32.1. At the time of the Asbury and McKie prosecutions the Scottish criminal justice system was working to the '16-point standard' for identifications to be taken to court. This chapter describes this standard, including its history and its use in Scotland.
32.2. An account of the genesis of the 16-point standard was given in the report by Evett and Williams, 'A Review of the Sixteen Points Fingerprint Standard in England and Wales.'1 Though this was written from the vantage point of England and Wales, it would appear from the overlap with the corresponding narrative in the HMICS report2 that the historical outline is the same in Scotland.
A history of the 16-point standard
32.3. The first fingerprint bureau in the UK was established at New Scotland Yard in 1901.
32.4. Initially a 12-point standard was used in Great Britain until 1924 when the 16-point standard came to be recommended by New Scotland Yard following review of an illustration in a publication by Alphonse Bertillon, in France in 1912, which had been sent to New Scotland Yard by the New Zealand police. That illustration showed prints from two different individuals, masked in such a way as to obscure differences, but with 16 points of similarity marked on the revealed areas of the prints. New Scotland Yard doubted six of the points of similarity but accepted that ten were present. Given that the illustration was understood to show that two different individuals could share ten common points, it was decided that the 12-point standard was not high enough and it was raised to 16.
32.5. Until 1953, despite the recommendation by New Scotland Yard, practice varied and some fingerprint experts were prepared to give identifications on less than 16 points. In 1953 the Home Secretary arranged a meeting following a trial in which an expert gave evidence identifying two prints, one having 12 points of similarity and the other 15. The defendant was convicted but the Home Secretary was concerned that if evidence was given of identification on less than 16 points the defence might successfully challenge the evidence, thereby tending to discredit fingerprint evidence as a whole. The Inquiry was provided with a contemporaneous note of that meeting3 and the relevant parts of the conclusion were as follows:
"The meeting was not concerned with the number of points of resemblance which are accepted as adequate for the purposes of ordinary police investigation.
The following conclusions were reached:-
32.6. The exclusion of "ordinary police investigation" work confirms that the standard was aimed at the evidence that might be led in court or, as Evett and Williams put it, "court quality identifications". There were originally two national standards relevant to court: the 16-point standard for a single print and the "two ten" rule in the case of multiple prints from different fingers belonging to the same individual.
32.7. From 1984, at least in England and Wales, the court standard became subject to what became known as the 'dire and crucial' exception. The National Conference of Fingerprint Experts4 agreed that evidence might be given of an identification on less than 16 points in a case of particular importance, provided that the evidence came from an expert of long experience and high standing and it might have been necessary for the expert to say that, though he was personally satisfied as to the identification, it did not meet the national standard. This 'dire and crucial' exception appeared to apply only to serious crimes.
32.8. The exclusion of contributions towards ordinary police investigations led to a practice that some witnesses to the Inquiry referred to as the 'strong suspicion' exception. Fingerprint examiners would report to the police 'non-provable' identifications based on less than 16 points in coincident sequence, the object being to give the police a lead in their investigation while recognising that, save in cases falling within the 'dire and crucial' exception, the fingerprint examiner could not take that evidence to court. Evett and Williams reported that the lower threshold for 'strong suspicion' was generally eight points but that the Metropolitan Police used a threshold of ten. The justification for the use of eight points derived from the conclusion of the 1978 National Conference of Fingerprint Experts:
"There was general agreement amongst fingerprint officers that if there were eight points of resemblance present, that was sufficient to establish identity beyond all reasonable doubt."5
32.9. To the lay person it might be thought that there would be variations in the degree of confidence that a fingerprint examiner would have had regarding the certainty of his conclusion as between (a) an identification to the legal standard of 16 points and (b) a finding of 'strong suspicion' based on a lesser number of points as low as eight. That was not the case. The distinction between the two was a matter of legal practice, not a reflection of fingerprint practice. Evett and Williams emphasised that it was "important to understand that a fingerprint expert regards any identification as a certainty" and that was whether he had reached that conclusion on 16 points or less.6 That is consistent with the fact that the 1978 National Conference of Fingerprint Experts' conclusion was that eight points were sufficient to establish identity beyond all reasonable doubt.
Lack of substantive base for the 16-point standard
32.10. Evett and Williams reported in 1989. As part of their study they located a copy of the original Bertillon publication.7 By scrutinising better quality copies of the prints than had been available to New Scotland Yard they found that the prints had been touched up. Professor Margot of Lausanne University confirmed that, of the ten points originally accepted by New Scotland Yard, five had in fact been altered by the addition of ink lines. This finding coincided with an article published in 1914 by Locard who said that "certain of the points are patently fabricated."
32.11. Consequently, Evett and Williams concluded: "The original decision to adopt a standard of 16 points at New Scotland Yard was clearly based on incomplete and misleading information."8
England and Wales: evidence other than in accordance with 16-point standard
32.12. As noted in paragraph 7, the 16-point standard was not universally insisted upon in English legal practice and Mr Leadbetter confirmed that there were instances in which evidence was led in English courts where 16 points had not been found. He gave an example of one case in which he had been involved where he gave evidence of identity based on only eight points.9
32.13. In that regard reference may also be made to the discussion of the Court of Appeal decision in R v Buckley in April 1999 upholding a conviction where the fingerprint evidence was based on nine matching ridge characteristics.10
The 16-point standard in Scottish practice
32.14. In 1997 the 16-point standard was the basis of legal practice in Scotland.
32.15. Mr MacPherson's evidence suggests that practice at SCRO mirrored that in England and Wales with the application of the two ten (or '10 and 10') rule, the 'dire and crucial' exception and 'strong suspicion' letters where there were between 10 and 15 points.11 The threshold of ten for 'strong suspicion' (as opposed to eight) is discussed later in the context of 'eliminations'.
32.16. That may, however, be an oversimplification.
32.18. In the 1998 letter, written by Mr Mackenzie, it was stated that there had never been a problem for SCRO adhering to the two ten rule, but Mrs Tierney (who started work in SCRO in 2000) told the Inquiry that she was told that it was not the practice to use the two ten rule routinely. She was told to mark any identification having 10 to 15 characteristics with an asterisk and often had casework returned by a verifying expert who would state they had in fact found 16 characteristics. This prevailed until 2004 when, as the training manager, she put forward a paper to Mr Ewan Innes, then Head of SFS, detailing the actual requirements of the 1953 standard and recommending that the practice of asterisking identifications with less than 16 characteristics cease. This was supported by Mr Innes and the practice was subsequently changed.14
32.19. Mr Pattison made enquiries and found that 'dire and crucial' was not a term used by either COPFS15 or the police.16 Neither Mr Mackenzie nor Mr Dunbar could give any specific example of a case in which it had been applied in Scotland.17 That is consistent with correspondence in 1994 in which SCRO plainly stated that its experts would not give evidence in court where fewer than 16 points were found, not even in a case where the procurator fiscal was showing interest in such evidence.18 Mrs Tierney spoke of one case reported to the procurator fiscal with less than 16 points when she worked in SCRO,19 which would be after May 2000.20
32.20. To complete the survey of practice in Scotland, it is necessary to note that SCRO examiners also spoke of 'eliminating' a mark as having been made by a specific individual, an 'elimination' being a finding that could be made on less than 16 points.
The detail of the 16-point rule
32.21. The Inquiry heard evidence of some variations in practice as to the details of the 16-point standard.
The definition of a characteristic
32.22. The 16-point standard operated relative to second level detail.21
32.23. The HMICS report recorded a lack of consistency worldwide on the description of characteristics that could affect the application of this standard. Thus, some practitioners might characterise the same feature as either (a) a short ridge (either end being a ridge ending and hence counting as two points) or (b) an island (one point).22
32.24. SCRO witnesses explained that examiners were taught that there were essentially two characteristics, bifurcations and ridge endings, with various permutations such as lakes, and islands or independent ridges.23 Mr MacPherson told the Inquiry in his oral evidence that SCRO counted some features such as a lake or island as two characteristics whereas he understood that they would count as one in England and Wales.24 Mr Geddes said that he did not know of a difference between Scotland and England until the Inquiry, though he was aware of the Netherlands regarding two features as one point.25
32.25. Mr Sheppard was asked about any difference in English practice and his reply was that a lake and an island each ordinarily counted as one characteristic but each could be counted as two if necessary to complete 16 points.26
32.26. The Evett and Williams report stated that at one time the standard for identification was "that mark and print should exhibit at least 12 points (characteristics, or minutiae) of agreement, and, of course, no points of disagreement)." (Emphasis added). Evett and Williams then explained that the standard was raised to 16. The inference to be drawn from the Evett and Williams report would appear to be that a 'point of disagreement' would be fatal.27
32.27. Notably, the formulation of the 16-point standard in the notes of the Home Office meeting in 1953 contains no explicit reference to points of disagreement or differences (explained or unexplained) but Mr Sheppard gave this fuller formulation of the rule:
"the common practice and the common application of rules, although not written down rules, is that to provide evidence in a court of law you must provide 16 ridge characteristics in coincident sequence with none in disagreement."28
32.28. At one level this formulation was agreed by the SCRO fingerprint examiners, including Mr MacPherson.29 Mr Dunbar indicated that for an officer to be satisfied in relation to any comparison, "there must be no areas of disagreement that cannot be explained."30 Mr Geddes said: "It is when the ridge characteristics fall in sequence and agreement, with none in disagreement, that I know I have an identification."31
32.29. On closer scrutiny, however, there came to be a subtlety to the evidence of the SCRO witnesses. A more accurate formulation of the 16-point standard is probably that there should be no unexplained differences and as discussed in chapter 28, the Inquiry heard evidence from some fingerprint examiners that the presence of 16 characteristics in coincident sequence proved identity beyond any doubt. Accordingly, they reasoned that any differences must have some explanation, even though the examiner was not aware what the specific explanation was; and by that circular process of reasoning effectively restored the formulation of the rule back to the 1953 statement of it, with no significant role ascribed to differences.
Identifications and eliminations
32.30. At SCRO there was a distinction between 'identifications' and 'eliminations'. An 'identification' related to identifying a mark as belonging to a suspect, and 'elimination' to a person who had a reason to be at the locus of the crime (for example the victim or a police officer involved in the investigation).32
32.31. An 'identification' would generally have to be made to the 16-point standard. An 'elimination' could be made on a lesser number of common points. Mr Mackenzie said that this had been in place when he joined the bureau in 1967.33
32.32. Mr Stewart said that there were two criteria. Normally in a volume crime case, an examiner would be willing to eliminate on a lower number than 16, once he had established identity. In a serious case, it was down to the examiner in charge of the case to decide whether to apply the 16-point standard even to eliminations or to apply a lesser standard to them. He was the one who had the information to decide whether 'elims' might be more significant in the case than suspects, for example.34
Number of points for an 'elimination'
32.33. As at 1997 an examiner making an 'elimination' of a mark had still to satisfy himself or herself as to the identity between mark and print but, as already explained in paragraph 9 a fingerprint examiner would often reach that conclusion to a level of personal certainty on less than 16 points. Indeed, an examiner could arrive at a personal conclusion on identity before counting the points.35
32.34. Evidence varied regarding the personal threshold for an 'elimination'.
32.36. Mr Dunbar said that for an elimination "the number of characteristics in sequence and agreement was probably viewed as somewhere around eight", but for quality assurance purposes he did not regard a strict numeric standard as being the appropriate approach, rather it was for the officer to be satisfied that the mark and print were made by the same individual.39 Ms McBride's evidence was the same.40
32.37. Mr Foley said eight to ten depending on the quality of the mark, if he had to go for one number it was ten.41 Mr Bruce said a minimum of ten (also speaking of a minimum of double figures but he did not know if that was official) unless it was "ultra clear".42
32.38. Mr Padden's recollection was that the threshold was 12.43
32.39. It may be fair to infer that Mr Padden's recollection is probably mistaken given that his colleagues consistently spoke to a lower number. The variations among the other witnesses, ranging from eight to ten, probably reflect the fact that the number of characteristics was an irrelevance44 or at the very least secondary to the nature of the characteristics and the relationship between them.45
3 CO_1599 pdf pages 33-34
4 The meeting in 1953 between representatives of major fingerprint bureaux, the Director of Public Prosecutions and the Home Office also decided on an annual meeting of representatives of all fingerprint bureaux. Evett & Williams noted that The National Conference of Fingerprint Experts no longer existed - CO_1375.
5 CO_1375 pdf page 11
6 CO_1375 pdf page 30
7 CO_1375 contains a reference to the original article: Bertillon, M.A. 'Les Empreintes Digitales' (1912) Archives d'Anthropologie Criminelle de Lacassagne, volume 27, pages 36-52
9 Mr Leadbetter 23 October pages 20-22
10 See chapter 31
11 FI_0055 para 80 Inquiry Witness Statement of Mr MacPherson
14 FI_0152 paras 13-15 Inquiry Witness Statement of Mrs Tierney
15 FI_0195 para 14 Inquiry Witness Statement of Mr Pattison
17 Mr Mackenzie 2 October page 129, Mr Dunbar 6 October pages 61-62
18 CO_1599 pdf pages 62-65 (and see also pages 6-9)
19 Mrs Tierney 12 November pages 162-163
20 FI_0152 para 2 Inquiry Witness Statement of Mrs Tierney
21 See chapter 2 para 11ff
22 SG_0375 para 5.11.2
23 Mr Geddes 26 June pages 6-7 and Mr MacPherson 27 October page 17
24 Mr MacPherson 27 October page 15
25 Mr Geddes 26 June pages 15, 18
26 See chapter 10 para 110
27 e.g. see Champod C. and Chamberlain P. Fingerprints, in: Fraser J. and Williams R. (eds) Handbook of Forensic Science, Willan Publishing, 2009 page 72
28 Mr Sheppard 8 July page 137
29 Mr MacPherson 27 October page 131
30 FI_0053 para 71 Inquiry Witness Statement of Mr Dunbar
31 FI_0031 para 22 Inquiry Witness Statement of Mr Geddes
32 FI_0039 paras 26 and 27 Inquiry Witness Statement of Ms McBride
33 FI_0046 para 53 Inquiry Witness Statement of Mr MacKenzie
34 Mr Stewart 5 November page 28
35 Ms McBride 6 November page 137
36 Mr Graham 9 July pages 65-70
37 Mr Swann 21 October page 41
38 Mr Leadbetter 23 October page 21
39 FI_0053 para 70 Inquiry Witness Statement of Mr Dunbar
40 Ms McBride 6 November page 98
41 FI_0051 para 7 Inquiry Witness Statement of Mr Foley
42 Mr Bruce 10 July pages 62 and 71-72
43 Mr Padden 23 June pages 63-64
44 Mr Geddes 26 June page 24
45 Mr Geddes 26 June pages 9, 23, 28ff, 39ff, 1 July page 18ff