following excerpt: "DNA profiling, like fingerprint evidence, depends to some extent on theory and statistics, and it is for the jury to assess the evidence and determine whether they accept it as sufficient to warrant conviction." The extent to which fingerprint evidence is supported by theory and statistics is discussed in chapter 41.15
31.10. Writing in 1999, the Canadian authority on fingerprint evidence, David Ashbaugh, referred to the fact that few "identification specialists" were challenged in court and he observed: "Legal counsel shied away from dwelling on a science that was considered exact and infallible, a belief that was difficult to dispel without adequate and structured literature being available. Most challenges were haphazard at best, usually ill-prepared, and often confusing. The majority were doomed to fail. Each failure further entrenched the infallibility of the science."16
R v McNamee and R v Buckley
31.11. By 21 April 1999 (when the trial in HMA v McKie commenced) there had been discussion in the courts of England and Wales about the admissibility of evidence where a fingerprint examiner was satisfied about identity but there were less than 16 matching points. Cases referred to in R v Buckley17 show that by 1995 the prosecution in England and Wales was seeking to lead fingerprint evidence even though it did not meet the 16-point standard. Between 1995 and 1998 evidence was admitted based on twelve similar ridge characteristics, and in one case about two fingerprints which had eight and fourteen characteristics respectively.
31.12. On 30 April 1999, the Court of Appeal in R v Buckley refused an appeal in which there was a challenge to the admissibility of fingerprint evidence based on nine matching ridge characteristics. Rose LJ noted the potential introduction of the non-numeric approach and provided guidance for the assistance of judges who had to consider the admissibility of fingerprint evidence based on less than sixteen points. That guidance uses eight points as the lower limit of admissibility "save in wholly exceptional circumstances", with discretion to admit evidence based on more than eight points depending on a number of criteria. The evidence of Mr Sheppard18 supports the view that Buckley did not have a significant impact on practice in England and Wales, probably because of the move to a non-numeric approach, but it was cited by the Court of Appeal in 2011 in R v Smith19 at least as part of the historical narrative. The decision in Buckley did not impact on HMA v McKie because it was probably unknown at the time in Scotland and, in any event, the Crown relied on the combination of 17 points in the SCRO productions for Y7.
31.13. Also, by the date on which the trial in HMA v McKie commenced, there had been one high profile case in England involving a conflict in fingerprint evidence. The case was R v McNamee,20 decided by the Court of Appeal on 17 December 1998.
31.14. Mr McNamee had been convicted of conspiracy to cause explosions in connection with the Hyde Park bombing in 1982. The evidence against him included three fingerprints found on tape and a battery in various caches of bomb making equipment found after the bombing. The Court of Appeal records that two defence experts examined the prints before trial but they were not called to give evidence, the inference being that they did not challenge the Crown's identification. At trial there was no challenge to the identification of any of the three prints and Mr McNamee sought to explain the presence of his prints by evidence that he might innocently have come in contact with the items during the course of his work. He was convicted. At his first appeal the court allowed fresh evidence to be led relating to the identification of the mark on the battery. At that stage the defence instructed two experts, Mr Waghorn and Mr Swann. Initially Mr Waghorn spoke to only ten matching characteristics but in cross-examination, he agreed with the 16 matching points that the Crown witness had shown on enlargements prepared for the appeal. Mr Swann was not called to give evidence and the appeal was refused. In due course the case was referred back to the Court of Appeal by the Criminal Cases Review Commission after the prints of another man, a convicted bomber, were found on the bomb making materials. The second appeal was argued on a number of grounds including fresh evidence relating to the identification of one of the marks, the mark on the battery. The Appeal Court heard conflicting evidence from 14 fingerprint experts about the analysis of the mark, including some who were witnesses to this Inquiry.21 The question for the Appeal Court was whether the verdict was safe and that involved considering what the jury would have made of the fresh evidence. The court found it impossible to say with confidence what conclusion the jury would have reached on the fingerprint evidence. In addition to the doubt about the fingerprint evidence the court held that there had been a material non-disclosure by the prosecution on another point and these two matters combined justified the conclusion that the verdict was unsafe and the second appeal was allowed and the conviction set aside.
31.15. The basis of the decision in R v McNamee was that the court was uncertain what conclusion the jury would have reached on conflicting testimony. The court did not make a positive finding of misidentification. That decision, which has never been reported in any series of law reports, understandably appears to have had no bearing on the McKie case.
Routine evidence: sections 280 and 281 of the Criminal Procedure (Scotland) Act 1995
31.16. As discussed in chapter 30, the form of the fingerprint reports prepared by the SCRO officers for the trials in HMA v Asbury and HMA v McKie and the revised style now used in current practice22 were influenced by the statutory provisions in sections 280 and 281 of the 1995 Act.
31.17. The history of these provisions is of some relevance, as it correlates with evidence to the Inquiry that fingerprint officers had to appear less often in court after they were introduced. They date back to section 26 of the Criminal Justice (Scotland) Act 1980, though what is now section 280(4) of the 1995 Act initially only applied to summary trials. It was extended to solemn proceedings in 1995 by section 22(3) of the Criminal Justice (Scotland) Act 1995 and the provisions came to be consolidated in the Criminal Procedure (Scotland) Act 1995.
31.18. The rationale for the introduction of section 26, noted in chapter 30,23 is also instructive: "This section contains a number of provisions designed to make it easier for the prosecutor to establish certain matters which are rarely in dispute¿"24
Commentary
31.19. Prior to HMA v McKie, fingerprint evidence was rarely, if ever, challenged in Scotland and had come to be treated as 'routine'. Statute had, as a consequence, exempted fingerprint evidence to some extent from the full rigours of the legal requirements otherwise applicable to expert evidence. However, the Glasgow office of SCRO at that time followed an exceptional practice of producing charted enlargements to illustrate 16 points in coincident sequence for Y7, QI2 Ross and certain other marks involved in HMA v Asbury and HMA v McKie. This provided some transparency and, in the event, was sufficient to enable Mr Wertheim and Mr Grieve to take issue with the conclusion in relation to Y7 and, as in HMA v Rolley, the charted enlargements were available to the jury for them to assess the evidence that the witnesses gave.
31.20. There was a second compounding factor at play in the build up to HMA v Asbury and HMA v McKie. Fingerprint evidence had come to carry the aura of 'infallibility'. It may be difficult to perceive any substantial distinction between 'infallibility' and 'practical infallibility' and the more relevant consideration may be that since the case of Hamilton in 1933 the Appeal Court in Scotland had endorsed fingerprint evidence as carrying a presumption of truth. In that respect Scotland was not alone.
31.21. With the benefit of the subsequent insight provided by Simon A. Cole in an article entitled 'Is Fingerprint Identification Valid? Rhetorics of Reliability in Fingerprint Proponents' Discourse'25 and other modern sources it is possible to recognise now that there were practical limitations to the two propositions spoken to by the fingerprint experts in Hamilton: (1) no case of identity in the ridges of the skin of the fingers of two different persons has been discovered; and (2) over a period of time of significant duration, the fingerprint method of identification has never once proved to be unreliable. Even if those two propositions were correct in 1933, given the complex judgments that fingerprint examiners have to make during a comparison, it does not necessarily follow that the fingerprint examiners who are giving evidence in any particular case have been able, reliably, to observe and interpret a sufficient number of matching characteristics in mark and print to prove that a particular individual is the donor of the particular crime scene mark.26 In any event, the second of those propositions has been overtaken by events: instances of erroneous identification have occurred.
31.22. Among the authorities to discuss this matter more recently was the English Law Commission. The Commission does not directly address the reliability of fingerprint evidence but such evidence is of a kind that involves the combination of a scientific base and experience-based application.27 The Law Commission suggests that there will be experience-based lines of expert evidence where the wealth of experience possessed by the expert is likely to provide a sufficient guarantee of reliability in the round for much of the expert evidence founded upon it28 but it may be added that this ought to proceed on a proper understanding of the underlying basis of the expert's opinion.
31.23. In the case of fingerprint evidence the scientific underpinning is limited to the first of Cole's two propositions: i.e. that no two individuals have been known to share the same fingerprint, hence fingerprints can provide a reliable basis for identifying an individual.29 The reliability of evidence of identification given by fingerprint examiners in any particular case calls for the assessment of a complex mix of distinct experience-based or subjective judgments, including (1) whether the patterns of ridge detail in mark and print are in 'agreement' and (2) whether the quality and quantity of the matching characteristics are 'sufficient' to point uniquely to the donor of the print as the maker of the mark. The reliability of fingerprint evidence depends on the robustness, and ultimately on the accuracy, of those judgments.
31.24. Four of the generic criteria that the Law Commission recommends for the assessment of the reliability of expert evidence generally seem particularly apt to the assessment of fingerprint evidence:
"(a) The extent and quality of the data on which the opinion is based, and the validity of the methods by which they were obtained.
(b) If the opinion relies on an inference from any findings, whether the opinion properly explains how safe or unsafe the inference is (whether by reference to statistical significance or in other appropriate terms).
(c) If the opinion relies on the results of the use of any method (for instance, a test, measurement or survey), whether the opinion takes proper account of matters, such as the degree of precision or margin of uncertainty, affecting the accuracy or reliability of those results.
(d) The extent to which any material upon which the opinion is based has been reviewed by others with relevant expertise (for instance, in peer-reviewed publications), and the views of others on that material."30
Fingerprint evidence: unable to exclude
31.25. Traditionally in Scotland fingerprint officers have given evidence only if they have been able to reach one of two positive findings: either (1) that a specified person has been identified as the maker of the mark; or (2) that a specified person has (or specified persons have) been excluded as its maker. The Inquiry has heard evidence that some fingerprint officers in England have given evidence on an intermediate finding, where there is no material difference between mark and print so the donor of a known print cannot be excluded as the maker of the mark; but insufficient matching ridge characteristics to satisfy the officer that the mark can be uniquely identified as that of any particular person. That intermediate finding could be expressed in one of two ways: (a) the ridge characteristics in the mark are consistent with those in the print and therefore the donor of the known print is one of a number of persons who could have made it; or (b) that donor cannot be excluded as the maker of the mark. The second of those formulations has been termed 'unable to exclude' evidence and evidence of that kind has been led in England.31
31.26. I am advised that there is no rule of law that prohibits such evidence being led in a Scottish case. It could, in theory, provide corroboration for the prosecution; and such evidence might be of assistance to the defence, for example, if it supported the incrimination of another person. That is not to say that such evidence would invariably be admissible. The problem is that there is no statistical data showing the incidence of combinations of ridge characteristics in the population and therefore no scale by which to assess the weight to be placed on it.32
31.27. 'Unable to exclude' conclusions are considered further in chapter 38.
1 Hamilton v HMA 1934 JC 1; and HMA V Rolley 1945 JC 155 (a case concerning a palm print). Hamilton v HMA was applied in a murder case as recently as 1989 in Langan v HMA 1989 JC 132.
2 R v Castleton (1909) 3 Cr App Rep 74
3 Reilly v HMA 1986 SCCR 417; see also Campbell v HMA 2008 SCCR 847
4 Hamilton v HMA
5 Literally a "proved proof", a fact that is not permitted to be impugned or challenged.
6 Hamilton v HMA page 4
7 Hamilton v HMA pages 5-6
8 HMA v Rolley
9 HMA v Rolley page 158
10 Walker A.G and Walker N.M.L. The Law of Evidence in Scotland.1st Edition, Edinburgh: W. Green, 1964.
11 Walker A.G and Walker N.M.L. The Law of Evidence in Scotland. 1st Edition, 1964. Para 12, pages 12-13.
12 1992 SLT 193
13 Welsh v HMA page 196L
14 Welsh v HMA page 197D
15 At para 8ff
16 Ashbaugh D. R. Quantitative-Qualitative Friction Ridge Analysis: An Introduction to Basic and Advanced Ridgeology. Boca Raton, Florida: CRC Press, 1999. Page 4.
17 [1999] EWCA Crim 1191, 163 JP 561 CA
18 Mr Sheppard 7 July pages 180-182
19 [2011] EWCA Crim 1296 and [2011] 2 Cr App R 16
20 [1998] EWCA Crim 3524
21 Mr Swann and Mr Leadbetter gave evidence for the appellant and Mr Mackenzie and Mr Dunbar gave evidence for the prosecution.
22 CO_4109
23 Para 50
24 Annotation to section 26 of the Criminal Justice (Scotland) Act 1980 in 1980 Current Law Statutes Annotated, volume 2 (Sweet & Maxwell)
25 Cole S.A. Is Fingerprint Identification Valid? Rhetorics of Reliability in Fingerprint Proponents' Discourse. Law & Policy, 2006; 28(1): 109-135
26 See chapter 2 para 30ff
27 The Law Commission. Expert Evidence in Criminal Proceedings in England and Wales, The Stationery Office, 2011, LC325, paras 3.46-47. URL:
http://www.official-documents.gov.uk/document/hc1011/hc08/0829/0829.pdf
28 The Law Commission. Expert Evidence in Criminal Proceedings in England and Wales, 2011, LC325, para 5.79.
29 See chapter 2 para 30ff
30 The Law Commission, Expert Evidence in Criminal Proceedings in England and Wales, 2011, LC325, Schedule Part 1, para 1, to draft Bill at pdf page 166
31 e.g. Mr Chamberlain 18 November 2009 pages 105-106
32 e.g. Mr Pugh 24 November 2009 pages 12-14