THE FUTURE
OF FORENSIC
BIOINFORMATION
Carole McCartney
Robin Williams & Tim Wilson
MAY 2010
The Nuffield Foundation is a charitable trust with the aim of advancing social well-being. It funds
research and innovation, predominantly in social policy and education. It has funded this project, but
the views expressed are those of the authors and not necessarily those of the Foundation. More
information is available at www.nuffieldfoundation.org
ii
The Authors
Dr Carole McCartney: Senior lecturer in law at the University of Leeds. Carole has written
on Australian justice, Innocence Projects, and DNA and criminal justice, authoring Forensic
Identification and Criminal Justice: Forensic Science, Justice and Risk (2006). She
established an Innocence Project at the University of Leeds in 2005, and was project
manager for the Nuffield Council on Bioethics report ‘The Forensic Uses of Bio-information:
Ethical Issues’ (2007). She is currently an EU Marie Curie Fellow (2009-2012), working on a
project entitled: ‘Forensic Identification Frontiers’, which sees her based at the Centre for
Forensic Excellence at Bond University, Australia until September 2011.
Professor Robin Williams: Emeritus Professor of Sociology, University of Durham and
Professor of Forensic Science Studies at the University of Northumbria. He has been
researching the police uses of forensic science since 2000, beginning with a study of crime
scene examiners in one police area, extending this to Home Office supported work on crime
scene examination in several police forces, and then carrying out two Wellcome Trust
funded studies of forensic DNA databasing in the UK and across the EU. He was a member
of the Nuffield Council on Bioethics working party on ‘The Forensic Uses of Bioinformation:
Ethical Issues’ (2007).
Professor Tim Wilson: Visiting Fellow at PEALS (Policy, Ethics and Life Sciences),
Newcastle University and, after the commencement of this project, Professor of Forensic
Science and Public Policy at the Northumbria University Centre for Forensic Science. As a
former Senior Civil Servant he has extensive experience of policy making on forensic
science, the restructuring of forensic provision and regulation (including as a member of the
NDNAD Strategy Board). While at the Home Office he initiated work that culminated in the
2004 announcement that the DNA Expansion Programme would be replaced with a broader
and more integrated policy. He has also worked on international EU (AEGIS) and G8 cooperation projects for both the Home Office and ACPO.
Acknowledgements: The authors would like to thank all seminar attendees for giving so
generously of their time and expertise throughout this project. We would also like to thank
our respective institutions, Benedict Wilson for drawing figures 4.2, 4.3 and 6.1, and the
Nuffield Foundation for their unstinting support.
iii
The Future of Forensic Bioinformation
Contents
Page
1
2
3
Introduction
o
Forensic Bioinformation: Locating the Issues
o
Method
o
Report Themes
The Current Context
o
The Legal Context
o
Crime and Arrest Rates
o
Fingerprints and IDENT1
o
DNA and the NDNAD
o
Cost
o
The International Context
14
Evidence and Assessment
21
o
Evidence and Assessment
o
Forensic Bioinformation: Existing Data Sources
4
1
Fingerprints
DNA profiles
o
Police Force Returns
o
Forensic Bioinformation: Bespoke Research Studies
o
Cost-effectiveness and Forensic Science Utility
o
Conclusion
Internationalisation and Exchange
39
o
Introduction
o
‘Dissolving boundaries’ and ‘Securing the UK Border’
o
International cooperation
o
Bioinformation held outside the criminal justice system
iv
o
The significance of Scotland
o
The legal position in Scotland
Data sharing between jurisdictions
The interplay of policy & technological developments
Validity, technological development and DNA sample retention
False or adventitious matches, false eliminations and sample retention
Ensuring the validity of dactyloscopic information
The foreseeable technological obsolescence of SGM+
o
5
Conclusion
Governance & Accountability
o
o
The current network of forensic bioinformation governance
Issues and Prospects
Scientific and Organisational Integrity
6
73
Quality Standards
Data Protection and Security
Scientific Research and Technological Development
Public Confidence and Trust
o
Criminal Justice Integration
o
Conclusion
Summary and a Proposal
109
o
Beyond S & Marper
o
Evidence and Assessment
o
Internationalisation & Exchange
o
Governance and Accountability
o
A Proposal
v
1
Introduction
Forensic Bioinformation: Locating the Issues
1.2.
The collection and use of bioinformation in support of criminal investigations and
counter-terrorism measures is an important feature of contemporary efforts to
ensure public safety and maintain national security. The two most important forms
of forensic bioinformation - DNA profiles and fingerprints - were both initiated by
British scientists and police officers, and the UK commitment to the collection and
use of these kinds of forensic bioinformation has been unrivalled internationally.
The National DNA Database (NDNAD) is widely acknowledged to be a worldleading innovation in the development of the forensic application of genetic
technology and information management. The IDENT1 platform, hosting the
national fingerprint and palm print databases, also continues to grow in size and
technical capability. The scale of investment in forensic bioinformation collection
and the technical infrastructures that support its investigatory and prosecutorial
functions continues to be significant, reflecting public confidence and political belief
in its contribution to crime detection, the administration of justice, and the risk
management of known offenders.1
1.3.
Previous studies have examined the ways in which scientific, legislative and
budgetary innovations have been brought together to materialise the rising
aspirations of a number of key criminal justice and forensic science stakeholders.2
There have been extensive ethical and political reflections of these developments
in the UK and elsewhere. Academic research has both supplemented, and been
informed by, the work of relevant Parliamentary Committees, Advisory Bodies and
other civil society organisations.3 More recently, the deliberative landscape over
which various forensic bioinformation claims have been advanced and defended
has been irrevocably altered by the decision of the European Court of Human
Rights on the legality of the current legislative framework for the collection and
management of forensic bioinformation in England & Wales.4
1
By early 2006 there had been investment of £150 million in IDENT1 and £300 million in the DNA
Expansion Programme: Home Office, DNA Expansion Programme 2000-2005: Reporting Achievement
(London: Home Office, 2007) pp 19 and 4.
2
See for example: McCartney, C. (2006) Forensic Identification and Criminal Justice, Cullompton: Willan;
Williams, R., and Johnson, P. (2008) Genetic Policing, Cullompton: Lynch, M., Cole, S., McNally, R. and Jordan,
K. (2008) Truth Machine: The Contentious History of DNA Fingerprinting. Chicago: Chicago University Press;
Cole, S. A. (2001) Suspect Identities: A History of Fingerprinting and Criminal Identification. Cambridge Mass:
Harvard University Press.
3
See, for example (inter alia): House of Commons. (2005) Forensic Science on Trial. Committee on Science and
Technology. London: HMSO; Human Genetics Commission (2001) Whose Hands on Your Genes? London:
Department of Health; Nuffield Council on Bioethics (2007) ‘The Forensic Use of Bioinformation: Ethical Issues’
Cambridge; Human Genetics Commission (2009) ‘Nothing to Hide, Nothing to Fear’ London, Department of
Health; Home Affairs Committee (2010) ‘The National DNA Database’; London, House of Commons.
4
S & Marper v. The United Kingdom (App.no. 30562/04) [2008] ECHR 1581 (4 December 2008).
1
1.4.
Rapidly evolving technology and legal reforms continue to stimulate debate about
the efficacy and ethics of the role of forensic bioinformation within the criminal
justice system in England & Wales and elsewhere. The collection, retention and
use of biological materials, usually without the consent of those from whom they
were taken or retrieved, raises a range of policy questions. These include the
scope of powers necessary for the effective and ethical collection and use of such
materials, and how a balance may be achieved between the exercise of these
powers by the police and the rights of individuals. However, almost all who urge
caution, and many of those who promote, the future development of these
technologies now seem to agree that there is a paucity of independent and
authoritative research on how, and the extent to which, the information derived
from them directly impacts on criminal investigations or usefully supplements other
forms of information held by the police and other relevant agencies.5 There remains
a lack of robust evidence and critical assessment of the benefits and costs of
rapidly increasing expenditure in this area. It is widely argued that this shortcoming
prevents informed decision-making and makes problematic the necessary efforts to
justify particular levels of investment in forensic bioinformation within the legal
system domestically and trans-nationally,6 albeit this is a systematic feature of all
police resource allocation.7
1.5.
The relevance of much domestic discussion on issues surrounding forensic
bioinformation has been overtaken by the judgment of the European Court Of
Human Rights in the case of ‘S’ and Marper versus the UK which was delivered in
December 2008. In 2001, two individuals, ‘S’, an eleven year old, and Mr Marper,
were arrested in separate incidents, and had their DNA samples and fingerprints
taken. Neither were subsequently convicted of any offence and duly applied to their
local Chief Constable to have their DNA samples and profiles destroyed. This
request was denied and an application for judicial review of the denial was refused.
In 2002 the Court of Appeal upheld this decision, and the case proceeded to the
House of Lords in 2004. The Lords considered whether the continued retention of
the DNA of ‘S’ and Mr. Marper, who remained unconvicted of an offence, was a
breach of their Article 8 right to privacy and their Article 14 right against
discrimination. The Lords decided that their privacy may have been breached (one
Lord concluded that it was, the others remained doubtful), but even if it was, it was
a ‘modest’ breach outweighed by the wider benefits to society of retaining DNA and
fingerprints. The purpose of retention – to prevent and detect crime – was provided
for by Article 8, and as such, the retention was ruled lawful.
5
The Magee Report defines information on criminality as: ‘any information which is, or may be, relevant to the
prevention, investigation, prosecution, or penalising of crime.” Sir Ian Magee, ‘The Review of Criminality
Information’ (The Magee Report) (July 2008) Home Office, London. p.4.
6
“A complex network of organisations is involved in the protection of the public… An effective public protection
network demands that all work together, nationally and internationally, to improve public protection. The public
expects them to do so efficiently, cost-effectively and with proper regard to their rights.” ibid., p.3.
7
Public Bill Committee, 26 January 20010, Q 79 at www.publications.parliament.uk accessed 24 March 2010
2
1.6.
The European judges in ‘S’ & Marper v UK took a diametrically opposed view to the
House of Lords. Their unanimous decision held that Article 8 rights to privacy were
indeed breached, and concluded that:
“…the blanket and indiscriminate nature of the powers of retention of the
fingerprints, cellular samples and DNA profiles of persons suspected but not
convicted of offences.., fails to strike a fair balance between the competing
public and private interests and that the respondent State has overstepped
any acceptable margin of appreciation in this regard. Accordingly, the
retention at issue constitutes a disproportionate interference with the
applicants' right to respect for private life and cannot be regarded as
necessary in a democratic society.”8
1.7.
The Marper judgment is significant for its decisive contribution to normative
debates about the use and governance of forensic bioinformation. UK instances of
these debates had been taking place, but had not directly influenced Government
policy. Also, with a few notable exceptions, the questions central to the European
judgment had failed to obtain more than superficial consideration from English
judges. These have been reliant upon a pragmatic refuge in ‘a balance of
judgment’, weighted in the Government’s favour by a failure to examine with any
rigour the claims of Treasury Counsel. It is salutary to recall a similar string of
ECtHR reverses relating to the ‘relaxed, some might say complacent attitude to
proactive policing methods in the English courts and how this came to be resolved
through the Regulation of Investigatory Powers Act 2000 (RIPA).9
1.8.
Mixed UK reactions to Marper may reflect a particular cultural and economic
context that influences attitudes towards the forensic use of bioinformation. For
over one hundred years, English scientists and officials have initiated many key
developments in this field, providing a model for applying modern technologies of
biometric identification in a systematic and commoditised manner. Jeffreys’
discovery in 1984 of what he termed ‘genetic fingerprinting’, resulted in the use of
DNA in immigration and paternity disputes, then in a murder investigation in 1986.
A century on from Henry’s introduction of fingerprints, this new technology of
identification was never a public sector monopoly. Indeed, private DNA analysis
companies in the UK have been instrumental in the expansion of the forensic use
of bioinformation. Even if profit-oriented, the pluralistic nature of this development
served the cause of human rights. For example, since its creation, Cellmark
Diagnostics has helped to reunite families by providing evidence of family biological
relationships to successfully challenge government immigration decisions.10
8
S & Marper v. UK [2008] para.125.
Roberts, P. (2007) ‘Law and Criminal Investigation’, in Newburn, T., Williamson, T. and Wright, A. (eds.)
Handbook of Criminal Investigation (Cullompton, Willan) p.99.
10
This is not necessarily always the case. A private sector monopoly or even technically competitive market that
is wholly dependent for income on government, is unlikely to have an equally benign effect by empowering
9
3
1.9.
These more recent technological developments occurred simultaneously with the
emergence of a competitive forensic marketplace. Within England and Wales the
use of DNA as an investigative technique was not held back by the problems in
traditional public sector forensic science laboratories, which are ‘almost universally
characterised by backlogs’.11 This may influence the content of the criminal law.
Most countries are likely to maintain symmetry between what can be described as
the legislative and technical domains: unless government forensic laboratories
have the resources, appetite for or ability to deliver additional DNA analytical
capacity there is little point in passing legislation to permit the extensive forensic
use of DNA.12 Indeed, the introduction of national arrangements for the use of DNA
for forensic purposes frequently precedes laws regulating such activity. Such use
almost universally precedes specific database legislation. Although within the
European Union there is now a pattern of enacting database legislation before the
database itself is created, there are still at least six countries that may have
databases but no database legislation, not even in the more limited form.
1.10.
This aspect of UK bioinformation development, with the exception of reference to
‘privatisation’ or ‘commercialisation’ per se, is ignored in virtually all of the
sociological, ethical and jurisprudential discourse on the use of bioinformation. It
has two consequences for critics of the UK position pre-Marper. Firstly, it can result
in inaccurate or incomplete comparisons between various countries. England and
Scotland are an important example. Secondly and more significantly, it could mean
that the margin of justification is a much more fragile basis for ethical and political
analysis than it seems at first sight. A comparative survey of the law in different
states may reflect different stages of technological development and investment
rather than jurisprudential philosophy or principled political decision making. The
extent to which the use of DNA and the creation of databases precede legislation
suggests that conclusions drawn from comparisons of statute books have to be
treated with caution. On the street and in the laboratory, criminal justice may
operate differently.
1.11.
For the advocates of a more maximalist approach to DNA profiling and retention,
such as represented in the UK Government’s response to the Marper judgment
(‘Keeping the Right People on The DNA Database: Science and Public Protection’)
the consequences are equally significant. First, in a commercial forensic market,
how can some of the key parties, whose voices are critical and dominant because
of their privileged access to expertise and information, demonstrate that whatever
they advocate is not influenced by commercial interests or professional ambitions?
individuals to defend their human rights against an oppressive state or its agents such as the police, on, the other
hand, a state forensic laboratory monopoly would equally leave citizens disadvantaged. Ultimately, this requires
an effective separation of powers within the executive, legislative, judicial and regulatory structures of the state.
11
Fraser, J. (2006) ‘The application of forensic science to criminal investigation’ in (eds.) Newburn, T.,
Williamson, T. and Wright, A. Handbook of Criminal Investigation (Cullompton: Willan)
12
Wilson, TJ, Forensic science and the internationalisation of policing’ in Fraser, J. G. and Williams, R. (Eds)
Handbook of Forensic Science (Willan Publishing), 2009, pp.509-512 and Figure 18.9.
4
For example, the Government must surely bear in mind the effect that its proposals
for responding to the Marper judgment could have on the income to be realised
from its plan, in due course, to sell some or all of its equity investment in the FSS.
More generally, for those who piloted the introduction of current arrangements was
there a risk of seeking to demonstrate professionally what could be done when,
exceptionally within the forensic sciences, resources were available rather than
delay this for greater deliberation about governance and proportionality? In any
case, would the political climate have allowed such a pause?
1.12.
The Marper judgment - partly as a result of timing with an impending General
Election - increased the political significance of the debate by placing the
Government under a legal obligation to reform the laws governing the retention of
forensic bioinformation. This obligation has required the Government to review and
justify its forensic bioinformation policies as well as explain some of the
assumptions underlying them. The judgment itself draws heavily on arguments and
analysis developed during this emergent discourse, particularly the report
published by the Nuffield Council on Bioethics (NCOB), whose authority it
specifically acknowledges. This vindicates the constructive and principled criticism
of current DNA law and usage to ensure that the inadequacies in these
arrangements are acknowledged. It did not mean, however, that the issues and
recommendations for change elucidated over the years would be reflected in new
legislation. A linear progression from the principles expressed in, for example, the
NCOB report and subsequently endorsed at Strasbourg to the enactment of new
legislation at Westminster was not guaranteed. The reason for this was explained
by The Joint Committee on Human Rights:
‘... judgments of the European Court of Human Rights leave a
considerable amount of discretion to the State concerned as to precisely
how it amends its law, policy or practice to meet these obligations. The
process of implementing a judgment of the European Court of Human
Rights is therefore an unavoidably political process, constrained by the
legal obligations (to stop the breach, provide a remedy for the individual
concerned and to prevent new or similar breaches), but a political process
nonetheless.’13
13
Joint Committee on Human Rights (2010), Enhancing Parliament’s Role in relation to Human Rights
Judgments, Fifteenth Report of Session 2009-10, HL 85 & HC 455 (London, The Stationery Office), para.15.
5
1.13.
The force of this observation is demonstrated in the nature of the Government’s
legislative proposals and how they got through Parliament despite weaknesses in
its claims exposed by well argued opposition and the scepticism of experts. Despite
a long period of consultation and deliberation, the Government’s draft legislation
was published within months of a General Election. They were a disappointment to
those who hoped for a more considered response to the Marper judgment or a
greater willingness to work towards a degree of consensus while the draft
legislation was before Parliament. Consequently, it is likely – irrespective of the
passage of the Crime and Security Act 2010 - that the outcome of these debates
will not be settled until the next Parliament unless the present Labour Government
returns to power with a working majority or, if a minority government can rely on a
coalition of supporters to block further reform. Such an outcome would only endure
in the opinion of many authorities (see below) until the new retention legislation has
been referred back to ECtHR.
1.14.
The Bill contained some concessions for critics. An unequivocal change is that the
deletion of volunteers’ profiles has been brought into line with Scots law: consent
may be withdrawn at any time.14 The effects of some other changes are less clearcut. The Government initially proposed that there should be a legal duty of deletion
where, in the judgment of the relevant chief constable, a DNA profile was obtained
because a person was arrested as a result of, for example, mistaken identity or
unlawful action by the police.15 During the Commons stage of the Bill, because of
disquiet about current inconsistency in decisions concerning the deletion of data,
the Government tabled amendments to centralise decision making, placing
responsibility for ‘guidance’ with the NDNAD Strategy Board.16 However, it is
unclear how these different and potentially contradictory provisions will work in
practice, although another Report stage concession demanded that the Board itself
be subject to limited Parliamentary scrutiny.17 Also, not all questions raised about
how the new arrangements will work have been addressed. At present when a
DNA profile is deleted from the database information relating to the profile is also
removed from the PNC. This second deletion does not appear to be guaranteed by
the arrangements in the new legislation.18
1.15.
An even more startling lacuna in the Act is that, despite provisions to have
cautions, and also reprimands and warnings under the Crime and Disorder Act
1998 treated as the equivalent to a conviction for the purposes of permanently
retaining bioinformation, there appears to be no provision to reverse permanent
14
See the Crime and Security Bill as introduced into the House of Lords, clauses 14 (2) in respect of the
amendment to PACE by the insertion of clause 64ZL
15
Hansard 18 Jan 2010, Column 34: See the Crime and Security Bill as introduced into the House of Lords,
clauses 14 (2) in respect of the amendment to PACE by the insertion of clause 64ZI.
16
Hansard, 8 Mar 2010, Column 65: See the Crime and Security Bill as introduced into the House of Lords,
clause 23 (2)-(3).
17
Hansard, 8 Mar 2010, Column 118: see the Crime and Security Bill as introduced into the House of Lords,
clause 23 (4) and (6) which do not however, give Parliament any powers to vote on NDNAD governance rules.
18
Public Bill Committee, 26 January 20010, Q137 www.publications.parliament.uk, accessed 24 March 2010.
6
retention triggered by a conviction where that conviction has been quashed on
appeal.19 There were a total of 1019 convictions overturned on appeal between
2000 and 2005 in the Court of Appeal (Criminal Division), with 439 successful
appeals against conviction in 2009 alone. In addition, in Crown Courts hearing
appeals from the Magistrates Courts, there were 3,651 appeals allowed in 2005,
meaning that this is not an insignificant oversight.20
1.16.
Limitations in the information provided, and the time for debate available to
parliamentarians meant that some questions could not be followed through. These
included issues raised several times at the Commons Report Stage about the
greater risk of adventitious matches as databases are enlarged or information is
exchanged under the Prüm Treaty, particularly with countries using a different
multiplex to the SGM+ multiplex used in the UK. There were also concerns
expressed about whose DNA would be made available under the Treaty to other
countries. Moreover, claims of ‘national security’ will impose the indefinite retention
of the samples of unconvicted persons on Scotland. This will be wholly at the
discretion of Scottish chief offices of police. This provision undermines the principle
in that jurisdiction where retention, in the absence of a conviction, had hitherto
always been subject to procuratorial or judicial review.21
1.17.
The core political controversy and the key subject of debate at all stages, however,
remains that of the circumstances under which a DNA profile might be retained in
the absence of a conviction. The issue has to be addressed because of the Marper
judgment: how the law governing this might be revised so that such arrangements
are no longer ‘blanket’ and ‘disproportionate’. The Government first proposed to
keep such data for up to 12 years,22 but it finally elected for a six year retention
period in most cases involving the arrest of an adult. The Conservatives propose
similar arrangements to those in Scotland where retention is determined by the
charge for which a person arrested, is normally limited to three years and in all
cases is subject to procuratorial or judicial review.23 The Liberal Democrats
advocate that no profiles should be retained in the absence of a conviction.24
1.18.
During the Commons Report Stage both Opposition options, especially the former,
were commended by the Chairman of the Police Federation, as more likely to
maintain public confidence, in preference to the Government’s proposals.25
19
2010 c. 17 is not any easy piece of legislation to read, but no account appears to be taken in the relevant
sections, that is s. 14 (2) 64ZD (destruction of data relating to a person not convicted) and 64ZI (equivalence to
conviction definitions and the provision to delete e.g. where the arrest was unlawful)
20
See http://www.hmcourts-service.gov.uk/ for Court of Appeal statistics and http://www.officialdocuments.gov.uk/document/cm69/6903/6903.pdf for Crown Court statistics for 2005.
21
The Crime and Security Bill as introduced into the House of Lords, clause 16.
22
Hansard, 8 Mar 2010: Column 39.
23
Hansard, 18 Jan 2010, Column 45-46.
24
ibid. Column 60.
25
Public Bill Committee, 26 January 2010, Q39-49.
7
Ministers argued repeatedly that those proposals alone, even compared with the
Scottish legislation (hinted at as a possible solution by ECtHR), reflect an evidence
based approach. This claim was not validated by the President of ACPO, however,
who instead suggested that six years is ‘reasonable’ or ‘about right’ for reasons of
‘simplicity’ or ‘professional judgment.26 ACPO also made it clear that their views
were influenced by cost and a preference for IT solutions. ACPO witnesses
advocated a simple rule to govern data deletion informed by the assumption that
this would enable the process to be managed by algorithm. The Scottish system,
with its emphasis on an examination of individual circumstances, ‘was rejected
because it would take up an inordinate amount of police resources’.27 They were
not questioned about why, in a country where expenditure on law and order, as a
percentage of GDP, exceeds that in any other G8 member state and two-thirds of
this money is spent on policing, the options were so limited.28 The House of
Commons Home Affairs Committee, having studied the material published in
support of the Government’s claims, could only refer to the ‘complexity of the
issues’ and ‘conflicting evidence’. Nevertheless, there was cross-party agreement
within the Committee that, in the absence of conviction, there is not a case for the
retention of a DNA profile for more than three years and that research was needed
to evaluate the forensic use of DNA.29
1.19.
MPs and The Joint Committee on Human Rights noted a Government Minister’s
statement: “We have obviously considered the [Marper] judgment and how far we
can push the boundary of the judgment in relation to our wish to have protection for
the public”. The Joint Committee found such a ‘very narrow approach’ in order ‘to
maintain the main thrust of its original policy on the retention of DNA’ unacceptable,
and likely to result in a further successful legal challenge.30 The latter view was
shared by the Equality and Human Rights Commission. It wrote to all MPs to
advise them that the Government’s proposals would be in breach of Article 8.31
This is consistent with views expressed by the Secretariat of the Council of
Ministers of the Council of Europe. It has opined that the automatic period of six
years proposed by the Government would not comply with the Marper judgment.32
26
ibid, Q112-117.
ibid, Q93 and 97-100.
28
Soloman, E., Eades, C.,Garside, R. And Rutherford, M. (2007) Ten Years of Criminal Justice Under Labour:
An Independent Audit, London: The centre for Crime and Justice Studies at Kings College London, available at
www.kcl.ac.uk/ccjs pp. 18-23
29
Home Affairs Committee (2010), The National DNA Database, Eighth Report of Session 2009–10, HC 222-1
(London, The Stationery Office), especially paras 35-37.
30
The Joint Committee on Human Rights (2010), Legislative Scrutiny: Crime and Security Bill; Personal Care at
Home Bill; Children, Schools and Families Bill, Twelfth Report of Session 2009-10, HL Paper 67 and HC 402
(London, The Stationery Office), paras 1.8 -1.10 and 1.72.
31
Hansard, 18 Jan 2010, Column 37.
32
Public Bill Committee, 26 January 2010, Q149-150; the Secretariat statement is quoted in The Joint Committee
on Human Rights (2010), Twelfth Report of Session 2009-10, HL 67 & HC 402 (London, The Stationery Office) at
para. 1.47 and note 47.
27
8
1.20.
The Government’s reluctance to materially change its policy was expressed in a
claim made by the Home Secretary. In the mandatory statutory statement about the
compatibility of the provisions of the legislation with the ECHR, while accepting that
the taking of fingerprints and samples is likely to constitute an interference with a
person’s right to a private life under Article 8, he went on to cite in aid of his claim,
that the interference with a person’s physical integrity is short-lived and could be
characterised as ‘minimal’ - the House of Lords Marper judgment.’33 As The Joint
Committee on Human Rights commented:
‘... the reasoning of domestic courts may still be relevant even after a
decision of the European Court of Human Rights, but we would sound a
note of caution about the Government’s continued reliance on the House
of Lords judgment: the Grand Chamber’s decision took account of the
decision and reasoning of the House of Lords and came to a different
conclusion about the ambit of the right to private life in Article 8 ECHR and
what constitutes an interference with that right, and the Government
cannot now rely on the House of Lords decision to the extent that it is
inconsistent with the decision of the European Court of Human Rights.’34
The Home Secretary’s statement was matched by a different form of denial by one
MP, Mr Cash. He did not acknowledge the importance of the domestic discourse
that gave rise to and influenced the ECtHR judgment, and how this had reversed
previous failures domestically to give these important and complex issues serious
and consistent attention. Instead he criticised the employment of ‘abstract
principles that are decided in European Courts’ in matters that ‘could just as well
have been decided’ by English courts on the basis of UK legislation’.35
1.21.
This brief analysis of the recent political process at Westminster does not, of
course, address the wider political considerations that, for the present at least have
proved to be the decisive factor. The temptation to seek party political advantage
from recent events was demonstrated by former Prime Minister Blair. In a carefully
choreographed intervention in the pre-General Election campaign he asserted:
'On law and order the Tories.... even want to restrict the use of the DNA
database. This employs the advanced technology of DNA tracking and
matching, to provide incontrovertible evidence of guilt or innocence. Its
use so far has resulted in extraordinary breakthroughs..... Yet the Tories
oppose it.'36
33
Crime and Security Bill, Explanatory Notes (referring to the Bills introduced in the House of Commons on 19
November 2009), para. 216.
34
op.cit. n.30, para. 1.20
35
Hansard, 8 Mar 2010, Column 63
36
The Guardian, 'Tony Blair election campaign speech – full text’ accessed on 31 March 2010, at
http://www.guardian.co.uk/politics/2010/mar/30/tony-blair-election-campaign-speech-in-full .
9
1.22.
This partisan and, many would say, misleading exploitation of the issue for
electoral advantage was echoed by the Prime Minister. Standing alongside the
mother of a murder victim, he sought to convey the impression that the ability to
detect the killer would have been frustrated by anything less than the Government’s
proposals. The foreword to a Home Office consultation document issued previously
had also implied that the change required as a result of the Marper judgment might
‘undermine a system that helped trap the [killer of Sally Anne Bowman]’. In fact, the
procedure under which Ms Bowman’s killer was detected, following the arrest of the
perpetrator for a minor offence following which he was linked to the crime scene
after his DNA was speculatively searched, was not affected by Marper. This had
been raised with the Government as early as August 2009. The Prime Minister’s
assertions were challenged by both Liberty and the Liberal Democrats who were at
pains to stress that they had not advocated changes in the use of speculative
searching on arrest.37
1.23.
These tactical political manoeuvres need to be seen in the context of a General
Election where the Government was judged by most commentators to be unlikely
to win. A much greater problem that is currently systemic within English politics
remains to be addressed whenever the recent legislation is revisited. As the then
European Human Rights Commissioner, Signor Gil-Robles, observed in 2004 and
anticipating the Government’s ‘push the boundary of the judgment’ philosophy:
‘The United Kingdom has not been immune... to a tendency increasingly
discernable across Europe to consider human rights as excessively restricting the
effective administration of justice and protection of the public interest....’38
1.24.
Some commentators have ascribed this to the reaction in the United Kingdom to
the events of 11 September 2001. The British Government has attempted to curb
civil liberties and reduce the transparency that should be the hallmark of a
democratic government, not always successfully, despite periodically keeping the
unease prompted by terrorism ‘at a certain pitch’.39 Other writers have noted this as
a consequence of growing support for a more communitarian political ideology.
Wilson for example, has compared Labour’s response when it came to power to
the sweeping economic and social changes under the previous Conservative
administration. Because government had ceded much of their power over the
economy, there was a greater disposition to intervene much more in the minutiae of
national life. One of its primary tasks became that of making neighbourhoods safe.
37
For warnings against the misuse of this case see: for example, see Northumbria University Centre for Forensic
Science (2009), Response to the Home Office consultative document ‘Keeping the Right People on The DNA
Database: Science and Public Protection’, Engineering and Information Sciences, and the School of Law of
Northumbria University, p.5. For more recent criticism by Liberty and the Liberal Democrats see: BBC, DNA
database debate is ‘confused’, http://newsbbc.co.uk/z/hi/uk_news/politics/election_2010/8611278.stm .
38
Wilson, B. (2009) What Price Liberty? London, Faber and Faber,p. 319.
39
Grayling, A.C. (2008) Towards the Light, London, Bloomsbury, pp. 262-269.
10
1.25.
The Labour Government was also greatly influenced by the private sector’s
apparent success in using information technology to improve effectiveness, for
example, in anticipating and influencing consumer demand through ‘dataveillance’.
The ultimate goal of many key policy initiatives – in health, child protection, revenue
collection and crime prevention, over the next decade therefore, had been the
convergence of data to make possible speedy and cost-effective intervention,
including pre-emptive action, by a smaller and more distant state. If Wilson is
correct in this analysis and that privacy is ‘a significant barrier to data-sharing,
strategic-pre-emption and resource allocation, something which the government felt
made it less powerful than private business’,40 the Government’s reluctance to
concede greater changes in response to Marper becomes more intelligible.
1.26.
The Government is undoubtedly correct in recognising that the parameters for any
debate about rights have changed. Dataveillance is a fact of everyday life, not least
for academics with a Google dependency and for a younger generation that may
respond to the frequent impersonality of modern life by exposing the private
spheres of their lives to casual strangers in a way unacceptable to older
generations. It would be wrong however, to seek to diminish the validity of a rights
based discourse: it is difficult to identify a more defining characteristic of a law
observing state. Hopefully the period when statements were made about how crime
and terrorism ‘mock’ a system, built not for another decade but another age,41 has
passed. Moreover, there is now a shared awareness in the political parties and
some law enforcement interests, such as the Police Federation, of the need to
rebuild public confidence. Indeed, this is the single national performance target for
the police set by the Home Office. Possibly there is room for convergence between
the Government and its critics on the basis of ideas developed in the field of
bioethics for examining how citizens place trust in others and institutions.42
1.27.
In this report therefore, with the concern to secure public confidence and trust as
our ultimate aim, we have attempted to systematically review the issues of
evidence, and the significance of international exchange, along with some inter-UK
comparisons in order to consider a variety of technological claims and foreseeable
developments. We also seek to illuminate the range of current and emerging
governance issues that need to be addressed. In doing so we hope that this will
prove to be a measured and disinterested, albeit modest, contribution to the debate
about the future of bioinformation that has already been joined at Westminster, but
- as we anticipate at the time of writing - is quite possibly likely to be resumed
during the next Parliament.
40
op.cit. n.38, pp.317-337.
Downing Street Press Briefing, 15 May 2006; quoted in Wilson, B. (2009) op.cit. n.38 p. 351.
42
O’Neill, O. (2002) Autonomy and Trust in Bioethics, Cambridge, CUP.
41
11
Method
1.28.
This project began with a review of the academic literature on forensic
bioinformation, alongside a reading of annual reports of key agencies and
documents produced by bodies that have considered the application of forensic
bioinformation innovations to the criminal justice process in England & Wales.
Following this work, a series of key issues was identified and summarised.
These summaries were then used to support four meetings of law enforcement
professionals, practitioners, government representatives, academics and other
key stakeholders.
1.29.
Each of three seminars began with scene-setting by the investigators with the
remainder of the day dedicated to open (directed) discussion and debate.
Finally, an international symposium provided an opportunity to test and further
refine these ideas with relevant academic, policing and scientific experts from
beyond the United Kingdom, including individuals working in European, North
American and Asian criminal jurisdictions. The seminars were held on 13
January 2009, 25 March 2009, and 13 May 2009. The international symposium
was held on the 28th and 29th of July 2009. Further details on all of these
meetings can be found on the project website at:
www.law.leeds.ac.uk/research/projects/bioinformation.php .
1.30.
Experts working in specific domains of technological innovation have much to
contribute to the wider social debates that surround their work. However, their
views have not often been the subject of direct research.43 This is certainly the
case for those working within forensic bioinformation. In the course of this
report we draw on the many dialogues that were occasioned by these meetings.
We do not identify the contributions of individuals, as all discussions were held
under the Chatham House Rule, but seek instead to represent a variety of
expert views expressed by those with differing scientific, legal, policing and
policy perspectives.
1.31.
The critical examination of such practices that we present in the following
pages, informed by the operational and policy perspectives of a large number of
seminar and symposium participants, is largely focused on England & Wales,
but also draws on international experience where that seems especially
relevant. It pays particular attention to policy issues which are currently
unresolved or contested and suggests how informed debate can be taken
forward. It aims to identify those features of the present system and potential
future developments, both nationally and internationally which either merit
encouragement or give rise to concern.
43
‘The views of experts have not been sufficiently evaluated, however, even though… they play an important
role in shaping future practice.’ Capron et al, ‘Ethical Norms and the International Governance of Genetic
Databases and Biobanks: Findings from an International Study.’ (2009) Kennedy Institute of Ethics Journal,
19 (2) 101-124, p.101.
Report Themes
1.32.
44
Several major themes emerged from reading the written material prior to the
meetings and from the deliberations of participants in the meetings themselves.
They are summarised in the three bullet points that follow.
o
Evidence and assessment. A recent review of the NCOB report noted
that: “A leitmotif in the report is the emphasis on further, more detailed
evidence. Sound ethical advice and good governance depends upon solid
facts and data.”44 Insufficient research and the unsatisfactory way in
which, sometimes, case studies and data have been presented means it
is difficult to see how the bodies set up to utilise and oversee the NDNAD
and IDENT1 can give adequate consideration to possible reforms unless
this deficiency is addressed. There is little statistical evidence or rigorous
comparative analysis of the facts in individual cases to show how and
why forensic bioinformation is useful and to what extent is cost effective in
either inceptive or probative terms. There is scant evidence about the
costs of retrieving bioinformation in terms of individual cases.
o
Internationalisation and Exchange. The international utilisation of
forensic bioinformation and law enforcement co-operation is a growing
consideration for those who think about the future of forensic
bioinformation. Initiatives such as the Prüm Treaty are being implemented
to facilitate the exchange of data for law enforcement purposes. Such
exchange raises important practical questions for investigators and
technical questions for scientists, but also issues of the cost-effectiveness
and affordability of these initiatives. There are also questions to be asked
about the harmonisation of different bioinformation systems and the
possibilities of uniform processes across myriad jurisdictions with differing
policing and judicial systems.
o
Governance and Accountability. There is a need to demonstrate that
reliable forensic bioinformation is being properly and consistently
deployed, both within domestic policing priorities policies and practices,
and also within emerging international systems for criminal investigations,
surveillance (intelligence gathering) and control over the movement of
persons. Consideration needs to be given to the requirements for ‘good’
(effective and ethical) governance of this socio-technical domain. This
includes the future ‘steering’ of forensic bioinformation policies as well as
current management. Issues such as the facilitation of transparency and
accountability mechanisms, including the new demands of emerging
counter-terrorism policies, all need further examination.
Dierickx, K, ‘A Belgian Perspective’ (2008) Biosocieties, 3, p.97.
13
2
The Current Context
The Legal context
2.1.
The police have had powers to take fingerprints of convicts since the 1891
Penal Servitude Act first provided for the measuring, photographing and
fingerprinting of convicted prisoners,45 although a warrant from a Magistrate was
then required to take fingerprints from those on remand and if discharged, the
prints were required to be destroyed.46 It was not until 1948 that the Criminal
Justice Act permitted the fingerprinting of suspects, although these still required
a warrant from a Magistrate and were destroyed if the person was subsequently
acquitted or not committed for trial. The Police and Criminal Evidence Act 1984
(PACE) introduced powers for the police alone to regulate the taking of
fingerprints (terrorism, as well as immigration legislation, have also conferred
powers on authorities to take fingerprints from individuals).47 The power to take
DNA samples was included in PACE, and sit alongside those powers to take
fingerprints (and photographs), the powers being identical. PACE has been
amended several times to facilitate the taking, use and retention of forensic
bioinformation from widening categories of persons.
2.2.
Fingerprints and DNA will ordinarily be taken with consent, but PACE also gives
police the power to take samples without consent where the person has been
arrested for, convicted of, or given a warning, caution or reprimand in respect of
a recordable offence or told that he will be reported for one, or is answering bail
at a police station and there is some doubt over identity,48 or if they are detained
following acquittal on grounds of insanity or finding of unfitness to plead. An
officer must inform the person that their fingerprints/DNA will be subject to a
speculative search (s61(7a)). PACE also preserves the power of compulsory
fingerprinting contained in immigration and terrorism legislation while s27(3)
gives the police the power to request someone to attend the police station to be
fingerprinted, with a power of arrest for the purpose of taking fingerprints for a
recordable offence.49
2.3.
The DNA database was set up, without dedicated primary legislation, in 1995.
45
Regulations for the Measuring and Photography of Criminal Prisoners, SR & O 1896/762. This provision
being continued by the Prison Act 1952 ss16 and 54(3).
46
Leigh, L.H. Police Powers in England and Wales (2nd edn.) (London, Butterworths,1985) p.228.
47
See Terrorism Act 2000 Schedule 8 Part I s10; Anti-Terrorism, Crime and Security Act 2001 s89 (ss201);
Asylum and Immigration Appeals Act 1993 s.3; Immigration and Asylum Act 1999 s141.
48
Authorisation can only be given for the purposes of identification if the person refuses to identify
themselves or the court, or an officer of at least the rank of inspector, authorises them to be taken if there
are reasonable grounds for believing that he is not the same person; or the person who has answered to bail
claims to be a different person from a person whose fingerprints were taken on a previous occasion.
49
A recordable offence is one that can attract a custodial sentence upon conviction, however it also includes
other offences that do not, including begging, selling alcohol to a person who is drunk, public drunkenness,
and running onto a football pitch for example. The full list of recordable offences is found in National Police
Records (Recordable Offences)(Amendment) Regulations 2005 (SI 2005/3106).
14
The power to speculatively search the database was provided for by an
amendment to PACE in the Criminal Justice and Public Order Act of 1994,
permitting DNA profiles and fingerprints to be speculatively searched against
one another and crime scene profiles/prints. The powers to take and retain DNA
(and therefore expand bioinformation databases) were significantly extended in
2001, and again in 2003, by removing the requirement to destroy profiles where
there was no subsequent conviction, and expanding sampling powers so that
the police could take fingerprints and DNA at the point of arrest, rather than
when individuals were charged with offences. Section 64 of PACE then permits
samples to be retained:
“after they have fulfilled the purposes for which they were taken but
shall not be used by any person except for purposes related to the
prevention or detection of crime, the investigation of an offence, the
conduct of a prosecution or the identification of a deceased person or
of the person from whom a body part came.”
2.4.
There remain additional powers to take and retain fingerprints under terrorism
and immigration legislation. Most recently there has been an additional power to
take DNA from people under ‘control orders’ which can be searched against law
enforcement databases. In a change under the Counter-Terrorism Act 2008,
s.14(5)(a) samples may now be used “in the interests of national security” in
addition to those purposes detailed in PACE s.64. The Act also permits covert
sampling, with the seizing of bioinformation without the knowledge of individuals
who may be under surveillance. These samples are not necessarily held on
NDNAD though they are retained, (it is not clear where), raising serious
questions over the (separate?) governance of these samples. The Crime and
Security Act 2010 permits 2 year extensions of retention for anti-terrorism
purposes, which can be repeated. The Counter-Terrorism Act 2008 also
includes provision for the use of DNA for the identification of an individual (not
just bodies or body parts as under PACE 1984). This is a significant departure
from the powers in PACE, and may have implications for human rights.
2.5.
In the latest government proposals (detailed later) – measures were introduced
to keep volunteers off the NDNAD – with reference to a ‘separate process’ for
volunteers. However, no details were forthcoming about the nature of the
separate process or of its regulation and oversight, and the Crime and Security
Act 2010 permits the withdrawal of volunteers’ consent but provides no further
detail on how this is to be managed. As yet – there is no agreement on who is a
‘volunteer’ as presently there are a variety of categories (police register
volunteers; sex offenders; individuals worried about honour-based violence;
missing persons etc.). In the case of fingerprints, there is a separate
‘elimination database’ for volunteers and a ‘special case’ facility, both of which
are managed differently.
15
Crime and Arrest Rates
2.6.
England and Wales, like much of Europe, has experienced a drop in recorded
crime levels from the early 1990s. More recently, since 2004/05, the annual
total of crimes recorded by the police has been falling, reaching approximately
4.9 million crimes in 2007/08.50 However, despite the drop in crime, the number
of arrests annually has been rising. In 2007/08, there were 1,475,266 arrests
(albeit a slight dip of 0.5% in the usual year on year rises). Nearly a quarter of
all arrests (24%) result in a caution, and these cautions contribute significantly
to the overall detection rate, which in 2007/8 was 27%.51
2.7.
The standardisation of arrest criteria in 2005 make comparisons of arrest rates
across years problematic. However, between 2002 and 2006 there was a rapid
escalation in numbers of children and young people being arrested – a rise far
greater than in the adult population, and most marked in the 13-15 age
bracket.52 Many of these arrests, it is claimed, are for ‘trivial’ reasons, and some
have asserted, most plainly the past Head of the Youth Justice Board, Rod
Morgan, that many were arrested to reach police arrest ‘targets’ and that young
people were increasingly being criminalised.53
Fingerprints and IDENT1
2.8.
Fingerprint evidence has been used in criminal investigations and judicial
processes for more than a century, initially to confirm questioned identity, and
subsequently to assist in the identification of individuals whose fingermarks
were recovered from scenes of crime. Current legislation means that any
person arrested in connection with a recordable offence in England, Wales and
Scotland, can have their fingerprints and palm prints taken. The fingerprint
images, along with nominal data, will be entered onto the national fingerprint
database and linked to a corresponding arrest record which is entered on the
PNC in England & Wales, or the CHS (Criminal History System) in Scotland.54
2.9.
Fingerprints were first automated nationally in England & Wales in 1987 and
significant technological developments are still continuing. IDENT1 stores the
fingerprint and palm print databases (as well as the footwear impression
database), and uses an Automated Fingerprint Identification System (AFIS) to
compare fingerprints held on the national fingerprint database with those
database holding unidentified marks from crime scenes.
50
http://www.homeoffice.gov.uk/rds/pdfs08/hosb0708summ.pdf .
Home Office, 30 April 2009 Statistical News Release: Police Powers and Procedures England and
Wales 2007/08. Available at http://www.statewatch.org/news/2009/may/uk-ho-stats-police-prel.pdf .
52
Home Office Statistical Bulletins: “Arrests for Recorded Crime’ 02/03 – 05/06.
53
See Centre for Crime and Justice Studies, ‘Summary Justice: Fast but Fair?’ (August 2008). Available at
http://www.crimeandjustice.org.uk/summaryjusticefastbutfair.html .
54
Scotland removes these records on acquittal.
51
16
2.10.
LIVESCAN electronic scanners linked to the National Automated Fingerprints
Identification System (NAFIS) are also available in most police stations for
almost instantaneous searching of fingerprints taken from individuals against
the over 8 million sets of fingerprints held on IDENT1. Mobile scanners are also
available for use on the street to search individuals against the Police National
Computer. Earlier LANTERN hand held mobile fingerprint consoles are now
being replaced, firstly under Project MIDAS (‘Mobile Identification at Scene’),
and now by the UK’s ‘National MobileID’ service.
2.11.
IDENT1 has been designed, developed, and is managed by Northrop
Grumman Information Technology (until at least 2013), a contractual service
which is overseen and controlled by the NPIA. This service is also increasingly
used by the UK Border Agency for checking of biometric visas and passports
and processing of visas. Up to 8,500 visa applications a day are anticipated to
be using IDENT1 for identification in the near future, and IDENT1 is expected to
play a crucial role in the creation of links with the Schengen SIS II, a Europeanwide searching system for law enforcement agencies.
DNA Profiles, Samples, and the NDNAD
2.12.
DNA profiling has been used by the police to investigate crimes for over 25
years.55 Despite some early and well-publicised difficulties in its acceptance by
the Courts, especially in the United States, it is now more often portrayed as the
standard-bearer of forensic science quality, with claims made for its unrivalled
power to inculpate the guilty as well as exculpate the innocent. In order to
facilitate the use of DNA profiling, and maximise its utility, the National DNA
Database (NDNAD) was established in 1995, growing rapidly with the aid of
legislative amendments extending police powers to take and retain DNA over
subsequent years. There has also been significant financial investment in the
technology. The history of the NDNAD, assisted by the comparatively greater
openness of this database than other police or government bioinformation
systems, has already been charted by a number of studies.56
2.13.
Until the recently passed Crime and Security Act 2010, there was no statutory
basis for the governance of NDNAD. Upon establishment, it was initially
operated and overseen by the Forensic Science Service. However,
Custodianship is now in the hands of the NPIA, which was established by the
Police and Justice Act 2006, commencing operations in April 2007. The NPIA is
a police-led, central resource, governed by a tripartite arrangement – with the
Home Office, the Association of Chief Police Officers (ACPO) and the
Association of Police Authorities (APA). All NPIA work is overseen by a Board,
55
The scientific background to forensic DNA profiling is described in a number of publications. A useful and
short introduction may be found in Chapters 1 & 2 of the Nuffield Council on Bioethics: The Forensic Use of
Bioinformation: Ethical Issues’ (September 2007).
56
Williams. R. & P. Johnson, ‘Genetic Policing’ (Willian Publishing, Cullompton, 2007).
17
which must ensure that they reach their aims and objectives and that they
comply with statutory and administrative requirements for the use of public
funds.57 As with predecessors however, the NPIA is not immune to change or
disbandment, with suggestions that the NPIA may be removed by a
Conservative government and the recent announcement of the impending
departure of the Chief Executive.
2.14.
A ‘full DNA ‘profile’ held on the NDNAD consists of the 20 numbers that are
measured by the SGM+ multiplex, together with a sex marker (Amelogenin).
Each number represents a measurement of the repeat frequency (‘peak’) for
each of the two alleles at the positions or loci along the chromosome selected
for analysis or typing as a DNA profile within the multiplex (sometimes called
‘marker’) kit or system. Thirty-six additional non-genetic data fields are available
for each sample profile, although not all records will have entries for all 36 fields
as this will depend upon their origin.58 At the present time, the DNA ‘sample’ –
the DNA extraction made from the cellular material provided by individuals or
collected from crime scenes – is also retained by the processing laboratory on
behalf of the police force that originally collected the material.
2.15.
The NDNAD is linked to the Police National Computer (PNC), maintained and
operated by the NPIA, and provided for by s.27(4) of PACE. Regulations made
under that section also regulate the information that is to be kept, including
details of people, vehicles, crimes and property. The PNC can be accessed by
the police and other criminal justice and related agencies with approximately
185 million transactions during 2008. Transactions are currently growing by
approximately 10% per year.59
2.16.
Changes to policy, requirements to keep data to support the retention of DNA
and fingerprints, a recent Court ruling, and a failure to follow this issue through
during the Commons Committee stage of the Crime and Security Act 2010
mean that records on the PNC are no longer ‘weeded’, and old or minor
convictions, or arrests that do not result in a conviction (i.e. arrests with NFA)
are ‘stepped down’ (i.e. not immediately accessible to non-police agencies), but
are retained by the police and may be revealed on a ‘standard disclosure’ to
non-police agencies. This has implications for all individuals who are arrested,
as these records will remain permanently and being ‘a person known to the
police’, may impact upon visa and job applications for example.
57
NPIA Business Plan 2009 – 2010. p.15. Available at: http://www.npia.police.uk/en/7403.htm .
For a description of all the data entry fields, see: Home Office, ‘Keeping the Right People on the DNA
Database’ (May 2009).
59
www.npia.police.uk accessed 1 December 2009.
58
18
Cost
2.17.
Individual police forces pay their forensic provider (with whom they have a
contract) for DNA testing. Each force must then budget for their needs and will
spend varying amounts on DNA testing and associated activities (such as crime
scene recovery). Fingerprinting may be harder to accurately cost as fingerprint
bureaux are operated locally by each force. It has been commonly believed that
because of this ‘in-house’ provision of fingerprint analysis, fingerprints are
‘cheaper’ than DNA testing, though this may not necessarily be the case or a
particularly meaningful statement.
2.18.
The cost of running the NDNAD, including the monitoring and accrediting of
forensic DNA suppliers increases as the number of suppliers increases, though
in recent years, some of this cost has been passed onto suppliers. However,
there has still been an increase in costs, notwithstanding the one-off 2008/09
cost of the transition of the NDNAD operation from the FSS to the NPIA.60
Maintenance costs are approximately £200,000 per annum while development
costs for future years are yet to be publicly made available by the NPIA. The
total cost of the NDNAD in 2008-09 was £4,290,500 (this includes both capital
and running costs; it is not possible to separate the two, as well as almost £1m
in costs relating to the transfer to the NPIA).61
2.19.
The NPIA will not reveal the costs associated with IDENT1/ Livescan because it
forms part of a private contract with Northrop Grumman and is therefore
considered commercially sensitive. On this basis, the NPIA has claimed
exemption from the right of access in accordance with sections 41(1) and 43(2)
of the Freedom of Information Act.62 However, in 2004, it was reported that
Northrop Grumman won a US$244 million contract (approx. £125m at 2004
exchange rates) to provide advanced biometric identification technology to the
UK police. The contract was for eight years with three additional option years,
and followed the NAFIS contract that it held from 1995.63 The NPIA recently
awarded Detica an £18.1m contract to deliver an Automated Fingerprint
Identification System (AFIS) to provide a nationwide mobile biometric
identification capability through a selection of different hand-held devices.64
60
NPIA website, accessed on 1st December 2009.
Lord West of Spithead, Written Answers - HC Deb, 2 December 2009, c51W.
62
Section 41(1) exempts information if it was obtained by a public authority from any other person and the
disclosure of the information would constitute an actionable breach of confidence. This exemption is
absolute so does not require a ‘public interest’ test to be undertaken. Section 43 also exempts information if
it is likely to prejudice commercial interests, although this is a qualified exemption, and the balance of public
interest and the benefit to the community at large, should be considered before disclosing the information.
63
Northrop Grumman Wins U.K. Fingerprinting Job, The Washington Post, December 20, 2004; Page E04.
64
‘$25m mobile biometric ID deal awarded by UK’s NPIA’ Biometric Technology Today, March 2010, p.1.
61
19
The International Context
2.20.
The Secretary General of Interpol has observed that the use of DNA on a
significant scale has been restricted to a club of wealthier countries.65 While this
may remain true, the latest Interpol survey shows that over 50% of countries in
all regions, except Africa, use DNA profiling in criminal investigations. There are
120 Interpol countries using DNA profiling and 54 of these have national DNA
databases with 26 countries planning to introduce a DNA database.66 The
numbers of fingerprint databases across the globe are not detailed, but will
likely be far higher, with the century of use and ‘affordability’ of fingerprinting.
2.21.
International cooperation involving the exchange of DNA, however, is in its
infancy. Information is scarce and despite publicity for some successful cases,
cooperation appears to be far from regular procedure even for investigating
serious crimes. Within the EU this might change as a result of the Prüm Treaty,
although there is evidence to suggest that the introduction of technologically
advanced searching systems and databases will not necessarily result in major
increases in activity. Indeed, the Council of the European Union has already
taken action to restrict the level of dactyloscopic searching under Prüm,
presumably to prevent analysts being overwhelmed by additional searches.
This in turn suggests that the amount of cooperation involving the use of
fingerprints is much lower than is commonly imagined, but while information
about the volume of UK involvement in international cooperation is rare, data
about the exchange of fingerprint information appears to be non-existent.
2.22.
There are various technical obstacles to greater international cooperation the
most significant being the imminent technological obsolescence of the SGM+
multiplex. In contrast to the Government’s earlier strategy for upgrade of
profiles from SGM to SGM+, the absence of retained samples as a result of the
Crime and Security Act 2010 may mean that retained profiles may frequently
have little value for international comparison purposes well before the normal
minimum six year retention period under that act expires.
2.23.
The modest ambitions for enhanced criminal justice cooperation internationally
are dwarfed by the anticipated scale of activity intended for the purposes of
border control. Inevitably this will result in the vertical integration of searching
systems and databases with criminal justice resources and operations.
Potentially the amount of information held by the state, as a result of (quasi-)
consensual access to bioinformation, could make existing arrangements based
on forfeiture increasingly irrelevant, particularly as fingerprints and, possibly,
DNA are used for border control purposes.
65
Quoted in Wilson, T (2009) ‘Forensic Science and the Internationalisation of Policing’ in Fraser, J. G. and
Williams, R. (eds.) Handbook of Forensic Science (Cullompton: Willan Publishing), p.506.
66
INTERPOL, Global DNA Profiling Survey 2008, Key Facts. Available at www.interpol.int .
20
3
EVIDENCE AND ASSESSMENT
Evidence and assessment
3.1.
In the course of its deliberations on the legality of the UK forensic DNA
collection and retention regime that has been in place since 2001, the
European Court of Human Rights considered the quality of the evidence
presented to the Court concerning the usefulness of the retention of DNA
samples and profiles from those who, although arrested by the police, remained
unconvicted of any crime. In its judgment, the Court observed that
“... the statistical and other evidence, which was before the House of
Lords and is included in the material supplied by the Government
appears impressive, indicating that DNA profiles that would have been
previously destroyed were linked with crime-scene stains in a high
number of cases.”
However, this seemingly impressive appearance did not withstand detailed
scrutiny, and the Court concluded that:
“neither the statistics nor the examples provided by the Government in
themselves establish that the successful identification and prosecution
of offenders could not have been achieved without the permanent and
indiscriminate retention of the fingerprint and DNA records of all persons
in the applicants' position”.
The judges in this case were evaluating a very narrow range of claims made by
the Government; that related largely to the usefulness of retaining forensic
bioinformation from individuals who had not been convicted of criminal
offences. Nevertheless, their comments, both here and elsewhere in their
judgment, give pause for thought about the quality and quantity of evidence that
currently exists concerning the effectiveness of the uses of bioinformation more
generally. It is this issue that will be the focus of this section of the report.
3.2.
The dismissive view of the Court surprised many key actors in the UK network
of forensic bioinformation advocates and users, especially since strong claims
about the usefulness of forensic bioinformation for the investigation of crime
has been a constant factor in the UK policy debate. Early FSS estimates of the
usefulness of the NDNAD for the detection and reduction of volume crime were
undoubtedly influential in securing funding for the expansion of the NDNAD
through increases in collection rates from subjects and crime scenes between
2001 and 2005.67 The subsequent modification of the ‘FSS General Crime
67
Although the DNA Expansion Programme was established in order to fund increases in the collection of
DNA, it quickly became clear that the work of the new cadre of crime scene examiners employed through
21
Reduction Model’ by the ‘MHB Model’ provided by the authors of the
‘Pathfinder’ study did not substantially change this optimistic view, and official
statements concerning the NDNAD have further emphasised its significance to
the detection of both volume crime and serious crime.68 Assertions made in a
key document that summarised the effectiveness of the Expansion Programme,
especially those that favourably compared the overall detection rates for
volume crimes to the detection rates when DNA crime scene samples were
loaded onto the NDNAD, are frequently repeated in the UK and elsewhere in
support of the establishment or expansion of DNA databases elsewhere in the
world.69 Whilst most such claims focus only on the detection, rather than the
prosecution of offenders, the recent Magee Report suggested that “improved
use of biometrics in prosecuting and convicting could lead to 1% more
convictions of those tried at court.”70
3.3.
However, alternative voices in the UK debate on forensic bioinformation have
often argued that there is a need to improve the research base from which to
launch any such assertions. The House of Commons Science and Technology
Committee called for more research in their report ‘Forensic Science on Trial’ in
2005, and it was also the view of the Nuffield Council on Bioethics, as cited by
the European Court of Human Rights. This view has subsequently been
endorsed by other authorities, including the Human Genetics Commission.71 A
series of commentaries and papers by Genewatch, often in response to specific
Ministerial statements, or as submissions to consultation exercises, have
provided very detailed critiques of data quality as well as the credibility of the
interpretations of existing data on the use of DNA profiles in criminal
investigations offered by official spokespersons.72
this programme would result in an increase in fingermarks as well biological samples collected from crime
scenes.
68
Both models are fully outlined in Burrows, J. et.al. (2005) Forensic Science Pathfinder Project: Evaluating
Increased Forensic Activity in Two English Police Forces. London: The Home Office.
69
See pages 16-17 of Home Office Forensic Science and Pathology Unit (2006) DNA Expansion
Programme 2000-2005: Reporting Achievement. London: The Home Office.
70
Sir Ian Magee, ‘The Review of Criminality Information’ (The Magee Report) (July 2008) Home Office,
London. p.17.
71
Statements to this effect can be read both in the HGC response to the Consultation Paper and, in more
detail, in their 2009 Report: ‘Nothing to Hide, Nothing to Fear’.
72
A full collection of these papers can be found on the Genewatch website: http://www.genewatch.org/sub539478 accessed 16th April 2010.
22
3.4.
Following the delivery of the European Court judgment, the UK Government
commissioned additional work which sought to provide further evidence on the
usefulness of retaining fingerprints and DNA profiles from those not convicted
of crimes.73 This work was explicitly used to support proposals for changes in
the retention regime that would meet the Court’s demands, but following its
publication, the study was subjected to a series of critical commentaries. Its
standing now remains unclear. Rather than engaging directly in this dispute we
want briefly to outline the currently available official data sources on which any
such evaluations can be based, as well as what can be learned from some of
the existing academic studies of the police uses of forensic bioinformation.74 In
the course of doing so, we want to explain why so many observers find the
existing data, and many of the current studies, wanting in the support they
provide for an authoritative account of the effective uses of forensic
bioinformation in criminal investigations and prosecutions. We begin by
describing the different kinds of information that circulate in the public realm
before offering a commentary on its quality and relevance.
3.5.
The most basic level of facts relevant to understanding the operation of forensic
bioinformation regimes in any criminal jurisdiction relate to the amount of such
information held by relevant authorities together with other significant facts
retained about the individuals whose biometric data have been captured for
inclusion. We provide below some recent figures on the number of fingerprint
and DNA profiles holdings in relevant databases in England & Wales.
3.6.
According to the most recent statistics issued by the National Policing
Improvement Agency, in October 2009, IDENT1 held:
•
•
•
•
•
•
8.1 million individuals' ten-prints
17.8 million sets of ten-prints
1.8 million unidentified marks
7.9 million palm prints
156,210 palm marks
4,396 serious crime marks
It can be seen – from the figures shown in the next section – that the size of this
collection is much greater than that of the current DNA database. This is of
course an historical artefact resulting from the longer history of the collection of
fingerprints from offenders and recovery of fingermarks from scenes of crime.
However, it is important to note that the size of the fingerprint collection has
often be used to anticipate the eventual size of the NDNAD, and since its
73
Especially the study by Ken Pease appended to the Consultation Paper: Annex C: DNA Retention after S
and Marper: Ken Pease, Jill Dando Institute. April 2009.
74
Whilst our description of existing data sources are limited to those held in England & Wales, we also
introduce some examples of recent research from other jurisdictions to indicate the kinds of studies that
might be possible here in the future.
23
establishment in 1995, the collection of DNA subject samples and fingerprints
from those arrested on suspicion of involvement in a criminal offence has been
treated as an equivalent necessity. Accordingly, since that time and with some
variation resulting from local arrangements, the number of new DNA subject
samples loaded onto the NDNAD has been equivalent to the number of new
tenprint records loaded onto the fingerprint database. At the same time,
recovery rates of these two different forms of bioinformation from crime scenes
remains very different: the number of fingermarks recovered from crime scenes
remains considerably higher than the number of biological samples recovered,
but the attrition rate is also much greater. In view of these facts it seems
surprising that more efforts have not been made properly to assess the
usefulness of fingerprints to crime detection, particularly as enhancing their
value to investigators was one of the intentions of moving from the DNA
Expansion Programme to the Forensic Integration Strategy.75
3.7.
The most recently available figures on the NDNAD76 show that the database
held the following:
•
•
•
•
5.6 million subject sample profiles
4.9 million (est.) individuals who correspond to those sample profiles77
354 thousand unmatched crime scene sample profiles;
39 thousand sample profiles from volunteers.78
In recent years, approximately 50,000 subject samples have been loaded onto
the NDNAD each month. Crime scene profiles are deleted from the NDNAD at
the request of the police following a conviction for the offence or when a
decision is made that it will not be investigated further. Between 2007 and
March 2009, 52,688 crime scene profiles were removed from the NDNAD.79
3.8.
Individuals who wish to have their profiles removed from the NDNAD must
make their request to the Chief Constable of the force who obtained their DNA.
Between March 2007 and March 2009, 445 subject profiles were deleted from
the NDNAD under the ‘Exceptional Case Procedure.’ However, the law
regarding DNA retention differs in Scotland so during the same period, 35773
Scottish subject profiles were deleted from the NDNAD.80
75
Home Office (2006), DNA Expansion Programme 2000-2005: Reporting Achievement, London, Home
Office, p. 20.
76
Taken from NPIA website, accessed on 1st December 2009.
77
It is currently estimated that as at 16 October 2009 13.8% of the subject profiles held are replicates.
78
These figures do not include sample profiles held on the Police Elimination Database (PED). PED
searches are carried out only if a senior investigating officer requests a comparison of profiles from a
specified officer or officers with a DNA profile from a specified crime, unlike NDNAD searches which
compare all profiles from crime scenes with all profiles from known individuals. Since 1 August 2002, all new
recruits to the police service have been required to give a DNA sample as a condition of employment. As at
30 September 2009 there were 118,699 records held on the PED for England and Wales police forces.
79
NDNAD Annual Report 2007-2009, p.25.
80
ibid. p.25.
24
3.9.
On 24th April 2009 there were an estimated 986,185 persons included on the
NDNAD who had no conviction recorded on the PNC either because their PNC
record has been ‘weeded’ and has therefore been deleted, or because
proceedings against them are still ongoing, or because or they have never
been convicted of an offence.81 These subject profiles without corresponding
PNC records are often confusingly referred to as ‘innocent’ people on the
NDNAD although it is only the third of these groups that would correctly be
called ‘innocent’ (and those in the second group are yet to be decided upon so
may or may not join this group). We have already noted that the retention of this
group of subject profiles is subject to considerable contestation, and we will
return to discuss this group later in this section of the report.
3.10.
It may seem intuitively obvious that the larger the number of profiles and
fingerprints held in any forensic database, the more useful will be these
holdings to the process of criminal investigation since they will eventuate in
more matches and identifications. The rising match rate observed throughout
much of the time that the NDNAD has grown, may seem to support this general
idea, although there have been stutters in the upward trajectory. However, the
most recent comparative data on DNA database sizes and match rates
provided by ENFSI shows no simple association between the two measures,82
and it is likely that important determinations of such rates include the extent to
which legislation and practice eventuate in the collection of bioinformation from
the most relevant category of individuals, and the assiduousness with which
biological material and fingermarks are collected from scenes of crime.83 In
addition, care has to be taken when comparing match rates between different
DNA databases (and sometimes between the same database at different points
in times) since match counting rules may differ.84
Forensic Bioinformation: Existing Data Sources
3.11.
IDENT1 is used to compare fingermarks at crime scenes with fingerprints held
in the National Fingerprint Collection. Automated searching – and the candidate
identification of crime scene marks with the fingerprints of individual suspects –
is supplemented by expert evaluation and decision-making. The NDNAD is
used to automatically compare crime scene profiles with profiles obtained from
suspects for the offence in question as well as profiles obtained from individuals
arrested in connection with a recordable offence. Some data are provided by
81
ibid. p43.
ENSFI (2009) DNA-Database Management Review and recommendations at
http://www.enfsi.eu/get_doc.php?uid=345, accessed 6 March 2010. We later take up the issue of the
relationship between match rates and the more difficult question of the significance of, and uses made, of
such matches.
83
Attentiveness to this categorical issue can be seen in the various descriptions of the population that the
NDNAD was hoped to capture. These have included ‘the active criminal population’, ‘the active known
criminal population’, and ‘the known active suspect criminal population’.
84
See comments on match counting rules for scene-to-scene matches in particular by Buckleton, Bright and
Walsh (2009) ‘Database crime to crime match calculation’ Forensic Science International: Genetics 3: 200201.
82
25
IDENT1 and the NDNAD on the number of identifications and matches that are
provided to the police each year. Examples of these are shown below.
Fingerprints
3.12.
From April to October 2009, IDENT1 made 47,783 crime scene ‘identifications’,
averaging 85000 identifications a year. In addition, it verifies the identity of over
1.5million arrestees per year. There are no data on the number of identifications
that led to detections or convictions. Presently, there are 2000 identity checks
being processed via mobile devices per month, and the UK Borders Agency
uses IDENT1 to check over 4000 identities per week.85 However, no data are
centrally provided from IDENT1 on the uses made of these identifications to
support the detection or prosecution of offenders.
DNA Profiles
3.13.
Crime scene profiles may ‘match’ with both subject profiles as well as with other
crime scene profiles. Subject profiles can also match with each other.86 In 200809, the average match rates between crime scene and subject profiles when a
new crime scene profile was loaded onto the NDNAD was 58.7%; when a new
subject profile was loaded it was 2.3%.87
3.14.
A ‘match’ refers simply to two profiles being found to match on the database.
These will include cases where an individual whose DNA profile is found at a
crime scene may have a legitimate reason for their biological material being
present (e.g. they are a victim, a witness, or otherwise legitimately present). In
such instances, there will be no ‘detection’ resulting from the match. A ‘DNA
detection’ refers to cases where a crime has been cleared up and a DNA match
was available to investigators. However, such ‘DNA detections’ cannot be
assumed to have resulted from the DNA match itself. A supplementary category
of ‘Additional DNA detections’ includes cases in which an offender being
presented with his DNA ‘match’ for a case, then admits to further offences.88
3.15.
‘Multiple matches’ also occur in cases where a crime scene matches with more
than one subject profile. Such multiple matches normally arise because a
partial crime scene profile matches against a number of full subject profiles or
because the same person’s profile is held on the NDNAD under different names
(a replicate). Between May 2001 and March 2009, 306,379 crime scene profiles
have been identified in 282,861 match groups.89 For 226,393 of these, a single
subject was reported as linked to the crime scene by their DNA.90
85
NPIA Business Plan 2009-2010, p.31.
In the case of ‘replicate profiles’.
87
NDNAD Annual Report 2007-2009, p.33.
88
ibid. p.35.
89
ibid. p.28. A ‘match group’ consists of two or more DNA profiles that are indistinguishable or compatible.
90
supra.
86
26
3.16.
Since 1995 (until March 2009) 556,794 crime scene profiles have been loaded
onto the NDNAD and during the decade 1998/99 to 2008/09 there have been
410,000 crime scene to subject matches, which may have been useful to the
investigation, if not directly identifying the offender. In 2008/09, there were
36,727 matches between crime scene and subject profiles, of which 17,607 led
to DNA detections, and 14,602 'additional' detections, a total of 32,309 crimes
detected in 2008/09 in which a DNA was available.91 While taken as a
proportion of the number of crimes reported (approx. 4.9 million), this number is
very low, yet there are proportionally few crimes where there are ‘scenes’ that
can be usefully searched, and many other factors that will preclude the use of
DNA evidence (including in most instances all those involved already being
identified). In 2008-09, just 796,780 crimes were subject to a crime scene
examination (17% of recorded crimes), meaning approximately 2.21% of all
examined crime scenes led to DNA detections in 2008/09.92
3.17.
There are very considerable difficulties in interpreting the significance of these
kinds of figures provided by the NDNAD when attempting to assess the general
usefulness of genetic forensic bioinformation for the investigation of crime.
Ministers have frequently used answers to Parliamentary Questions to provide
seemingly better summary data, but even here, the same difficulties recur. In
the following paragraphs we consider one example of such a House of
Commons answer in order to illustrate these issues. In this case, the Secretary
of State for the Home Department provided a table of results (shown as Table
3.1 overleaf) to the House of Commons on 25th January 2010 (HC 642W).
3.18.
The written statement includes a number of claims, caveats and explanatory
notes to the table. Perhaps the most important of these is the assertion that the
data in the table ‘under-represent the overall contribution of DNA matches to
the detection of serious crimes such as murder and rape.’ The reason for this is
that the figures for detections do not include those cases in which ‘one-off
speculative searches of the NDNAD’ played a part in a successful detection,
nor do they include those cases in which only laboratory-based DNA profile
comparisons were deployed. The former are usually carried out when crime
scene profiles are insufficiently complete to be suitable for NDNAD loading, and
the latter are carried out without the direct deployment of the NDNAD. The
Secretary of State also asserted that these DNA routines are deployed more
commonly in serious crime investigations and it is the neglect of these that may
result in an under-representation.
3.19.
However, since no data are provided on the number of such comparisons, it is
obviously difficult to assess their effect on the overall picture provided. Other
caveats included in the written answer include the standard remark that
91
92
NPIA website accessed on 1st December 2009.
NDNAD Annual Report 2007-2009, p.27.
27
‘detections are achieved through integrated criminal investigation, not through
DNA alone’, but it is interesting to note that in this instance, the usual preceding
modifier ‘serious crime’ has not been used. Finally, it is noted that ‘The data
provided are management information and have not been formally assessed for
compliance with the Code of Practice for Official Statistics.’ The statistics have
been compiled from individual police forces’ forensic performance data, albeit
standardised data collected according to Home Office pro-forma’.
Table 3.1
28
3.20.
Some general observations on these data are in order. First, that the category
of ‘DNA Detections’ (defined as detected crimes ‘in which a DNA match was
available’) remains heterogeneous. The table gives no information about the
role played in an investigation by the DNA match, although there are data fields
on the relevant Home Office pro-forma which require some indication of the role
played. Even without this information, it seems intuitively obvious that the
availability of DNA matches in detecting 70 of the 4,583 homicides detected in
2008-2009 will differ according to the nature of the varied cases (including
murder, manslaughter, infanticide, attempted murder, the destruction of a viable
unborn, threat to murder, and causing death by dangerous driving, etc) and the
dynamics of particular investigations.
3.21.
Second, it should be noted that since the role of a DNA match in aiding the
original detection remains unexplicated, it is difficult to know what significance
to place on figures given in the adjacent column of ‘Additional detections arising
from the DNA match’. Without this, and despite the Government claim that
since these detections ‘arise from a crime with a DNA match; and are therefore
detections where a DNA match played a part in solving the crime’, it hardly
seems necessary to add the concept of ‘Total DNA–related detections’, which
combines together the first two categories, increasing confusion.93
3.22.
If these observations give some sense of the difficulty of reaching simple
conclusions from these data, this difficulty is multiplied when questions are
asked about the effect of the retention of particular categories of persons.
Many such questions have been asked of Ministers in Parliament, but a brief
survey of their replies indicates the level of ambiguity and imprecision that
surrounds them.94
3.23.
Better summaries and analysis may be found elsewhere. For example, the
ACPO Criminal Records Office (ACRO) recently has become involved in the
analysis of existing data on behalf of the NDNAD Strategy Board. One of their
recent studies determined that in 2008/09, there had been 818 subject profiles
matched with homicide or rape crime scene profiles. In 82 of these matches,
the subject did not have a prior conviction, and their DNA was matched when
the crime scene profile was loaded. Senior detectives involved in these 82
cases reported that in 43 of them, the match was ‘important’, although the
various meanings that may be captured by the use of this term remain open to
93
There is one other major issue about these data: the use of proportions in the final two columns may be
statistically inappropriate since figures for recorded crime and for detections relate to the same year, but not
necessarily to the same data-set. Force detections achieved in the year 2008-2009 may include the
detection of crimes recorded in earlier years, and crimes recorded in 2008-2009 may be detected in
subsequent years. It is conventionally argued that this ‘rolling proportion’ remains roughly accurate, but it
does not seem to represent good statistical practice. Once again it is something that can be corrected if and
when proper case-tracking data become available for analysis.
94
Once again, Genewatch papers can be consulted for the details of many of these.
29
speculation.95 More recently, the government presented five case studies
intending to demonstrate the power of retaining the DNA of ‘innocent’
individuals, in subsequently detecting further offences. Of these five however,
two were the same case, and of the four remaining cases, one of the rapes
occurred inside the suspect’s bedsit, and a burglar had recently been dismissed
from employment at the burgled premises and he had been caught on CCTV,
leaving the studies lacking much persuasive force.
Police Force Returns
3.24.
Is it possible to develop a better grounded understanding of current police uses
of forensic bioinformation through the analysis of locally collected data? All
Police Forces in England & Wales submit periodic returns to the Home Office
that include data on their uses of forensic bioinformation. The purpose of these
returns is largely to monitor the performance of Scientific Supports Units, and
this purpose shapes decisions about what information to collect from whom,
and how it should be analysed, reported and used. These same data may also
be examined within forces in order to compare the performance of different
units (and sometimes different individuals) within their own organisation. The
focus of all of these national and local efforts has been to standardise data for
meaningful inter and intra-force comparisons, at least as far as the uses of
bioinformation to support the investigation of volume crime is concerned.
3.25.
However, even when such periodic returns are accurately recorded, they do not
necessarily allow an adequate estimation of the quality and quantity of the
contribution of forensic bioinformation to criminal investigations and detections.
Partly this is because not all instances of the collection of forensic
bioinformation are recorded. For instance, if a non-police officer recovers
biological material (i.e. a pathologist/ forensic medical examiner) then this will
not be recorded in the force periodic returns. There may also be many
instances in which the failure to match a DNA profile or identify a fingerprint to a
particular suspect may appear to be a negative result, but such negative results
may play an important part in excluding particular individuals from further
investigative inquiries and/or by giving further direction to an investigation.
3.26.
Individual forces may also supplement the data required by national agencies
with their own data, for example by recording the temporal order in which
particular forensic results were obtained, the time taken to achieve them, and
the time taken to act on them. But these more detailed efforts normally follow
the use of bioinformation only to the point of ‘detection’ in the criminal process,
and even here there can be difficulty in assessing the manner in which the
availability of bioinformation assisted the achievement of a detection. This is
especially true in the case of serious crime investigation, but is also the case in
95
NPIA website, accessed 1st December 2009.
30
routine volume crime investigations where administrative systems do not
necessarily permit nuanced recording of the forensic contributions to the
investigative process.
3.27.
Prior to the introduction of The Scientific Work Improvement Modelling
Programme (SWIM), the Police Standards Unit (PSU) utilised a simple linear
model in which it was assumed that simply attending more scenes to collect
additional forensic material would result more detections. SWIM was developed
in response to concerns that where there were backlogs in processing or acting
on bioinformation and other forensic intelligence, greater scene attendance
would only add to the backlogs. Greater attendance would not necessarily
improve detections or to the degree suggested by some commentators. The
assiduous collection of fingermarks and biological samples from scenes of
crime has to be matched by the adoption of equally rigorous and efficient
processes all the way through the CJS. Any analysis of the effective uses of
bioinformation based on police force returns will have to be able to map the
changes in significance of forensic bioinformation as progress is made through
successive stages of the criminal process.
3.28.
An alternative source of data - The Police National Computer (PNC) was not
designed to facilitate the collection of data relevant to questions about the
effective uses of forensic bioinformation, or deliver large ‘chunks’ of data. The
PNC data are limited largely to demographic/ operational and conviction data.
Forensic Bioinformation: Bespoke Research Studies
3.29.
If the analysis of police force returns currently does not provide a sound basis
for assessing the effective uses of forensic bioinformation, it is disappointing to
note the seeming absence of an accumulating body of knowledge derived from
existing research studies in this field. In the UK, research has been mostly
undertaken/ funded by the Home Office and disseminated through Home Office
channels. Whilst some studies, like ‘Pathfinder’ have sought to track individual
cases through the criminal justice process, this process has proved to be
difficult and time consuming. Other Home Office funded studies have sought to
throw light on the police uses of forensic bioinformation, but it seems that
enthusiasm for commissioning such studies by independent researchers has
waned over the last few years. There is insufficient space to provide details of
all of these studies, but readers interested in their details can consult a
systematic review of findings relating to volume crime investigations published
by the Home Office in 2005.96 Additional analysis of, and arguments
96
Bradbury, S-A. & Feist, A. (2005) The Use of Forensic Science in Volume Crime Investigations: A Review
of the Research Literature. London: The Home Office.
31
concerning, such studies can be found elsewhere.97 While all suggest that the
role played by forensic bioinformation in detecting crime is significant, the
detailed trajectories of its uses – its successes and failures - remain poorly
articulated. Additional uncertainties exist about the uses of forensic
bioinformation in serious crime investigations. There is hardly any published
research that focuses on these kinds of cases, and this is especially difficult to
understand given the Government’s reiteration of claims for the importance of
DNA databasing for the detection of murders and rapes.
3.30.
In the absence of independent peer-reviewed research on the use of forensic
bioinformation it is difficult to know the extent and quality of the knowledge base
on which police and other criminal justice actors base operational and policy
decisions in this domain. Whilst NPIA Guides and ACPO and other Manuals
may seek to represent the ‘best’ utilisation of forensic bioinformation, such
prescriptive accounts may depart from actual practice in many cases. Manuals
are not a substitute for empirical research on this topic. There is little
collaboration between the police and external researchers, and internal reports
and documents are not widely circulated outside of operational circles.
3.31.
Whilst it may be difficult to undertake research in this field (especially in the
investigation of serious crime) there is scope for doing so if all parties to such
work understand its potential benefits. However, it would be important that such
work was not overwhelmed by the immediacy of political preferences and
instead was able to develop through a joint commitment to the production of
high quality evidence relevant to the operational uses of bioinformation.
3.32.
It will also be very useful to compare the situation in England and Wales with
policing research elsewhere in the world. There may be better integration of
academic research in other European jurisdictions, although this assumption
needs to be tested. The situation in North America is also unclear, although NIJ
funding has been used to facilitate some research on the uses of forensic
science to support criminal investigations. The recent NAS Report may
encourage the further development of academic/police partnerships.
3.33.
Two recent US studies provide a sense of what might be possible given
appropriate levels of funding and the development of such partnerships in
England & Wales. Both are ambitious pieces of work, the first being the only
known instance of the use of an experimental design to determine the effects of
the use of DNA profiling on volume crime investigations, the second being a
wider survey of the use of a larger range of forensic technologies, including
97
For example, Williams, R. and Johnson, P. (2007) Trace Biometrics and Criminal Investigations, in
Newburn, T., Williamson, T. and Wright, A. (eds.) Handbook of Criminal Investigation. Cullompton: Willan,
pp. 357-80., and in Williams, R. (2008) Policing and Forensic Science, in Newburn, T. (ed.) Handbook of
Policing (2nd Ed). Cullompton: Willan, pp. 760-93.
32
forensic bioinformation.98 It remains to be seen whether there is the political will
to encourage and fund such studies here.
3.34.
In the US, lessons are being learnt from DNA exonerations. It may be that a
similar de-construction of miscarriages of justice in England and Wales may be
useful and map where mistakes occur in the criminal process. Study of
exonerations in the US,99 have exposed the limitations of techniques such as
serology and bite marks, and also how scientific testimony might not be credible
or be distorted in cross-examination or when a case is summed-up. This is
something within the remit of the CCRC but their only attempt to conduct such
research was a study of 80 initial referrals within which sample there were said
to be 26 examples of flawed or problematic forensic science. The absence of a
database of such cases that could be interrogated has been an obstacle to
systematically following-up concerns about expert evidence exposed by cases
such as the Cannings appeal. The Omagh trial was seen to be a clear example
where significant errors had been made during the trial – not just in respect of
the actions of police officers - but also in understanding the scientific issues
within the court room.
3.35.
The communication of forensic evidence in court is crucial and yet remains
under-researched. The significant issues of how forensic bioinformation is used/
portrayed/ dealt with during trials are highly complex. Consideration of these
issues has in turn raised questions about the ability of the adversarial system to
cope properly with expert evidence. The Law Commission have recently
examined this area and proposed reforms to the reception of expert evidence,
although their deliberations were incomplete at the time of writing.100
Cost-effectiveness and Forensic Science Utility
3.36.
The interest in the criminal justice system consumption of GDP has been
manifest from the mid-nineties onwards. While there has been a recent retreat
from the use of a battery of performance indicators (PI) in policing, demands for
the evaluation of cost-effectiveness are likely to intensify as the recession
intensifies fiscal constraints. With the highest level of CJS expenditure in G8
countries when measured as a proportion of GDP, UK police forces and
forensic providers can expect increasing scrutiny from budget managers and
their accountants.
98
The first of these studies has been published (see Roman, J.K. et al (2008) The DNA Field Experiment:
Cost Effectiveness Analysis of the Use of DNA in the Investigation of High-Volume Crimes. Washington:
Urban Institute. The second, carried out by Peterson and colleagues at the University of California is
currently under review and awaits publication.
99
Garrett, B.L. and Neufeld P.J. (2009) ‘Invalid Forensic Science Testimony and Wrongful Convictions’
Virginia Law Review, 95:1-97
100
Law Commission of England and Wales (2009) The Admissibility of Expert Evidence in Criminal
Proceedings in England and Wales: A New Approach to the Determination of Evidentiary Reliability.
Consultation Paper no.190. London.
33
3.37.
One of the major obstacles to cost evaluation in policing identified at the
beginning of this project is the problem of measuring inputs. Initially this arose
because of the difficulties in measuring police expenditure on different activities.
In recent years ACPO had effectively created a competitive market for many
forensic science services. At first sight this should result in better input
information. It was suggested by some participants in the seminars however,
that one of the results of the procurement exercise had been to create a false
sense of the cost of forensic provision, with inadequate account by procurement
managers of the ability of smaller providers to guarantee timely delivery and
quality of results. The procurement exercise may have led some to believe that
‘value for money’ was synonymous with ‘lowest price’. Also, while the cost of
DNA use could be measured (albeit imperfectly) from such prices, it is unlikely
that the cost of fingerprint work could be made available on an approximately
comparable basis.
3.38.
The two key measures that have been proposed for assessing value for money
in policing are ‘cost-effective analysis’ (CEA) and ‘cost-benefit analyses’ (CBA).
The former is used to compare input costs. The latter is more complex. Having
valued all inputs, outputs and outcomes in monetary terms, it compares
competing options for spending money to identify the highest achievable net
benefit. Further complications arise when there is inconsistency in calculations
regarding the inclusion of staff costs as well as laboratory costs for example.
More recently ONS (The Office of National Statistics) has been seeking to
develop a methodology to measure CJS productivity. This is defined as ‘a ratio
of CJS outputs to CJS inputs at constant prices’. This approach appears to
have been based on CBA; it places a greater emphasis on the importance of
outcomes compared with outputs. The ONS paper on this subject contains two
caveats about their proposed approach:
3.39.
•
It may be unclear which observed or recorded changes can be attributed
directly as outcomes to improved CJS productivity or result from other
factors (e.g. is a reduction in the theft of or from vehicles the result of
improved CJS detections achieving the detention of offenders or,
alternatively, better security by manufacturers and owners)?
•
In contrast to the ONS approach, the Administration of Justice (AoJ)
methodology that the Home Office began to develop earlier recognises the
interdependence of CJS agencies, for example, improved detection of
offenders by the police may be negated by the absence of effective followup by the CPS, courts and NOMS (National Offender Management
Service).
The ONS have also acknowledged that some outcomes have a greater value or
address more serious concerns than others, for instance, car crime is generally
more tolerable than burglary. During seminar discussion, further limitations in
34
the ability to measure value for money or the reliability of the results of such
work were identified:
•
The sometimes dispersed (temporally and geographically) benefits from
acquiring bioinformation, for example, cold case reviews might result in the
detection of crimes by using forensic information collected and then stored
for several decades, and international cooperation might result in criminals
being identified from databases created and maintained by other states.
•
Unanticipated benefits including the utilisation of investigation capabilities
for DVI purposes (e.g. the international response to the 2004 Tsunami).
•
Evaluation methodologies may also need to qualify reduced input values by
making an adjustment for any consequential loss of output in terms of
research and development investment where the ability to engage in R&D
is curtailed by procurement strategies or decisions based solely on price.
3.40.
There are many precedents in other areas of econometrics where research
combines objective and subjective measures and seek to evaluate short-term
and long-term impacts. Consideration should be given to weighting outcomes
by something equivalent to cost-utility analysis in health economics (e.g. ‘quality
adjusted life year’ that reduces inherent discrimination in CBA against people
with a lower earning capability). There have also been attempts to formulate a
multi-attribute decision-analysis in energy/ environment and health arenas to try
and elicit value-based weightings in decisions.
3.41.
The most promising area for evaluating what can be achieved through forensic
science might be major crimes. Case reviews provide an opportunity to obtain
data from case support and scientific support units. With the analysis of a
sufficient number of cases it might be possible to robustly estimate the costbenefits of forensic science inputs. Although the full range of inputs/outputs
might be overlooked, miscounted or unquantifiable (e.g. a scene of crime visit
that did not yield trace bioinformation but resulted in crime prevention advice
being acted on).
3.42.
There are always going to be different levels of ‘value’ in such evaluations.
There may always be a cash value for the decisions facilitated by particular
types of forensic evidence. There will also be intangible/ (intrinsic) ‘values’ in
terms of the ability to successfully detect crimes (community values) and these
may be harder to capture and measure. Research may be undertaken but may
require caveats because of the complexity of the social and legal dimensions
that may affect interpretation. Moreover, many pilot studies or focused work
produced results that reflected a level of resource that was not available when
working normally and such results were unlikely to be replicated in national
trends. Also much research does not answer the questions that we are now
35
asking. Attrition studies are intended to identify where cases fell out of the
system not crime reduction or displacement, and for all forms of investigation
not just forensic science. We are now trying to fit bits of research into a very
different agenda and asking questions of earlier research that it cannot answer.
3.43.
One difficulty with evaluations is the lack of stability within the criminal justice
policy-making arena – very often politicians are short-termist, and decisions to
change priorities and redeploy resources may be presented in order to give the
impression that additional resources have been provided for CJS purposes.
Independent experts can build better models and can make evaluations about
long-term values/strategic objectives for the criminal justice system etc. (one
example that longer-term CJS transformation is possible was the abolition of
capital punishment against the opposition of large segments of the press and
the instincts of some senior politicians).
3.44.
Other problems in evaluation arose when benefits take time to be realised or
are dispersed. The Dutch were able to demonstrate back in 2004 that without
necessarily achieving immediate detections, they could build up via their DNA
database patterns (by offence and location) of collaboration between different
criminals that might later prove to be significant later in the light of new
(possibly quite different) information. The Belgian database has been used to
demonstrate that DNA could link serious offences committed in that country, the
Netherlands, Germany and France. The inputs and outputs for this activity,
however, would be scattered over four jurisdictions.
3.45.
In considering the three issues of data sources, research, and value for money
evaluation, participants in seminars often made comparisons between the
medical and criminal justice domains, particularly the role in the former played
by the UK National Institute for Clinical Excellence (NICE) which enjoys
significant technical, scientific and professional independence from Government
as well as from professional and commercial interests. During seminar
discussions there was some support for considering the possibility of learning
from this health model.
3.46.
The ‘ultimate issue’ for most of those contributing to seminars was the question
of what value forensic bioinformation can be shown to add to the criminal
process/ criminal justice system. It may be necessary to consider different
methods of data collection to give more information on different ‘outputs’. This
would depend on what the ‘outputs’ were considered to be and how they may
be measured. It may not be complex to create new measures and collate data
but would require substantial resources. It may also require significant human
intervention (which is costly) because the type of data required would
necessitate judgments of relevance to criminal justice processes. Any ‘rich’ data
36
collection would also require the support of other agencies – in particular the
CPS (Crown Prosecution Service). However, the CPS previously have
expressed no interest in collection/ returning data for these kinds of research
purposes.
Conclusion
3.47.
There is wide agreement of the necessity to strengthen the current evidence
base for the use of forensic bioinformation in support of criminal investigations
and prosecutions. This can be seen in Government statements, in the work of
advisory bodies and in the reports of Parliamentary Committees. It was also a
theme throughout many of the discussions held in the course of this project. We
endorse the view of the Human Genetics Commission that “…the compiling of
an evidence base that would demonstrate the utility of the NDNAD as an
investigative and probative resource is, we believe, an urgent and important
task.”101
3.48.
The Human Genetic Commission’s recent report ‘Nothing to Hide, Nothing to
Fear’, includes an important section dealing with the issue of research and
evidence on the uses of forensic bioinformation. In this section, entitled
‘Forensic Utility’, the Commission lay out a strong case for the necessary
development of a programme of research, recommending that
“....data supporting evaluation of the forensic utility of the National DNA
Database should be collected and published by the National DNA
Database Strategy Board or the National Policing Improvement Agency
National DNA Database Delivery Unit. An evaluation of such data should
be conducted by an independent body and placed in the public
domain.”102
3.49.
The provision of enhanced NDNAD information should clearly be an element in
any programme of research, although in several of our seminars, participants
pointed out that these kinds of suggestions for the provision of ‘better data’
place immense demands on existing information management systems, and
indeed sometimes could only be met by the introduction of new and expensive
supplements to them. Intensive work would need to be done to identify the
kinds of questions that could be asked about ‘utility’ as well as the kinds of data
that might permit them to be answered. This work would itself necessitate
discussions between a number of academic, operational and policy actors.
101
Human Genetics Commission, Response to “Keeping the Right People on the DNA Database” August
2009, para. 22.
102
HGC ‘Nothing to Hide, Nothing to Fear’ (Department of Health, London, 2009) p.70.
37
3.50.
In addition to the issue of IT support for the provision of standardised
information, a strong case can be made that a wider range of studies should be
commissioned and supported by key agencies, including relevant funding
bodies, forensic providers, police users, and the Home Office. Such an
enhanced forensic bioinformation research programme will need to take into
account other forms of bioinformation. It seems obvious that the uses of
IDENT1 should be considered alongside the collection of information about the
NDNAD. However, there are also good reasons for extending the reach of
research beyond a consideration of databases as such in order better to
understand the very wide range of uses of forensic bioinformation that occur in
the course of criminal investigations. Many of these uses may be highly
significant in supporting investigations but are not well captured by counting
DNA matches between scenes and subjects, or the successful identification of
fingermarks. Some highly significant uses of DNA profiling and fingerprint
technology may not use relevant databases at all, and even when they do, it is
not necessarily a matter of using them to identify an unknown offender but for a
series of other purposes of relevance to a particular investigation.103
3.51.
Other countries, especially the USA, seem to have begun the establishment of
such a programme more easily than the United Kingdom, although such work is
at a seemingly early stage and is in danger of dominated by a preference for
one particular methodology – randomised control trials.104 In the absence of a
UK body equivalent to the NIJ in the US, and in the light of probable budgetary
constraint, any call for enhanced funding for a wider range of studies on the
uses of forensic bioinformation may seem a tall order, but it is difficult to see
how the public debate on this topic called for by Prime Minister Brown in a
recent Parliamentary answer can be properly conducted in the absence of such
a background.
103
These are very varied, but may include for example, helping to determine which of a series of suspects
handled any object of interest, inferring what courses of action and patterns of movement have occurred
within a crime scene, or which of a series of known persons may have driven a relevant vehicle. It is also
possible to imagine other research which utilises such databases for studies related to the detection and
reduction of crime in general but which do not directly focus on the investigation of individual cases.
104
The very interesting response of Ruth McNally to the HGC consultation exercise which informed the
preparation of Nothing to Hide, Nothing to Fear resonates with some of ideas about the necessity for a wider
range of studies. See http://www.hgc.gov.uk/Client/document.asp?DocId=227&CAtegoryId=4 accessed 12
January 2010.
38
4
Internationalisation and Exchange
Introduction
4.1.
This chapter is concerned with the theme of international exchange: both the
exchange of information and ideas. While both the previous report of the
Nuffield Council on Bioethics and that of the Human Genetics Commission
touched upon the international exchange of bioinformation, the paucity of data
about such cooperation has restricted the consideration that international
exchange has received, despite its relevance to important arguments about
sample retention.105 Convening a series of UK seminars and an international
symposium created the opportunity to address this problem by bringing
together criminal justice and scientific experts from a range of countries.
4.2.
The seminars reviewed international biometric exchange arrangements in terms
of validity of purpose, technical obstacles, effectiveness and ethical
considerations. These elucidated concerns about the foreseeable technological
obsolescence of the SGM+ multiplex and the operation of the Prüm Treaty.
Seminar participants also had the privilege of hearing views on international cooperation from a father whose daughter’s murderer was identified through the
international exchange of DNA.
4.3.
From the first seminar it was apparent that there was a need to address the
second quite different aspect of this subject: the exchange of ideas (or in some
instances resistance or ‘seduction’106 by outside influences) and increasing
internationalisation of decision making and technological development. The
Marper judgment and the EU Council recommendation that has resulted in the
foreseeable technological obsolescence of the SGM+ multiplex,107 are the
principal decisions of this kind for the purposes of this study. It also became
clear that some cross-jurisdictional comparisons are made without an attempt
to understand contextual variations. Discussion during several meetings
reverted to the European Court’s attempts to discern the margin of
appreciation, particularly through its references to Scots Law. Equally important
for many participants was the cautionary note in the Marper judgment for
countries at the forefront of technological innovation:
105
Human Genetics Commission, (2009) ‘Nothing to Hide, Nothing to Fear’ London, Department of Health;
especially pp.70-71.
106
Dahl, J.Y. and Sætnan, A.R. (2009), ‘ ”It all happened so slowly” – On controlling function creep in
forensic science DNA databases’, International Journal of Law, Crime and Justice (2009), doi:
10.1016/j.ijcj.2009.04.002, where the authors provide an account of the influence or, as they see it,
‘seduction’ of UK DNA database developments on Norway.
107
2009/C 296/01: See para. 4.74.
39
‘...any State claiming a pioneer role in the development of new
technologies bears special responsibility for striking the right balance
between the use of modern scientific techniques in the criminal-justice
system and important private-life interests.’108
Beyond these obvious promptings however, there was some acknowledgement
among UK participants that technological development had outstripped other
considerations.
4.4.
What emerged from the seminar was consistent with a key insight in the recent
Human Genetics Commission report, which noted that technological
development could make traditional police databases only a small part of a
much wider system – ranging from genetic research to passport databases containing intimate personal data available to investigators. This could possibly
change public attitudes to the forensic use of genetic collections established for
quite different purposes.109 This prompts an immediate question: whether a
debate hitherto framed almost exclusively in terms of the forfeiture of privacy in
respect of forensic bioinformation might become irrelevant unless it also takes
account of the consensual or, perhaps more accurately, quasi-consensual
surrender of genetic privacy outside the criminal justice sphere? It has been
recently found that people may not be uniformly happy to hand over their DNA
for alleged ‘health benefits’ (note the apparent disinterest in available genetic
testing kits). There was also great opposition to the sharing of data included in
the original Coroners and Justice Bill 2009, and controversy over the testing
and storage of babies’ blood spots without consent in the US and Ireland.110
‘Dissolving boundaries’ and ‘Securing the UK Border’
4.5.
The Human Genetics Commission have noted the growing pressures for
sharing information:
‘... there are ... substantial pressures to create the conditions for the
horizontal integration of criminal justice databases, and the sharing of
information between national jurisdictions. Whereas the exercise of law
enforcement powers of European Union (EU) member states are confined
within national borders, for the transfer of information relevant to law
enforcement these borders are increasingly dissolving. We identify three
developments in the ‘third pillar’ of EU policy (police and judicial cooperation in criminal matters) that have contributed to the cross-border
linking of forensic bioinformation: the establishment of a system of cooperation based on a central information system making use of electronic
databases; the creation of a presumption and then an obligation in favour
108
S & Marper v the United Kingdom, para.111 and 112
op.cit. n.105 especially pp.79-80, 83, and 86-87
110
see www.genewatch.org.uk for further details.
109
40
of data sharing; and the commitment of the UK and other member states
to align with such a system’.111
The ‘dissolving boundaries’ paradigm has been reflected in numerous
government statements. These express anxiety about the increasing
seriousness and scale of transnational crime and indicate intent to counter it by
improving the machinery of cooperation.
4.6.
Much of the evidence given to the House of Lords EU Committee by the Home
Office is informed by this paradigm. The Government explained how the Prüm
Treaty would ‘speed up and improve the quality and quantity of information
exchanged’ about DNA, fingerprints and vehicle registration in order to identify
and bring to justice terrorists and criminals. Ministers informed Parliament that
the Government had negotiated hard ‘to get an outcome that enables [them] to
sign up to Prüm and get all the benefits in terms of fighting cross-border crime
and counter terrorism where so much depends on good data exchange and
intelligence led policing’.112 In turn, parliamentarians and ethicists emphasised
that ‘privacy-related issues concerning the use and transfer of DNA and other
data for inter-jurisdictional criminal matters must be considered and agreed in
parallel with arrangements for availability, exchange and linkage’.113
4.7.
There is also a paradox: that initiatives to harden rather than dissolve national
borders may prove to be more significant harbingers of the potential future
trajectory for the forensic use of bioinformation than the activities touched upon
in the Marper judgment. If so, this has major implications for the discourse of
rights and privacy. The ethical and legal debate about forensic bioinformation
has been conducted largely with reference to collections of genetic material and
information gathered as a result of forfeiture. It may also need to take account
of what is happening as a result of parallel collections created with the consent
or quasi-consent of citizens.
International Cooperation
4.8.
A study published by the University of Leiden and the Netherlands Forensic
Institute in 2000 reported that there was little systematically compiled or
published information about international cooperation.114 Half a decade later the
situation had not changed, but what could be gleaned did not indicate that a
large amount was being exchanged. A report on the sharing of DNA
111
op.cit. n.105 p.84.
House of Lords European Union Committee (2007)18th Report HL Paper 90, Session 2006-07 (London,
TSO), Ev.1-2 and Q8.
113
Nuffield Council on Bioethics (2007), The Forensic use of Bioinformation: Ethical Issues, p.105.
114
Joubert, C. and Bevers, H. ‘International Exchange of Police Data and Forensic Expertise’ in Nijboer, J.F.
and Sprangers, W.J.J. M. (eds), Harmonisation in Forensic Expertise: An inquiry into the desirability of and
opportunities for international standards, Amsterdam: Thelma Thesis 2000, p.485).
112
41
information by a DNA Expert Working Group of the European Network of
Forensic Science Institutes (ENFSI) reported:
‘During 2004 – 2005, the UK received only 149 DNA search requests.
These generated 21 potential matches although the majority of the
profiles were unreliable; this was due to poor discriminating power. The
USA had a similar experience – they received even fewer requests for
assistance even though they have a larger database with over 4 million
profiles.’115
4.9.
The reasons for this, in the opinion of the ENSFI experts, reflected
inadequacies in current international law enforcement arrangements as well as
different technical processes for deriving a DNA profile:
‘DNA has proved to be a useful method of identifying criminals and
bringing them to justice. However, there has been slow progress in
exchanging DNA information routinely between countries. There are
various reasons for the slow progress including the differences in Member
States’ domestic law and data protection arrangements together with the
technical complexities around DNA analysis and exchange... Although
there have been many successful developments and initiatives to help
Member States co-operate to exchange DNA information, the current
arrangements are inadequate in their present form. They are incapable of
development on a scale that will meet future requirements and they are
not always sufficiently transparent or capable of being audited.’116
4.10.
Summary statistics about the volume of international cooperation initially
appeared and have since ceased to be published in NDNAD annual reports.
(Compare, for instance, the reports for 2004-2005 and 2005-06 where there are
entries dealing with the International DNA Database and, in the former,
statistics for international search requests, with the absence of such information
in the reports for 2006-2007 onwards.) Last year the authors of another ENSFI
report referred to the present ‘occasional exchange of DNA profiles between
countries’.117 That appears to be an accurate description of the present volume
of international forensic cooperation involving DNA.
4.11.
This could change – within the European Union at least – when the Prüm
Treaty takes full effect. The best and most recent indicator of the potential scale
of cross-jurisdictional cooperation was revealed in January 2007. The Prüm
115
DNA Expert Working Group of the European Network of Forensic Science Institutes (ENFSI) (2007)
Search Request Network Study: Final Report (accessed on 15 April 2007) from the previous ENSFI website:
www.enfsi.org/ewg/dnawg/documents/Annex20F20AGISFinal20Report .
116
supra pp 5, & 2 (accessed on 15 April 2007).
117
ENSFI (2009) DNA-Database Management Review and recommendations at
http://www.enfsi.eu/get_doc.php?uid=345, accessed 06.03.10. p.8.
42
Treaty had been in force for the purpose of trials of the new arrangements
between Germany and Austria for less than two months, but:
‘…the automatic information exchange has brought about noticeable
operational success: for instance, the German authorities matched DNA
profiles of open cases against data held by the Austrian authorities and
found hits in 1,510 cases. In this context 710 open traces from Germany
could be attributed to persons known to the Austrian criminal prosecution
authorities. Broken down by types of crime, 41 hits in homicide or murder
cases, 885 hits in theft cases, 85 hits in robbery or extortion cases were
found….’.118
4.12.
Both advocates and critics of the Treaty recognise that these results would not
necessarily be repeated in the future. There would have been a large amount of
information about earlier serious crimes that was available for the first time to
the prosecuting authorities as a result of the Treaty coming into force.119
Nevertheless, this was a huge increase in the volume of cooperation involving
DNA. The potential of the Prüm Treaty for obtaining serious crime leads more
systematically and efficiently could prove to be as significant a breakthrough,
for progress in dealing with previously undetectable crimes, as the introduction
in England and Wales of cold case reviews of unsolved murders and rapes, or
the reversal of miscarriages of justice in the USA.
4.13.
The installation of the most recent Interpol AFIS system illustrates both how the
potential volume of cooperation can increase with improved automation and
suggests that this is not always fully exploited. The new AFIS system allows
1000 search requests to be carried out daily, compared with 160 previously.120
However, the database is reported to consist of some 100,000 ten prints and
3,000 latent prints, and the number of identifications made during 2009 was in
the order of 1,000.121 Despite initial appearances this appears to be a
significant level of cooperation for the exchange of fingerprints. Operating on a
sub-continental scale, but within a single unitary state it was reported in 2006
that the FBI IAFIS system had achieved 1,200 identifications with latent prints
after it went online in 1999.122
118
House of Lords European Union Committee, 18th Report of Session 2006–07, Prüm: an effective
weapon against terrorism and crime?, HL paper 90, (London: The Stationery Office Limited) pp. 15-16.
119
supra p.16.
120
Sagem Sécurité (2009) Press release: Interpol places its trust in Sagem Sécurité once again:
http://www.sagem-securite.com/references-263/criminal-justice-police-264/article/interpol-places-its-trust-insagem-securite-once-again, accessed 090310 .
121
Interpol Fact Sheet: Fingerprints http://www.interpol.int/Public/ICPO/FactSheets/FS03.pdf , accessed 9
March 2010.
122
Dror, I.E. and Mnookin, J.L., ‘The use of technology in human expert domains: challenges and risks
arising from the use of automated fingerprint identification systems in forensic science’, Law, probability and
Risk, Advance Access, published January 22, 2010, p.8. The FBI statistics are likely to reflect jurisdictional
responsibilities and like the Interpol data may significantly under represent the scale of cooperation, if much
of this is directly between lower level jurisdictions. It does, however, put the Interpol data in context.
43
4.14.
The relatively small scale or inherent difficulties of international cooperation
involving DNA and fingerprints can also be appreciated by contrasting this data
with the detection of criminals travelling with stolen or lost travel documents. At
the end of 2008 the Interpol Stolen and Lost Travel Documents (SLTD)
database had 16.7 million records and was being used by 145 countries.
Several countries are connected directly to the database which can be made
interoperable with national border control data systems. In 2008 law
enforcement and border control officers carried out 25 million international
searches of the database. This resulted in the identification of over 5,000
individuals travelling on fraudulent documents.
4.15.
Throughout this project, great care was taken to keep the internationalisation of
crime in perspective. Most criminals operate locally. Only a small number of
successful or dangerous criminals regularly cross borders to evade detection.
Investigators present during our deliberations reminded us however, that some
criminals who may have recently established themselves as ‘local’ in the UK will
have originated from other countries where records may exist relating to a
previous criminal career. Participants with direct knowledge of case cooperation
considered that for a minority of crimes (probably many more than the figures
above) there need to be systems in place to easily access information from the
suspect’s home country or countries of passage. This conclusion was
reinforced by consideration of a travelling murderer and sex offender who had
evaded detection for his most serious offences despite being in custody twice
subsequently. When linked to his offences by pure chance, his profile was
reanalysed in the USA (because of the use of different multiplex), to be
compared with DNA recovered from the crime scene in France.
4.16.
Our discussions about international cooperation sought to identify the
circumstances that might merit the exchange of forensic bioinformation. Under
the Prüm Treaty, there is no restriction by offence before a DNA exchange can
take place: every crime or arrest could in theory result in a search of all EU
biometric databases. In practice, there appears to be growing recognition that a
more structured and prioritised approach is needed. Policies developed to
prioritise or otherwise restrict cooperation will only apply to anonymous profile
comparisons. The results of such exchanges will be provided to the judiciary or
the police. They will decide whether to undertake further enquiries. The cost of
investigation and the severity of the crime (and perhaps whether the suspect is
already in prison) are then likely to determine whether the anonymous match
should be followed up.
4.17.
Within the UK such decisions are not determined, as in Germany, by the law of
criminal procedure or, alternatively, subject to judicial or procuratorial
supervision as in the Dutch model. Investigators and analysts have the
discretion to seek potentially useful information or intelligence. Participants
44
stressed that not limiting cooperation to specified offences is more likely to
assist the investigation of serious crimes. Moreover, when a foreign national is
in custody suspected of a crime, it may be in the interests of justice to check
whether forensic bioinformation links this person to crimes abroad at the same
time as IDENT1 and the NDNAD are searched. Sometimes international
cooperation is useful only because organised transnational criminals might
lower their guard in their home country. The self-styled ‘Pink Panther Gang’
was cited as an example of this. The gang had undertaken a series of highly
skilled armed robberies in UK, Belgium, France, Switzerland, Dubai and Tokyo,
but were difficult to detect as members withdrew to their safe havens in Serbia
and Croatia. However, some gang members committed minor offences in their
home countries and because the local police shared their information through
Interpol it was possible to arrest thirty or so members of the gang.
4.18.
Limiting international cooperation to serious or organised crime may result in
failures to identify such successful criminals or their activities. For example, a
series of thefts from lorries in Belgium and the South Netherlands were
considered minor property crimes. Over time however, analysis revealed a
pattern of events of much greater significance. Indeed some investigators
consider that the most effective way of solving serious crimes is for the police to
allocate adequate time and resources to investigating volume crimes. Those
responsible for serious (and often sexual) offences will normally have a
background of burglaries or property crimes. Sometimes they might deliberately
commit minor crimes in one country to obtain money, and serious crimes,
including serious sexual assaults in another.
4.19.
This does not necessarily mean, irrespective of available resources, that all
crimes should be investigated or that all requests to cooperate with an
investigation should be accepted. An initiative to categorise the comparative
nature of criminal offences, in terms of the substantial offence would be an
appropriate early step in developing an international framework that might give
greater coherence to international cooperation, rather than by simply trading
caps in the volume of cooperation that each country will agree to process. For
example, many countries may be confused by the nature of UK terrorist and
national security related requests. Murder or homicide is a serious offence,
irrespective of known or suspected terrorist involvement or motivation, but the
effect of some of the recent declaratory ‘terrorist’ offences may have been to
misleadingly brand individuals guilty of trivial transgressions or, in the case of
protests associated with the invasion of Iraq, those who expressed dissent from
what is seen by many authorities on international law as unlawful action by the
Government, as threats to international security.
45
Bioinformation outside the criminal justice system
4.20.
A project visit to China, including the Hong Kong SAR (Special Administrative
Region), caused us to reconsider our earlier lack of engagement with the use of
bioinformation outside the traditional criminal justice sphere. There DNA is seen
as much more important for border control than criminal justice purposes. Since
1980, immigration from the Chinese mainland to Hong Kong has been subject
to a strict quota. The number was originally limited to 75 per day. It was
gradually increased to 105 in 1994, and then to the current quota of 150 in
1995. Most of these migrants are allowed to enter Hong Kong in order to be
reunited with their families.123 Entitlement to entry under this scheme is
established through DNA testing.124 During 2007, 34,000 (on average 93 each
day) mainlanders joined their families in Hong Kong under the One-way Permit
Scheme.125 In contrast the annual number of arrests in Hong Kong analysed
quinquennially between 1990 and 2005, ranged between 41,000 and a peak in
1995 of 53,000.126 The use of DNA subject profiles for border or migration
control purposes, given that bioinformation is likely to be relevant to the minority
of offences, will be more frequent than for criminal investigations.
4.21.
Great Britain is significantly different to Hong Kong in terms of the levels of
controlled immigration from outside the free movement area of the EEA plus
Switzerland and recorded crime: proportionately (i.e. per 100,000 residents),
40% lower and four and half times greater.127 However, biometric collections of
data for border and migration control purposes are likely to become as
significant in the UK in future years, as they are in Hong Kong.
4.22.
During the current forensic biometrics debates and inquiries, little attention has
been paid to the evolution of UK border control policies and practice. Several
recent developments – bundled together in a 2007 Home Office document
entitled ‘Securing the UK Border’128 – and subsumed subsequently within the
UKBA’s e-Borders programme, need to come within the mainstream of forensic
123
Asia Pacific Migration Research Network (APMRN) (undated) Migration Issues in the Asia Pacific: Issues
Paper From Hong Kong http://www.unesco.org/most/apmrnwp7.htm, accessed 1 February 2010.
124
Project members’ discussions at government laboratories in Hong Kong and Beijing on, respectively, 9
and 13 November 2009.
125
Government of Hong Kong (2007), Yearbook for 2007, page 406
http://www.yearbook.gov.hk/2007/en/pdf/E20.pdf, accessed 1 February 2010.
126
Broadhurst, R., King, W.L. and Ching, LC (2008), ‘Crime trends’, p.62 in (eds.) Chui, W.H. and Lo, WL.,
Understanding Criminal Justice in Hong Kong, Cullompton, Willan.
127
The UK and Hong Kong data appear to be, respectively, in the order of (controlled immigration) 287 in
twelve months ending 30 September 2009: 490 in 2007 and (arrests: note England and Wales only) some
2,700 in 2008: 590 in 2005. This is based on UK data for the number of persons (excluding EEA and Swiss
nationals) granted settlement during the 12 months ending 30 September 2009: 176,270 (Home Office
(2009) Control of Immigration Quarterly Statistical Summary, United Kingdom: July – September 2009,
http://www.homeoffice.gov.uk/rds/pdfs09/immiq309.pdf Home Office (2009) and (at the time of writing) the
latest data for arrests in England and Wales: 1,475,000 in 2008 (Ministry of Justice 2010) Criminal Statistics:
England and Wales 2008 http://www.justice.gov.uk/publications/docs/criminal-stats-2008.pdf, p.16 and both
accessed 1 February 2010.
128
Announced in a Home Office Press Notice dated 28.03.07: Government To Strengthen "Off-Shore"
Border, http://press.homeoffice.gov.uk/press-releases/off-shore-border.html, accessed 2 February 2010.
46
biometric scrutiny. Two of these will result in database retention and the
incorporation of a standard set of Home Office biometric data within an identity
or travel document. This data will consist of a set of ten prints and digital facial
image theoretically capable later of being used within automated facial
recognition systems. Enrolment within this parallel system will be as follows:
•
•
A compulsory identity card for foreign nationals (ICFNs) was introduced in
November 2008 for migrants from outside the European Economic Area
(EEA) and Switzerland. The card is being introduced gradually and is
currently issued to a wide range of migrants and their dependants when
they apply to extend their stay in the UK and will also be used for
checking identity at border crossings.129
In addition to checking the biometric data in ICFNs, from 30 November
2009, fingerprint checks were introduced at the UK border for all
passengers required to use biometric UK visas and entry clearance
documentation.130 As a result every visitor with a nationality to which visa
requirements apply are now checked against watchlists to identify failed
asylum seekers or anyone who has previously been removed from the
country.131
4.23.
UK citizens and foreign nationals with residency or entry rights have been
encouraged to enrol in a fast-track border control system named IRIS. Between
September 2008 and the 2006 launch of this voluntary iris scanning
programme, 217,000 people had enrolled. However, this scheme may not be
extended beyond the original four airports. In addition to technical failures and
difficulties in providing access for enrolment, there are doubts about whether
iris scans can cope with high volume usage. UKBA are reported to be looking at
other biometric options such as facial recognition and fingerprints, possibly
‘privately funded rather than funded by the UK government’.132
4.24.
Migration statistics are complex, based on inadequate data collection designed
originally for other purposes and subject to significant methodological debate.
As a result they have been described by the Statistics Commission as ‘weak’.133
However, data published by the House of Lords Economic Affairs Committee
suggests that up to approximately 5.5 million immigrants (some 9% of the UK
resident population) originated from outside the EEA travel zone. Most will be
129
UKBA website information at
http://www.bia.homeoffice.gov.uk/managingborders/idcardsforforeignnationals/, accessed 1 February 2010.
130
UKBA Press Release 25 November 2009 Introducing fingerprint checks at the border,
http://www.bia.homeoffice.gov.uk/sitecontent/newsarticles/2009/november/introducing-fingerprint-checks,
accessed 1 February 2010.
131
Airport-technology. com (2008) Body Check: interview with Julie Gillis, of the UK Border Agency, on
website, http://www.airport-technology.com/features/feature41603/ , accessed 2 February 2010.
132
supra.
133
House of Lords Select Committee on Economic Affairs (2008), 1st Report of Session 2007–08: The
Economic Impact of Immigration, Volume I: Report, p.10 (London: The Stationery Office).
47
enrolled within in these new biometric collections.134 The number of enrolments
will be increased by travellers whose citizenship brings them within the scope of
the biometric visa scheme. Even conservative estimates suggest this is likely to
add an additional two million people annually to border control biometric
databases.135 This will be dwarfed however, by the standard sets of Home
Office biometric data collected with the introduction of biometric passports and
identity cards for British citizens. These are issued separately by the Identity
and Passport Service to UK citizens. While the number of identity cards issued
can be counted (at the time of writing) in thousands, some 5.25 million new
passports were issued in 2008-2009.136
4.25.
The latest reported increase in the number of personal profiles held on NDNAD
amounted to some 580,000 in the year up to 31 March 2009, resulting in the
retention of data relating to an estimated 4.8 million people (NDNAD Annual
Report for 2007-09). Post-Marper retention rules will result in a reduction the
scale of NDNAD operations and in the volume of data held. Irrespective of the
impact of this change, it is inevitable that the size of the criminal justice
biometric databases will soon be exceeded by the collections of data biometric
collected by the Identity and Passport Service and UK Border Agency. For
almost half a decade the Home Office has anticipated running ‘a highly reliable
biometric enrolment and verification system with 40-50 million people enrolled’,
although it may not have been able to operate at this scale by 2006, as
originally planned.137
4.26.
Biometric data held by UKBA can be shared with the police, security services,
customs and other agencies, possibly private contractors, within the UK and
abroad. Such cooperation is governed by the Code of Practice on the
management of information shared by the Border and Immigration Agency, Her
Majesty’s Revenue and Customs and the Police issued in 2008.138 This code is
clear and comprehensive about the collection of data and the purposes to
which it can be used, with requirements for access control, audit, and penalties
if information is mishandled or misused. In contrast to MOPI (see para. 5.34), it
was brought into force as secondary legislation (Negative Resolution
Procedure). Its chief limitation is the fact that it only applies to UKBA, police and
Customs staff. The code does not appear to apply to carriers, contractors and
security services.
134
ibid. p.12.
For details of potential numbers see: Office for National Statistics (2009) Travel Trends 2008: Data and
commentary from the International Passenger Survey (Newport, Office for National Statistics), Table 2.09,
p.33 accessible at http://www.statistics.gov.uk/downloads/theme_transport/Travel_Trends_2008.pdf.
136
Identity and Passport Service (2009) Annual Report and Accounts 2008-09, HC 629 (London: The
Stationery Office), p.4.
137
House of Commons Science and Technology Committee (2006) Identity Card Technologies: Scientific
Advice, Risk and Evidence, Sixth Report of Session 2005–06 (London: The Stationery Office), Ev119.
138
Available at: www.bia.homeoffice.gov.uk . accessed 3 April 2010.
135
48
4.27.
An even greater omission in the governance of non-criminal justice
bioinformation is that there do not appear to be any published regulations
governing access to or the sharing of biometric information held or collected by
the IPS in connection with identity cards and passports. For example, there
appears to be no information in the public domain about whether or not, and if
so, under what circumstances, fingerprint data held in those later two sets of
documents can be speculatively searched against latent prints held on NAFIS
or vice versa. The different controls, or absence of controls, over what appear
at first sight to be separate collections of data is made worse by the fact that
they are held on the same system. Initially at least, a single database holds
both the visa biometrics of foreign visitors from outside the EU, and the identity
card and passport biometrics of UK citizens. This collection of data is
interoperable with the basic demographic, employment and tax related national
insurance records of virtually UK citizens held on the DWP’s Customer
Information System (CIS).139
4.28.
In devising its standard set of biometric data, the Home Office, in contrast to its
pioneering role in DNA, was following a pattern set by other countries, the
International Civil Aviation Organisation (ICAO) and the European Union:
•
Following 9/11, the United States introduced biometric (fingerprint) visas for
visitors;
•
In May 2003, ICAO adopted a blueprint for the integration of biometric
identification information into passports and other Machine Readable Travel
Documents (MRTDs).This nominated facial recognition as the primary
biometric with iris and fingerprint as backup;
•
In June 2003, a border control programme based on iris scanning was
rolled out throughout the United Arab Emirates;
•
In December 2004, following European Parliamentary approval, a new
Regulation on passports in the Schengen States was adopted, providing
that newly-issued passports must include digital facial images (within 18
months) and fingerprints (within three years).140
This does not indicate a lack of ambition in the Home Office. A 2006 report by
the House of Commons Science and Technology Committee noted that none of
the above initiatives were on the same scale or simultaneously encompassed
three types of biometrics. As noted however, in connection with the pioneering
use of forensic bioinformation, issues of governance sometimes appear to have
taken second place to ambitions to provide a technological lead to the world.
139
Home office (2008), National identity Scheme Delivery Plan 2008, p25 (London: Identity and Passport
Service).
140
op.cit. n.136, p. 9 and 11.
49
4.29.
The use of DNA for border control purposes, on the Hong Kong and China
model has made a relatively late and problematic appearance.141 The initial
foray by UKBA into the use of genetic techniques for determining asylum was
described in an editorial in Nature as ‘scientifically flawed, ethically dubious and
potentially damaging to science’.142 In September 2009, UKBA announced the
launch of the ‘Human Provenance Pilot Project’ as a joint project with SOCA
and funded by the FCO. It is intended to deal with concerns about possible
‘nationality-swapping’, and to challenge Home office entry decisions, as in Hong
Kong, one element of the pilot is the use of DNA to verify relationships between
adults and children. According to information provided by the Home Office to
the House of Commons Science and Technology Committee Select Committee
this is intended, as one project objective at least, to evaluate the use of DNA to
prevent child-trafficking, as within China. The main aim however, appears to
evaluate whether DNA analysis and isotope comparison can test country of
origin claims. For example, would it enable UKBA to distinguish between a
Somali entitled to asylum status and someone from an adjoining East African
country who is not?
4.30.
UKBA indicate an intention to consult the Home Office Forensic Science
Regulator about ethical approval and scientific validity:
‘At the conclusion of this pilot we will review the results, including the
underpinning science and the ethical implications of the work. The
Forensic Regulator will also be consulted during the period of the 3 month
pilot. Only if the evaluation and regulatory review is positive, will UKBA
proceed to use the results of future tests to support the decision making
process in specific cases.’143
Participation was said to be voluntary. But the first language of participants is
unlikely to be English and given their vulnerable nature, informed consent
would be difficult to establish. UKBA caseworkers were initially at least,
encouraged to ‘draw a negative inference as to the applicant’s credibility’ when
making asylum decisions, if the asylum seeker refused to provide samples for
the trial. In the face of ethical and scientific concerns the status of the initiative
changed, becoming a proof-of-concept project.
4.31.
The project raised ethical and sociological questions about whether nationality
and family relationships are conferred by law, e.g. by official birth registration,
141
For what follows the facts are drawn chiefly from the Nature editorial; an analysis by the HGC Secretariat:
Human Genetics Commission in (2009), Paper HGC09/P25; and The UK Border Agency’s Human
Provenance Pilot Project, available at (accessed 02 February 2010)
www.hgc.gov.uk/.../HGC09%20P25%20%20Human%20Provenance%20Pilot%20Project.doc and Brice, P
(2009) A PHG Foundation news report dated 30 September 2009: DNA testing to assess nationality of UK
asylum seekers, http://www.phgfoundation.org/news/4855/ accessed 03 February 2010.
142
‘Genetics without borders’, Nature, vol. 461, Issue no. 7265.
143
The Government Office for Science (undated) from information provided by the Home Office for a House
of Commons Science and Technology Select Committee: Evidence Check (accessed 3 February 2010) p12.
50
or might be established through family practice, or alternatively, must be
validated biologically. The scientific methodologies – most of which have been
developed for the study of populations not individuals - are equally problematic:
•
Little research has been carried out on DNA markers and populations in
Africa. As a result comparison databases will be limited and less able to
provide an accurate picture of ancestry and certainly not on a local scale;
•
Y-chromosome analysis can be thrown off by a distant male ancestor;
•
SNP (single nucleotide polymorphisms) -based identifications can be
problematic, individuals whose parents come from two geographic regions
are often classed into a third region from which neither parent originated;
•
The proposed hair and nail isotopic signatures that were to be used can
probably only give an indication of place of residency for the past few years;
•
The Adam’s torso case, which appears to have been presented by UKBA
as proof that UKBA proposed to use an “internationally recognised”
comparison method was performed by bone analysis. Moreover, the
methodology in that case has not been explained in scientific publications
or examined in court for the purposes of establishing nationality.
Nature reported that UKBA claimed that the project had undergone peer review,
but that the agency did not provide details. Several geneticists contacted by the
journal had seen a preliminary proposal in 2007 and warned that it was unlikely
to work.
The Significance of Scotland
4.32.
Hepple has commented that the Marper decision strongly hints ‘that uniformity
with Scotland would be acceptable as being within ‘the margin of appreciation’
open to the UK Government’.144 Comparisons with Scotland were made during
the seminars and international symposium. It became increasingly apparent
during these discussions that there was little recent research on the use of DNA
in the two jurisdictions. In particular we noted an absence of efforts to place the
difference in law within the context of how different arrangements emerged, the
state of available evidence about the operation of the Scottish system and
broader criminological and other factors that are important when making crossjurisdictional comparisons.
4.33.
Williams and Johnson recognised in 2004 that ‘the kinds of operational issues
that will arise in the future for a number of police forces wishing to share
intelligence information across the EU are already visibly prefigured in the
current arrangements that exist to make possible the linkage of collections of
144
Hepple, (2009) ‘Forensic databases: implications of the cases of S and Marper’ Medicine, Science and
the Law, v. 49 no.2 p. 82.
51
forensic DNA profiles across jurisdictions within the United Kingdom’.145 Such a
comparison remains the only detailed source of information about crossjurisdictional biometric cooperation. With the passage of time it is possible to
provide a greater degree of analysis as a result of the wealth of information,
albeit not always consistent, that can be gleaned from NDNAD reports.
The legal position in Scotland
4.34.
Under Scots law a subject DNA sample can be taken upon arrest for an
imprisonable offence and speculatively searched against both the Scottish DNA
database and NDNAD. However, profiles must be destroyed if the individual is
not convicted or is granted an absolute discharge. Since the Police, Public
Order and Criminal Justice (Scotland) Act 2006, an arrest for certain sexual or
violent offences may permit the retention of the sample and profile for three
years from the date of arrest, even if the arrestee is not convicted. Upon
application by the chief officer to a sheriff, an extension of up to two years may
be authorized, where it has been demonstrated that there are reasonable
grounds for retention. The Scottish arrangements also anticipated the Crime
and Security Act 2010 with regard to volunteer samples. An individual may
withdraw consent for the retention of the sample and related profile at any time.
Data sharing between jurisdictions146
4.35.
The scale of personal data sharing and the annual volume of data transactions
across the jurisdictional boundary are summarised in Table 4.1.147 All subject
profiles taken in Scotland are uploaded to both the Scottish database and
NDNAD. However, a profile from a sample recovered from a Scottish crime
scene is only uploaded to the NDNAD if it did not result in a match on the
Scottish database.148 This is an effective and efficient arrangement for
administrators and the criminal justice systems on both sides of the border. As
such it is a potentially useful model for crime scene profile cooperation within a
devolved or federal structure, or internationally. The results obtained from such
interoperability are summarised in Table 4.2.149
145
Johnson, P. and Williams, R. ‘DNA and Crime Investigation: Scotland and the ‘UK National DNA
Database’ Scottish Journal of Criminal Justice Studies 10: 71-84.
146
Note to the tables in this chapter: 1.NDNAD Data has been taken from the relevant NDNAD report but
has not been adjusted to reflect the practice by which, from time to time, historical data has been revised in
subsequent NDNAD Reports. An example of this is the ‘slight difference’ in data about the annual uploading
of profiles at 31/03/09. Such changes are judged to be less significant than maintaining internal consistency
for all figures published in a specified year, not all of which have been subsequently republished in a
corrected form in line with other changes in the data. 2. Where the data cannot be read directly in the report
this is indicated by the addition of ‘(est.)’. All estimates have been calculated from other data in the report.
147
Sources: NDNAD Reports, for 03/04 pp. 10, 18 and 22, for 04/05 pp. 6,10-11 and16, for 05/06mpp.24, 29
and 31-32, for 06/07 pp. 12, 19, 21, 24 and 26 and for 07/09 pp. 10-15.
148
NDNAD Report for 2003/04 p. 22.
149
Sources: NDNAD Report for 05/06 pp.12-13 and 35, for 06/07 p.31 and for 07/09 p. 28 and Scottish
Government (2009) Prison Statistics Scotland 2008-09 and Beyond, p.16
52
Table 4.1: DNA database interoperability between Scotland and NDNAD: subject profile
(sp) uploading and deletion. Note: ACPO sp retention policy was changed in December
2007. This change only affected the removal of English and Welsh sp from NDNAD.
Subject samples
2003/04
Loaded (all sp)
2004/05
2005/06
2006/07 2007/08 2008/09
521,117
715,239
722,464 591,028 580,174
Scottish sp Loaded
(Note: no volunteer
profiles)
42,000
(est.)
(7% of
total)
43,315
(7% of
total)
46,856
(6.4% of
total)
55,339
(7% of
total)
Removals (E&W sp)
NA
NA
384
(est.)
Removals (Scotland sp)
NA
NA
Removals (Scottish sp) as
% of loads (in same year
/previous year)
(est.)
Proportion of sp on
NDNAD from Scotland at
end of year (cumulative)
5.8%
5.5%
NA
NA
488
(est.)
162
283
21,748
23,439
19,211
16,562
46.4%/
50.2%
42.4%/
50%
NA/
34.7%
5.1%
4.8%
(est.)
4.7%
4.6%
Table 4.2: DNA database interoperability between Scotland, Northern Ireland and
NDNAD crime scene profile (csp) matches.
All NDNAD csp matches
with subject profiles
2005/06
2006/07
2007/08
2008/09
49,247
44,244
40,406
40,687
Excluding matches with csp 48,402 (est.)
from Scotland and NI
(only Scottish
csp reported)
43,332 (est.) 39, 575 (est.) 39,697 (est.)
(only Scottish
csp reported)
NDNAD csp matches with
subject profiles from
Scotland (as % of all)
1,388
(2.9%)
1,330
(3.1%)
1,119
(2.8%)
1,357
(3.4%)
NDNAD csp matches with
subject profiles from NI (as
% of all)
Scottish database csp
matches plus NDNAD csp
matches relating csp from
Scotland
NDNAD csp matches
relating to csp from Scotland
(as % of data in serial
above)
Proportion of non-Scottish
residents in prison custody
in Scotland
134
(0.3%)
36
(0.1%)
66
(0.2%)
131
(0.3%)
3,971(est.)
4,653 (est.)
N/A
N/A
845
(21.3%)
912
(19.6%)
787
880
3.6%
(June 2008)
53
4.36.
Some weight has been placed upon an observation in the Nuffield Council on
Bioethics’ report:
‘The match rates between stored subject profiles and new crime scene
profiles, which is 52%, can be contrasted with that of the Scottish DNA
Database, which has a higher match rate of 68%. This demonstrates
clearly that the more limited retention policy in Scotland does not
negatively impact upon its subsequent match rates.’150
It is becoming apparent from data published since the Report’s publication that
the comparative utility of the different retention regimes cannot be as clearly
demonstrated as the NCOB believed at the time of its report.
4.37.
Irrespective of changes in the data, such comparisons need to be treated with
caution for two reasons. First, the two sets of match rates may be distorted by
the interoperability of the two databases, how the match rates are calculated,
and technical issues that may affect the generation and reporting of matches.
Second, they are also likely to be affected by significant criminological or
criminal justice organisational variations between the two jurisdictions.
4.38.
The interoperability of the two databases may reduce the NDNAD match rate.
As indicated above, all subject profiles typed in Scotland are uploaded to the
Scottish database and NDNAD. However, a Scottish crime scene will only be
uploaded to the NDNAD if it does not result in a match first on the Scottish
database. As a consequence, a proportion of NDNAD subject profiles (the
extent to which the proportion of Scottish subject profiles on NDNAD has
changed little in recent years is indicated at Table 4.1) are unlikely to be
matched against crime scenes. This view might not hold true if Scottish
criminals were unusually mobile. This possibility cannot be discounted, but sixty
years of research in the UK, USA and Australia indicates that burglary in
particular - the volume offence that generates a high proportion of DNA crime
scene profiles - tends to be committed within a small radius of the offenders
home.151 Hence the annual NDNAD match rates may need to be discounted by
a proportion of the 4.6% of its subject profiles currently from Scotland and
possibly also those from Northern Ireland (1.2%).152 Comparisons may also be
affected by technical issues in one jurisdiction. The failure by the FSS over an
extensive period to load some 26,200 crime scene samples to the NDNAD
delayed 1,168 match reports.
4.39.
In any case, there is increasing convergence between NDNAD and Scottish
database match rates. The latter appears to have declined slightly in recent
150
op.cit. n.113, para. 4.52.
Wiles. P., & A. Costello, ‘The Road to Nowhere: The Evidence for Travelling Criminals’, HO Research
Study no.207, (Sept. 2000).
152
NDNAD Annual Report for 2007-09 p.15.
151
54
years from 68% in 2005/06153 to 65% in 2008-09,154 whereas the NDNAD
match rate has steadily increased and averaged 58.7% in 2008/09.155 It would
be difficult to draw any reliable conclusions from this. The analysis of the
differences between the operation of the two criminal justice systems and
variations in criminological context would be a complex task. This justifies a
note of caution about attempts to draw conclusions between the utility of the
two retention regimes. For example, Scottish recorded burglary rates (per
100,000) have consistently been approximately half those in England and
Wales.156 As a result there will be significantly fewer opportunities (on a
proportionate basis) to utilise DNA.
4.40.
That Scotland has some 30% more police officers (per 100,000) and that the
number of offences ‘cleared up’ per police officer in Scotland is four times
157
greater than England and Wales indicates fundamental differences in the
patterns of recorded crime and the resources available within the criminal
justice systems as a whole. As Table 4.3 (overleaf) indicates, in addition to
significantly better ‘clear up’ rates, there appears to be much less reliance on
the DNA database within the Scottish criminal justice system than in England
and Wales in terms of the overall level of activity (as measured by recorded
crimes and detections) in proportion to the number of profiles uploaded.
4.41.
Despite the methodological challenges and the difficulties of obtaining data, a
comparative and detailed study of the use of DNA in England and Wales, and
Scotland should yield more information about cross-border cooperation,
providing valuable insights into the consequences of different sampling and
retention policies. The most promising area for research initially might be a
fairly modest exercise. As can be seen from Table 4.2, Scottish crime scene
profiles uploaded to NDNAD have a modest impact on NDNAD matches. If,
however, these NDNAD matches are counted within the overall number of
matches arising in relation to Scottish crime scenes (i.e. by adding them to
Scottish database matches), it is notable that subject profiles from NDNAD
contribute some twenty percent of Scottish matches. Research on a case by
case basis into the circumstances relating to the latter set of data might improve
our understanding of the effects of the different retention regimes and other
issues. The disparity of impact on this scale is certainly unlikely to be caused by
a disproportionate number of residents from England and Wales committing
crimes in Scotland for which DNA is used for the purpose of detection. It is
much more likely to be influenced by different sampling and retention policies.
153
NDNAD Annual Report for 2005-06 p.24.
SPSA Annual Report and accounts 2008-09 p.21.
155
NDNAD Annual Report for 2007-09, p.33.
156
European Source Book 2006, Table 1.2.1.11.
157
Approximately 319 per 100,000 resident population in Scotland and 261 in England and Wales:
calculated (as FTEs) from The Scottish Government (2009) Police Officer Quarterly Strength Statistics
Scotland, 31 December 2009, using the figure for December 2008, ACPO Police service strength in England
and Wales at March 2008 (data at March 2008) and ONS 2008 mid-year population estimates.
154
55
Table 4.3: Recorded crime and subject sample uploads, 2004/05-06/07.158
2004/05 (2004 for
English stats.)
thousands
2005/06 (2005 for
English stats.)
thousands
2006/07 (2006 for
English stats.)
thousands
England & Wales
Offences recorded
5,641
5,557
5,428
Offences ‘detected’
1,441
1,516
1,374
Subject profiles (sp)
loaded
Ratio of offences /
offences detected to
sp loaded
447
668
629
11.8/3
8.3/2.3
8.6/2.2
1,077
1,018
1,026
Offences and crimes
cleared up
Sp loaded
791
749
765
43
47
55
Ratio of offences and
crimes / offences and
crimes ‘cleared up to
subject profiles
loaded
24.9/18.3
21.8/16
18.5/13.8
Scotland
Offences and crimes
The interplay of policy & technological development
4.42.
Johnson and Williams’ review of the use of DNA in Scotland highlighted
differences in the culture of the Scottish system and views about the utility of
DNA databases to those in England and Wales. They stressed how important
distinctions were rooted in a Scottish Law Commission report in 1989 and the
Royal Commission on Criminal Justice in England & Wales report in 1993:
‘Both commissions focused on the essential need to allow the police to
use genetic technology in a ‘balanced and proportionate’ way that would
ensure the protection of individual rights and civil liberties as well as
maximize the potential for criminal detection. Both commissions also
placed great emphasis on the potential for DNA testing to exonerate
individuals during police investigations, coupled with the idea of DNA as a
definitive forensic method which could incorporate high statistical
probabilities of certainty (and thus a form of ‘objectivity’) into legal
proceedings. However, they differed in how this technology should be
translated into practices for policing.’
158
Sources: Ministry of Justice (2008) Criminal Statistics: England and Wales 2007, Figure 1.1, p.10,
National Statistics for Scotland (2009) Criminal Proceedings in Scottish Courts, 2007/08, Chart 2, p.7 and
NDNAD reports for 2004/05, pp.10-11 and 22, for 2005/06 pp.25, 29 and 32, for pp12, 19 and 26.
56
4.43.
The Scottish Commission, when considering the balance needed to enable
DNA to work effectively, and the rights of the individual, proposed that the
power to take samples without consent ‘should not include anything which
involves going inside a person’s body’. This recommendation was not reflected
in legislation. The 1995 Criminal Procedure (Scotland) Act allowed for the
taking of a mouth swab without consent, reflecting the approach taken in the
Criminal Justice and Public Order Act 1994. The only distinction was that this
had to be authorised by an officer with a rank no lower than inspector. The
2003 Criminal Justice (Scotland) Act removed this requirement.159
4.44.
Irrespective of the change in the law, ACPOS policy still places considerable
stress on a police constable’s individual judgment. It advises against the
universal sampling of arrestees. Instead this should only happen where a
person has been arrested or detained for crimes of violence, sexual offences or
theft or, at the arresting officer’s discretion, it is deemed ‘appropriate’.160 This
could be explained, in part at least, by the minimal central government support
for expanding the use of DNA, but there may equally be something in Scottish
policing culture or a different perception of the value of the utility of the DNA
database. Certainly some forces were reluctant to make use of the same
legislation to retain samples relating to individuals who had not been convicted.
When reviewing Scottish DNA retention policy, Fraser found that by December
2007, five of the eight Scottish police forces had still not made use of the
extended power of retention that had come into effect.161
4.45.
The technological development issues are easier to discern and it is possible to
see how they might have contributed to a different approach. The deliberate
restriction of sampling on arrest reflected a similar policy in England and Wales
that prevailed while the FSS had insufficient capacity to undertake such
work.162 This approach was only reversed by the then Prime Minister’s
announcement of the DNA Expansion Programme, the aim of which was to
have on record the DNA profile of ‘every known offender’. This new policy,
together with the ring-fenced central government funding, meant the volume of
subject profiles uploaded to NDNAD trebled during 2000–2005.163 The history
of the use of DNA in Scotland shows a similar pattern of initial capacity
constraint and subsequently greater encouragement to obtain DNA on arrest
once analytical facilities had been automated and capacity had been brought
back into balance with potential demand.
159
Johnson, P. and Williams, R. ‘DNA and Crime Investigation: Scotland and the ‘UK National DNA
Database’Scottish Journal of Criminal Justice Studies 10: 71-84
160
NDNAD Report 2004/05 p.8.
161
Fraser, J. (2008) Acquisition and Retention of DNA and Fingerprint Data in Scotland p.14 as published in
September 2008 at www.scotland.gov.uk/Consultations/Current .
162
Blakey, D. (2000) ‘Under the Microscope’, (London: Home Office), p.23.
163
Home Office (2006) DNA Expansion programme 2000-2005: Reporting Achievement (London: Home
Office), p.3-4.
57
4.46.
The introduction of DNA sampling had been delayed north of the border so that
by the time Scottish forces were submitting samples to the FSS for analysis
there were already major backlogs. This meant that the proposed processing
times offered to Scottish forces was unacceptable. Only then was it decided to
create a separate Scottish database populated by profiles from Scottish police
laboratories.164 Thereafter investment in automation was much slower though
its introduction was soon followed by expressions of concern about the level of
‘forensic capture’. The SPSA reporting that:
‘A marketing campaign has been introduced to increase awareness and
to promote the capture of criminal justice samples and fingerprints by
police officers. Posters have been designed and circulated via ACPOS to
custody suites across Scotland promoting the message of forensic
capture to ensure our DNA and Fingerprint databases are as up to date
as possible in order to provide more opportunity for successful hits and
identifications.’165
4.47.
The sampling and retention of child profiles also provides an insight into
different legal, cultural factors. Although few children are involved and the ages
of criminal responsibility are different, data suggest that the Scottish legal
system has been more willing to authorise the collection of bioinformation
relating to young people less than ten years of age than in England and
Wales.166 Legislation going through the Scottish parliament will expand
retention to children who admit guilt under the Children’s Hearing System, but
the age of criminal responsibility is to rise from 8 to 12 years.
4.48.
There is greater consideration of the Scottish position in Chapter 6, but
conclusions from this brief consideration of the Scottish arrangements include
the interoperability of NDNAD and the Scottish database, together with the
ability of the forensic services of the two administrations and Scottish criminal
justice colleagues to manage different retention regimes on a large scale as
indicated in Table 4.1, is clear evidence that forensic cross-jurisdictional
cooperation can be successfully managed on a large scale despite legal
differences. While this report identifies some gaps and lack of consistency in
the data and suggests where research is urgently needed, the scale and detail
of the interoperability analysis indicates that NDNAD has set an example of
greater openness nationally and internationally for criminal justice databases,
including the exchange of information about fingerprints, and possibly genetic
collections in the genetic research and health fields.
164
NDNAD Annual Report 2004/05 p.6
SPSA Annual Report and accounts 2008-09 p.21.
166
In 2008/09 the number of profiles for children under ten on NDNAD was 96 for England and Wales and
33 for Scotland, which if recalculated as per 100,000 population corresponds to 0.18 and 0.63 respectively.
NDNAD Annual Report 2008/09.
165
58
4.49.
Finally there is some evidence to support the contention made in this report that
a margin of appreciation based on no more than a survey of legislation is a
possibly more fragile basis for reaching views on proportionality than is
suggested in Marper. Scotland may have achieved a very balanced approach
to retention issues that still need to be settled south of the border. This appears
to have been as much a chance outcome, as one of good professional and
political judgment. It is far from clear however, from reading recent debates at
Westminster that what we would argue is the defining characteristic of Scots
law in this area – the exercise of individual discretion - has been fully
understood.
Validity, technological development and DNA sample retention
4.50.
Discussion about validity during the seminars caused us to focus on three
issues in particular:
•
•
•
False or adventitious matches, false eliminations and sample retention;
The challenge of ensuring the validity of dactyloscopic information;
The foreseeable technological obsolescence of the SGM+ multiplex.
The risk of these problems materialising is greatest when information is
exchanged between foreign jurisdictions, but is present when poor material has
been recovered from a crime scene and when using old records created with
either a lower discriminating power or to poorer technical standards.
False or adventitious matches, false eliminations and sample retention
4.51.
Different modes of individualisation based on bioinformation give rise to
different scientific and professional challenges, but common to all is the
recognition that mistakes have unacceptable consequences for individuals and
confidence in the criminal justice system. These issues are explored more
easily by an examination of the forensic use of DNA because of the greater
availability of information, but examples and specific points relating to
fingerprints are also included where possible.
4.52.
Within common law jurisdictions, debate about the validity of DNA matches has
focused on adventitious and false matches.167 An adventitious match is when
the DNA profiles of two individuals, who are not identical twins, match. This is
theoretically quite a rare occurrence within a DNA typing or multiplex that has a
sufficient discriminating power. The risk is minimised chiefly by using a marker
or typing system with an adequate number of loci for the size of the population
of the database or network of databases from which a match is being sought.
167
See especially Lynch, M. Cole, S.A., McNally, R. and Jordan, K. (2008) Truth Machine: The Contentious
History of DNA Fingerprinting, Chicago and London: Chicago University Press.
59
With the UK SGM+ multiplex (10 loci plus Amelogenin) the chance of an
adventitious match (for people who are not siblings) is reported to be 1 in 1,000
million.168 In this report another term is used in respect of both DNA and
fingerprints: a false match. This describes a more commonplace situation when
there initially appears to be a match, but closer examination shows that result to
be invalid. A false DNA match is likely to be the result of one or more of the
following potential sources of error:
•
•
•
Insufficient loci available for analysis from degraded DNA material
recovered from the crime scene resulting in an inability to take advantage
of the maximum discriminating power of the marker or typing system;
A misattribution of the source of a loci reading to a specific individual when
cellular material recovered from a crime scene belongs to more than one
person (a ‘mixed sample’);
An attempt to match an incomplete set of loci by using only the small
number of loci it has in common with different marker or typing systems
(e.g. during international exchanges or when samples generated by using
the current multiplex system are compared with profiles produced by
previous multiplexes that are less discriminating).
The third complication is both a source of error in its own right and a factor that
can compound problems arising in respect of the other two. Genetically an
adventitious match is a reliable result within the scientific parameters of the
typing system, but a false match is methodologically unreliable. For the
purposes of individualisation within legal proceedings both are invalid.
4.53.
During seminars experts referred to the regular and beneficial exchange of
fingerprint data between Austria and Germany over many years. However, very
similar problems can occur in respect of fingerprint matches, particularly when
what has been recovered from the crime scene is incomplete, degraded by
decay or fire, distorted when the trace was made, or has been searched for by
more than one proprietary algorithm. Problems can also arise from the use of
poor tenprints obtained manually using ink and paper. Despite the extensive
use of modern technology in wealthier countries, most tenprint records in use
today were obtained manually and this century old method is still the main
method used for enrolments, within criminal justice systems at least, in most
countries of the world. Problems are compounded when the quality of an image
is reduced in transmission internationally between different fingerprint bureaux.
4.54.
It was suggested during seminars that equal attention should be paid to false
eliminations, which arise for the same reasons as a false match. Indeed a risk
is present whenever DNA analysis is used if there are only a small number of
loci in common during a search involving profiles obtained from a multiplex with
168
NDNAD Annual Report 2003-04, p 33.
60
a lower discriminatory power (within the UK, the SGM system), or during
international searches involving multiplexes with a limited number of loci in
common. It is most likely to occur when a crime scene sample is degraded and
a limited range of alleles can be amplified. Problems may also arise because of
variation in PCR when different typing systems are used. One of the kits used
may not have resulted in one of a pair of alleles being amplified to a detectable
level, but the other system may have amplified both alleles.
4.55.
The limited evidence available from US fingerprint proficiency testing suggests
that false eliminations may occur on a smaller scale. Reports of the proficiency
testing undertaken by the American Society of Crime Laboratory Directors
indicate that 8% of the latent prints deemed by those conducting the tests to be
inadequate for identification were not so marked. In contrast, 2% of the latent
prints deemed adequate were scored as unsuitable for individualisation.169
4.56.
While the problem of false identifications is generally acknowledged, the
consequences have not necessarily always been analysed. For example, the
Human Genetics Commission suggested that a crime scene from Lewisham is
more likely to be linked to a local resident, than a resident of Lisbon or
Lugano.170 Hence, the investigator will know that the inquiry should concentrate
on the candidate match with a local resident, rather than one identified via, say
by a six loci match (the European and Interpol standard set) elsewhere in
Europe. In reality this would indeed be the case with a full SGM+ match (crime
scene to personal profile stored on NDNAD). In the absence of such a clear
result, it may be inappropriate to close minds to a possible foreign connection.
Criminal investigation within Lewisham has to take account of the nature of its
resident population. ONS research indicates that by 2008 almost a third of
Lewisham residents were not UK born, although Lagos had much greater links
with that borough than Lisbon or Lugano.171
4.57.
During the seminars it was acknowledged, as the Human Genetics Commission
suggested, that investigators will often be able to eliminate adventitious or false
matches fairly easily because it will be apparent with minimum checking that
the individual concerned has nothing to do with the crime under investigation.
Sometimes however, the issues may not be easy to resolve. This can result in
considerable intrusion into the private life of an individual. What has not been
recognised is that such a situation would be the equivalent to the suspicion
initiated by a close non-match obtained by the use of familial searching. The
accidental and unknown nature of the linkage with the investigation however,
means that none of the safeguards deemed essential for familial searching will
169
Pyrek, K.M. (2007) Forensic Science Under Siege (Burlington MA: Elsevier Academic Press), p. 269.
Human Genetics Commission, (2009) ‘Nothing to Hide, Nothing to Fear’ London, Department of Health
pp70-71.
171
Ellis, A. Population Trends 135, Spring 2009: UK resident population by country of birth, at
http://www.statistics.gov.uk/articles/populationtrends/PT135POPCOBARTICLE.pdf accessed 7 March 2010.
170
61
be in place. In the worst case scenario, someone may be trapped in a
nightmare situation because of a combination of a false match or near match
and circumstances judged to corroborate an unreliable biometric result.
4.58.
Dactyloscopic cooperation resulted in the most well known case of this kind:
that of Brandon Mayfield. A US born lawyer and former Army lieutenant,
Mayfield received $2 million in compensation from the FBI for what he
described as the ‘horrific pain, torture and humiliation’ caused to himself and his
family as a result of a false fingerprint match.172 He was detained (initially in a
single cell for up to 22 hours a day) after the FBI claimed that his prints
matched those found on a bag of detonators in Madrid following the 2004 train
bombing. This result was challenged by the Spanish police. The FBI examiners
later acknowledged that they did not re-examine their initial match decision as
carefully as they should have done. They knew by that time that Mayfield was a
convert to Islam, had an Egyptian wife and had represented a convicted Islamic
terrorist in a child custody dispute. An official investigation largely exonerated
the FBI examiners, focusing on the problems of dealing with ‘confusingly similar
prints’ identified by an automated database search, as an ‘unusually close nonmatch’. It concluded however, that the examiners had been willing to ‘accept an
extraordinary set of coincidences’ in explaining their identification, and that
internal reporting arrangements within the FBI had resulted in ‘implicit pressure’
on the examiner to declare a positive identification ‘when faced with a difficult
comparison in a case involving a particularly heinous crime’.173
4.59.
The reverse situation – a false elimination - may be equally damaging for the
administration of justice. A scientific or professional inability to match
bioinformation from the crime scene with data held under a different system
may result in an investigation being discontinued or being focused on the wrong
suspect simply because of a vague circumstantial fit with the limited information
about the crime. From our analysis of the political ambitions for the greater
exchange of forensic bioinformation internationally, it is reasonable to assume
that in the absence of technological, organisational and professional
improvements, that this is likely to be a growing problem in the future. It may be
however, that organisational improvement could be achieved simply through
investment in information technology, including automatic database searching.
This could be a costly mistake. With the increasing horizontal integration of
criminal justice data, staff currently working on international cooperation could
be overwhelmed by data and, as was seen in the Mayfield case, in major
investigations face considerable pressure to confirm ‘near close matches’.
172
Lichtblau, E. New York Times, 30 November 2006, ‘US Will Pay $2 Million to Lawyer Wrongly Jailed’, at
http://www.nytimes.com/2006/11/30/us/30settle.html?_r=1, accessed 6 March 2010.
173
US Department of Justice (2006) Office of the Inspector General Oversight and Review Division, A
Review of the FBI’s Handling of the Brandon Mayfield Case, especially pp.7, 9,11-12 and 20-21, at
http://www.justice.gov/oig/special/s0601/PDF_list.htm accessed 6 March 2010.
62
4.60.
The potential widespread distribution of automated searching for matches to
latent print matches (under Prüm this could yield 10 closest matches from 26
countries), could quickly produce candidate match ‘overload’. Greater reliance
on automated matching however, could exclude a potential match by not
putting it on the list of, say, ten candidate matches or closest near matches.
This might incorrectly eliminate the offender from the inquiry. Similar problems
were anticipated in respect of DNA by the ENFSI DNA Working Group in 2009.
They concluded that while the European Standard Set of Loci are sufficient for
the present ‘occasional exchange of DNA-profiles between countries’, the much
higher volume of exchanges of DNA-profiles made possible by the Interpol
DNA-database and the Prüm Treaty means that the chance of adventitious or
false matches will no longer be negligible, and may become significant.174
4.61.
The ENSFI report also provided clear examples of the risk of invalid results
because of misplaced reliance on automation alone or inadequate software
within an automated searching system. Some DNA-databases, such as CODIS,
contain mixed DNA-profiles and can be searched internationally. This can be
very helpful when investigating serious crimes. A numerical match between a
reference sample and a mixed profile however, must always be visually
checked against the plots of the DNA-profile because a numerical match may
not be a real match as shown in Figure 4.1. In addition, when two profiles have
been generated by using different multiplexes there is a risk of missing a match
due to the occurrence of so-called ‘null-alleles’. This is the result of a sampling
variation between two multiplexes resulting in the failure by one to amplify an
allele. One system might incorrectly detect only one allele present in the sample
(apparent homozygote or two identical alleles at a given locus), but the second
may have detected two alleles (apparent heterozygote or two different alleles at
a locus). The presence of a null-allele is identified by observing the unexpected
low peak height on the plots of the DNA profile of the apparent homozygote, but
this requires an attentive DNA-analyst or intelligent allele-calling software.175
Figure 4.1: Example provided by the ENSFI DNA Working Group illustrating how three
loci of a mixed stain and a reference sample which match on a numerical basis can be
seen not to match when the mixed profile is attributed to a single individual contributor
174
ENSFI (2009) DNA-Database Management Review and recommendations at
http://www.enfsi.eu/get_doc.php?uid=345, accessed 6 March 2010, p.8.
175
ibid. pp.12-13 and 8.
63
4.62.
The only effective way of dealing presently with invalid results when using DNA
appears to be, as recommended by ENSFI, that of retaining cellular material for
the same period as the profiles. It is also important to ensure that there are
sufficient DNA and fingerprint analysts to scrutinise candidate matches in
conditions where they are not placed under direct investigatory pressures. This
is clearly not a satisfactory long-term solution and many organisations have
significant privacy concerns. There are also cost implications. An interim option
would be to bring the UK into line with best practice in North America, where
demographic information is not sent to any FBI or RCMP forensic laboratory
that analyses personal profiles. Cost reduction ultimately depends on
considerable technological progress (considered below). In the meantime,
samples are being retyped extensively when matches are identified between
Germany and the Netherlands during their trials of the Prüm arrangements.
This extra work adds significantly to cost and speed, making it less effective for
finding forensic leads. However, these considerations may not detract from the
possibly greater benefits from not unnecessarily sending demographic details to
laboratories, a practice that opens possibilities for abuse and security lapses,
and has been criticised by those running the laboratories themselves.
4.63.
Similar obstacles can be anticipated in the progress of UK cold case review
work as degraded crime scene material is reanalysed using a new multiplex. In
either instance the problems of ensuring acceptable sample storage and the
cost of retyping personal profiles might prove, on balance, to be acceptable.
They need to be compared with the burden of potential investigative or legal
work to avoid or respond to legal challenges during a trial if all candidate
matches had not been exhaustively investigated for elimination purposes. Also,
a judgment needs to be made about the potential impact of sample destruction.
Not being able to follow-up ‘near matches’, and the ability to reanalyse old
crime scene material, may negatively impact upon effectiveness as a result of
foreseeable technological advances.
4.64.
The analysis of these issues should take account of the existence of a proven
option for reducing sample storage costs. The present storage of samples at
-20 degrees Celsius could be replaced by FTA paper storage in ambient
conditions. Regrettably it can be seen from the Impact Assessment published to
support the Crime and Security Bill 2010 that the consideration given to this
issue by the Government did not identify ambient storage as an option:
‘It costs around £0.90/year to store a DNA sample in a fridge. Under the
option of destroying all DNA samples, this cost would not occur. Given
the large number of samples that would need to be removed we have
estimated a saving of around £5.2m (one-off) and £7.7m/year.’176
176
Home Office (2009), Impact Assessment of the implementation of the S & Marper ECtHR judgment
regarding DNA profiles, DNA samples and fingerprints retention published at
www.crimereduction.homeoffice.gov.uk/crimeandsecuritybill/
64
4.65.
The impact assessment acknowledged however, that the options for dealing
with this issue had only received limited analysis during the consultation stage:
‘While most respondents failed to differentiate between samples and the
profiles which are derived from them, where they did, there was strong
support from the public and civil liberties groups for the proposal to
destroy samples. However, some caution was expressed by policing and
prosecutorial organisations in terms of possible operational and evidential
implications’.177
The Government also admitted that re-sampling ‘may not be practical or
appropriate in a large number of cases’ should new techniques be developed to
extract more information from crime scene samples or assistance be requested
internationally. It accepted that investigations might fail both domestically and
abroad due to the inability to produce the original sample.
Ensuring the validity of dactyloscopic information
4.66.
As a forensic technique grounded in professional expertise rather than
replicable and independently verifiable methodology, dactyloscopic examination
faces major challenges in the light of the McKie case in Scotland,178 and the
Brandon Mayfield case in the USA. As a result of problems documented in the
NRC 2009 report ‘Strengthening Forensic Science in the United States: A Path
Forward ‘, there may be a new round of challenges to fingerprint evidence and
other experienced based forensic disciplines. It is suggested in that document
that there now needs to be a clearer understanding of the basis for valid
fingerprint matches, together with research into causes of error and error rates.
4.67.
The massive expansion in the exchange of such information for border control
purposes, as well as the less dramatic growth of cooperation in the course of
traditional criminal investigations, means that new fingerprint standards for
establishing the validity of matches need to be applicable globally. The ENFSI
fingerprint working group is seeking to establish what is being done in the
different EU countries. Their analysis will cover national identification methods,
the testing of procedures and the assessment of personal competency. The
ENFSI fingerprint group and the International Association of Identification (IAI)
have links, but it was suggested that both organisations need to achieve
greater visibility and wield more influence internationally.
177
supra.
The Scottish Fingerprint Inquiry was established with the following terms of reference:
♦
To inquire into the steps that were taken to identify and verify the fingerprints associated with, and
leading up to, the case of HM Advocate v. McKie in 1999, and
♦
To determine, in relation to the fingerprint designated Y7, the consequences of the steps taken, or not
taken, and
♦
To report findings of fact and make recommendations as to what measures might now be introduced,
beyond those that have already been introduced since 1999, to ensure that any shortcomings are avoided in
the future. For current developments see: http://www.thefingerprintinquiryscotland.org.uk/inquiry/21.html
178
65
4.68.
During seminar discussion, developing a probabilistic basis for fingerprint
identification or perhaps reaching agreement on a numeric standard were seen
as the two most promising ways forward. This would enable the fingerprint
community to follow the lead of ENSFI in introducing proficiency testing and
audits. Whatever approach is adopted, it would be necessary to ensure that the
transition to more highly complex and statistically demanding approach is made
intelligible to existing practitioners whose existing working methods and
professional expertise will undergo a major change. Dror and Mnookin have
considered more systematically than was possible in our discussions, the
increased risks of examiners making erroneous identifications because of the
scale and speed of automated database searching. This arises from the
increased chances that automated searching will produce an increased number
of candidate matches based on incidental similarities and that some of these
may be an ‘artifact both of the relative similarity of the patterns being compared
and of the human cognitive architecture involved in pattern matching’.179
4.69.
Technical problems relating to the transmission of data and the preparation or
presentation of information for analysis theoretically at least present fewer
problems. The transmission of good digital images is critical. It should be
possible to reach an international agreement and introduce changes quickly
subject to available funding. This could include the minimal pixel quality of
exchanged images and the replacement of ink and rollers with digital image
capture such as LIVESCAN which is used in UK custody suites. However, it
was noted that little progress had been made internationally other than a very
limited agreement dealing with the file format for transmitting information.
4.70.
Automated search systems generally decrease reliance on individual decisionmaking while reducing costs and speeding up database searches. While the
manual scrutiny of candidate matches remains essential when dealing with
latent marks, it is less important when comparing a LIVESCAN reading from
someone brought into custody with an equally good set of tenprints on NAFIS.
The introduction of 24/7 automation has made a huge difference in the ability of
the police to verify the identity of persons in custody despite most UK fingerprint
bureaux operating only five days a week.
4.71.
The downside to the history of technological development in this field has been
how this has been based on the ad hoc development of different proprietary
search and image capture systems. Often different systems, sometimes even
different versions of the same system from a single manufacturer, are not
interoperable. Similarly, different algorithms used to retrieve stored prints in
accordance with a specific input map of features may require fingerprint
179
Dror, I.E. and Mnookin, J.L., ‘The use of technology in human expert domains: challenges and risks
arising from the use of automated fingerprint identification systems in forensic science’, Law, Probability and
Risk, Advance Access, published January 22, 2010, p.21.
66
examiners to learn system specific ways of annotating features on a latent print
in order to maximise the chances of a successful identification. This situation
was compared unfavourably in the NRC report with the high levels of
interoperability achieved for other distributed information networks, including
banking systems, such as ATM machines. The successful development of such
IT systems requires ‘finely crafted and agreed standards and protocols’. This,
the report concluded, would require strong, high level support from government
policy makers, and additional public funding.180
4.72.
Seminar discussions also acknowledged that there may also be development
aid, IPR, and strategic procurement issues that need to be addressed for the
benefit of developing and middle income countries. Some have recently tried to
consider systematically the increased risks of examiners making erroneous
identifications because of the scale and speed of automated database
searching. This arises from the increased chances that automated searching
will produce an increased volume of candidate matches that are based on
incidental similarities and that some of these may be an ‘artefact both of the
relative similarity of the patterns being compared and of the human cognitive
architecture involved in pattern matching’.181
4.73.
It is unclear within the UK how strategic coordination has taken place, even in
terms of basic system design. The Home Office decided to use LIVESCAN
tenprint scanning for subject enrolment for identity cards and passports. This is
in contrast to enrolment on IDENT1, where the standard is based upon rolled
prints. Both choices probably made sense in isolation. LIVESCAN enrolment is
likely to be quicker and cheaper than rolled prints, but rolled prints was the
standard judged necessary for comparisons with latent prints within the criminal
justice system. The critical question is whether, if or when, the identity card and
passport tenprints are accessed for criminal justice purposes, they will be
suitable for this purpose? This may simply be a matter of ensuring a more
intensive scrutiny of potential matches or near close matches when prints
obtained outside the criminal justice system are compared with latent prints.
At a minimum, there appears to be a prime facie case for interdependent
assurance that in such a situation adequate safeguards are in place. One
aspect that may have to be capable of being monitored is that there are a
sufficient number of competent experts in place to scrutinise automatic search
results, so that such reports can be properly verified or caveats attached before
being passed to investigators.
180
National Research Council of the National Academies (2009) Strengthening Forensic Science in the
United States: A Path Forward, Washington DC: The National Academies Press, pp.269-278.
181
op.cit. n.179, p.21.
67
The foreseeable technological obsolescence of SGM+
4.74.
During seminar discussion there was general agreement that the variety and
differing configurations of multiplex introduced over a decade ago are barriers
to efficient and cost-effective use nationally or exchange internationally. The
systems used most frequently, especially in the UK and in countries from where
the majority of known serious offenders in this country originate, are set out
overleaf in Figure 4.2. Within the European Union this problem is to be dealt
with in the medium term by introducing five new markers. These loci (Next
Generation Multiplex - NGM) have been developed for two reasons. Firstly,
they provide more discriminating power as national databases grow in size and
are searched more frequently alongside other national databases. Secondly,
they have a low molecular weight and will be more successful when used to
analyse degraded samples. This multiplex was introduced at the end of 2009
and their use in all member states by November 2011 was recommended in a
Council Resolution on 30 November 2009.182
4.75.
The cost of switching will reflect not only higher analysis costs (extra reagents
and additional processing), but also software upgrades for the NDNAD and
other DNA (provider and Interpol) databases. We were told that within the EU it
was estimated the five new loci would ensure sufficient levels of discrimination
for another five years.
182
EU Council Resolution on the exchange of DNA analysis results (2009/C 296/01).
68
Figure 4.2: Analysis of loci used in DNA multiplex in the countries with the most
criminological links with the UK.
4.76.
For the UK the loci issue can be summarised as follows:
•
•
•
Its multiplex, SGM+, is already technologically obsolescent, but adding the
five NGM loci could considerably reduce the risk of false matches,
adventitious hits or false eliminations (domestically as the NDNAD grows
and internationally as the volume of data exchange increases). It would
also increase the potential value of reinvestigating older crimes, particularly
murders and serious sexual offences.
Even the next upgrade, by adding the five NGM loci, may only be of value
for about another five years.
Even if the UK remains in step with other EU member states the origin of
the most serious criminals known to have committed offences in this
country means that in many cases the mulitplex used in their home country
69
and the UK do not have sufficient common loci to be confident of the
quality of any exchanged information or evidence being reliable. This
problem is illustrated at Figure 4.3.
Figure 4.3: The correlation (proportion of loci in common) of SGM+ with the mulitplex
used (where known) by countries with 200 or more citizens in custody in England and
Wales at 31 December 2008.
4.77.
International data indicates that SGM+ is a significant multiplex globally and
within the EU. Investment in five new loci is probably a good medium term
option. The issue that needs to be grasped now is what should be available in
five years time? In attempting to forecast possible trends the much closer
convergence of China (the fastest growing DNA database) and the USA (the
world’s largest DNA database) is probably the most strategic public policy issue
relating to the technological development of DNA multiplexes that the
governments of the UK and EU, together with the Council and Commission
have to address. China originally used ‘Identifiler’ (equivalent to CODIS plus
Amelogenin), the same multiplex as Hong Kong, but some time ago a new kit
was developed with 15 loci: Type 15 or ‘Sinofiler’. This is identical to Identifiler
except for the replacement of two CODIS loci and the addition of two loci not
used in other systems. At present it is believed that half the profiling work in
China is carried out with Identifiler, but eventually all analyses will use Sinofiler.
4.78.
The problem with the development of new multiplex within Europe is the risk of
being significantly out of step in terms of access to the benefits of technological
development; more effective international law enforcement cooperation; and
commercially with the predominant systems used elsewhere in the world. Links
70
with medium income countries through the Commonwealth will make no
difference. South Africa is the key country within sub-Saharan Africa for
undertaking DNA analysis for other countries in its region. It uses Pro+ (seven
common SGM+ and Identifiler loci, together with an additional three Identifiler
loci). Jamaica may be the country of origin for the largest proportion of foreign
nationals (predominantly male) in English prisons, but the impact of Jamaican
crime (as indicated by the deportation of Jamaican citizens for firearms and
violent offences) is greater in the USA rather than the UK and Jamaica uses
Powerplex systems not SGM+.
4.79.
ENFSI works with the relevant US and Australasian bodies, SWGDAM
(Scientific Working Group on DNA Analysis Methods) and SMANZL (Senior
Managers Australian and New Zealand Forensic Laboratories), but it is quite
difficult to co-ordinate and collaborate effectively with no regular funding and, as
a voluntary organisation for forensic providers, cannot have a formal mandate
to seek to promote coordination with North America or China. It exists on the
basis of minimal public funding limited to travel expenses and small research
projects, and no political support. Hence, progress in developing new multiplex
had only been possible because of the goodwill of scientists and the advantage
for two biochemical companies of ENSFI cooperation in developing a new
product that might be marketed beyond Europe. There is some optimism that
other countries will consider adopting the 5 loci that have been developed by
ENFSI when expanding their core loci. There is however, a major technical
barrier for achieving any kind of multiplex that would provide a more reliable
overlap between the various CODIS based multiplex and the different systems
developed in Europe. The kind of 20 loci multiplex that might make this possible
would require new biochemistry, at present it is technologically impossible to go
beyond a 15 or 16 loci system.
4.80.
The absence of coordinated EU national or Commission strategic involvement
also results in over dependence on commercial biochemistry. The absence of
public funding to create a public sector IPR for the new markers reflects the
weakness of the ENSFI, as the only EU forensic science body. We were
advised that even attempts to reduce analytical costs by reducing the market
dominance of the two dominant companies appear to have been blocked by
legal action to prevent a new company entering the market. On the other hand,
the limited number of commercial companies in the market has at least
prevented an even greater proliferation of multiplex systems. There is also a
more general economic issue arising in respect of IPR. For international
cooperation to work and to assist poorer countries achieve a modern criminal
justice system, it must be possible for less wealthy countries to access DNA
profiling. This problem needs to be recognised as being akin to the issue of
access to generic drugs for developing and middle income countries.
71
Conclusion
4.81.
While information is scarce, the amount of cooperation involving the use of
DNA appears to be occasional even when investigating serious crimes. Within
the European Union this might change as a result of the Prüm Treaty, although
the Council of the European Union has already sought to restrict the extent of
future searching under Prüm, presumably to prevent analysts being
overwhelmed by search requests and the need to verify results.
4.82.
The interoperability of NDNAD and the Scottish database, together with the
ability of the forensic services of the two administrations and Scottish criminal
justice colleagues to manage different retention regimes on a large scale is
clear evidence that forensic cross-jurisdictional cooperation can be successfully
managed on a large scale despite legal differences. While this report identifies
some gaps and lack of consistency in the data and suggests where research is
urgently needed, the scale and detail of the interoperability analysis indicates
that the NDNAD has set an example of greater openness nationally and
internationally for criminal justice databases, including the exchange of
information about fingerprints, and possibly genetic collections in the genetic
medical research and health fields.
4.83.
While the issues of validity, technological development and sample retention
have been considered in the context of international cooperation, it is important
to bear in mind that transnational crime is only a small proportion of recorded
crime, particularly those offences where bioinformation can be obtained from a
crime scene.183 The advantage of considering such issues however, serves to
identify some of the key strategic issues that will need to be addressed for
domestic reasons and acts as a reminder that when engaged in these internal
deliberations the possible global implications should not be neglected. An
example of this is illustrated by the difficulties in reconciling strongly held
bioethical positions with the practical needs of the criminal justice system when
determining sample retention policy. There appears to be a consensus between
the Government and civil society organisations about no longer retaining
samples. The consequences of this will not necessarily just be, as police and
prosecutors have indicated, missed investigative opportunities; it may also
result in incorrect matches being identified. Limited data is available about the
conversion of SGM personal profiles by using the SGM+ multiplex: that is an
increase in maximum discriminating power from an estimate of about one in 50
million to something in the order of one in one billion. Of the 2,000 samples that
were reanalysed in 2003/04, 19% were found to be invalid when the retained
sample was reanalysed, but it was estimated by the NDNAD management team
that the adventitious match rate could have been as high as 26%.184
183
Wilson, T. (2009), ‘Forensic Science and the Internationalisation of Policing’ in Fraser, J. G. and Williams,
R. (eds.) Handbook of Forensic Science (Cullompton: Willan Publishing), pp. 493-500.
184
NDNAD Annual Report for 2003/4 pp. 16 and 19.
72
5
Governance & Accountability
5.1.
The arrangements for the governance of forensic bioinformation collections in
the UK are under unprecedented critical scrutiny. Recent years have seen
significant changes in governance arrangements, especially of the NDNAD, but
there remain concerns that the present governance arrangements may be
inadequate for securing confidence in the ways in which such sensitive
information is collected and deployed by the police.
5.2.
The requirements for ‘good’ (effective and ethical) governance remain poorly
understood in the context of forensic bioinformation, yet it is essential that
consistent management, cohesive policies and robust processes can be
established and evidenced. This includes the future ‘steering’ of forensic
bioinformation policies as well as the mechanisms and consequences of their
current delivery. The adequacy of transparency and accountability
mechanisms, including the relationship between these matters and counterterrorism policy need examination. Further consideration needs also to be given
to questions of integrity, including the setting and monitoring of quality
standards and data protection and security measures. Further, there is a need
for explicit deliberation about the positioning of forensic bioinformation within
wider networks of personal information gathering undertaken by both State and
non-State agencies.
5.3.
There are concerns that governance arrangements of forensic bioinformation
collections in the UK may be inadequate and lessons available from other
bioinformation and biometric data collections in the UK and elsewhere are not
being heeded. This is not because medical biobank governance is always
exemplary; in fact the law governing biobanks in England and Wales has been
found to be “highly complex, confusing, uncoordinated, and inadequate”,185
leading to the conclusion in 2007 that “there is no clear, appropriate,
proportionate or effective framework for governing genetic databases, however
reasonably defined, in England and Wales at the present time.”186 Recent
analysis of the governance of biobanks within medical research,187 has found
regulatory structures to be “... outmoded and seriously deficient” with a
governance patchwork which is “duplicative yet incomplete.”188
185
Gibbons, S. J. Kaye, A. Smart, C. Heeney & M. Parker. (2007) ‘Governing Genetic Databases: Facing
Research Regulation and Practice’, Journal of Law and Society 34 (2) pp163-189. p.171.
186
ibid. p.188.
187
E.g. ELSAGEN at www.elsgen.net; Ethox Centre ‘Governing Genetic Databases’ project at
http://www.ethox.org.uk/research/research-archive/governing-genetic-databases; ESRC Genomics Forum
at : http://www.genomicsforum.ac.uk/
188
Gibbons, S. ‘Regulating Biobanks: A Twelve-Point Typological Tool’, Medical Law Review, 17,
(2009) pp313-346. p.314.
73
5.4.
Research into biobank governance has rarely given detailed attention to
forensic genetic collections,189 governance of biobanks for medical purposes
being seen as separate from criminal justice purposes. Yet, as has been
articulated within the biobank regulation arena:
“...where fundamental rights, values, competing interests, risks and
(public) concerns are implicated, unfettered freedom is neither
beneficial nor wise. In such areas, legitimacy, fairness, justice, and the
statutory ‘five principals of good regulation’ arguably demand a much
more focused, coherent, properly tailored, internally nuanced, yet
suitably flexible and responsive regulatory regime.”190
5.5.
This idea of the distinctiveness of forensic bioinformation collections is
reinforced by the Human Tissue Act, (which created the Human Tissue
Authority). This Act was intended to bring oversight to collections of human
tissue, yet does not cover tissue or analysis obtained by examining or testing
material that may be considered to be evidence or information to further
enquiries under PACE. This reflects an important and practical division of
responsibility between medical regulation and the criminal justice system to
ensure that, among other objectives, there is no conflict between this system of
regulation and the law relating to criminal evidence. Until the passage of the
Crime and Security Act 2010, the NDNAD could be described as “essentially
self-regulated.”191 Recent conclusions concerning the regulation of biobanks
may then be even more forcefully put when referring to the governance of the
NDNAD: “[there is] a de facto over-dependence on informal systems, selfregulation, and ‘soft’ regulatory techniques; and a worrying legitimacy deficit.”192
5.6.
Research into the ‘regulatory space’ of medical biobanks has found that
regulatory systems are reliant upon voluntary compliance with systems
operated by medical professionals, for medical professionals, thus raising “the
sceptre of ‘regulatory capture’ and the possibility that existing governance
structures lack sufficient independence and neutrality.”193 Such a regulatory
structure can also exclude other stakeholders and alienate the public, which
has negative impacts: “as many studies have found, public accountability is vital
for bolstering public trust, confidence, support and participation.”194 The
possibility of ‘regulatory capture’ is a serious risk within the field of forensic
bioinformation, with databases run by the police, for the police, and overseen
189
However, the ESRC Genomics Forum and GENEBANC project both have considered ‘forensic’
databases as part of their research. The former has responded to the recent HGC consultation, and the
latter has published a report on forensic databanks. For an account of the latter, see GENEBANC: Genetic
Bio and Data Banking: Confidentiality and Protection of data. Towards a European Harmonisation Policy.
Available at: http://www.genebanc.eu/ .
190
op.cit. n.188. p.346.
191
op.cit. n.185. p.178.
192
op.cit. n.188. p.314.
193
Kaye, J. & S. Gibbons, ‘Mapping the Regulatory Space for Genetic Databases and Biobanks in England
and Wales’ (2008) 9, Medical Law International, pp111-130. p.126-7.
194
ibid. p.126-7.
74
largely by the police. However, the focus should remain on good governance
with visible accountability and appropriate levels of external involvement, rather
than demanding the police cede control of the databases altogether, as some
commentators, most notably Sir Alec Jeffreys, have suggested. As ‘owners’ and
‘users’ of the databases, the police should always remain at the centre of any
arrangements since:
“intelligent accountability requires more attention to good governance and
fewer fantasies about total control. Good governance is possible only if
institutions are allowed some margin for self-governance or a form
appropriate to their particular tasks, within a framework of financial and
other reporting.”195
5.7.
The government laid out ‘principles of regulation’ in the Legislative and
Regulatory Reform Act 2006, s.21(a) which states that “regulatory activities
should be carried out in a way which is transparent, accountable, proportionate
and consistent”. This Act was intended to bring into effect the Better Regulation
Commissions’ ‘five principals of good regulation’: regulation should be
transparent; proportionate; accountable; consistent; and targeted only at cases
in which action is needed.196 However, to fully appreciate regulation, it is
important to not simply focus on just legal ‘rules’ concerning an activity or
process etc., or the legal powers invested in official ‘regulators’. There can be
other limitations or curbs on activities that, while they are not invoking ‘law’
directly to exert control, can nonetheless often have a regulatory effect.197
The current network of forensic bioinformation governance
5.8.
There are numerous bodies, agencies and organisations that offer resources
for the setting of ethical standards for collections of bio-information,198 such as
the Human Genome Organisation (HUGO); the Council of Europe; and the
United National Educational, Scientific and Cultural Organisation (UNESCO).
Most European States also have a National Ethics Council,199 and while the UK
does not have such an entity, both the Nuffield Council on Bioethics, and the
Human Genetics Commission have reported on forensic bioinformation.200
Many international organisations have outlined ‘principles’ of governance for
genetic collections (mostly bio-banks for medical research purposes).
According to Capron et.al, these have “...certain common assumptions,
195
O’Neill, Onora, ‘The Reith Lectures 2002 – A Question of Trust.’ Lecture Three, available at:
www.bbc.co.uk/radio4/reith2002 .
196
Better Regulation Task Force, Regulation – Less is More: Reducing Burdens, Improving Outcomes: A
BRTF Report to the Prime Minister (2005) Annex B, see also the Legislative and Regulatory Reform Act
2006, ss2 and 21.
197
See Baldwin, R. & Cave, M. Understanding Regulation: Theory, Strategy and Practice. (Oxford University
Press; Oxford, 1999).
198
Known variously as ‘biobanks’; ‘genetic databanks’; and/or ‘genetic databases’.
199
In the US there is the National Bioethics Advisory Commission (NBAC).
200
Human Genetics Commission ‘Nothing to Hide, Nothing to Fear?’ (2009) and Nuffield Council on
Bioethics ‘The Forensic Use of Bioinformation: Ethical Issues (2007).
75
numerous points of difference, and several lacunae regarding practical issues
that are not fully addressed.”201 Indeed, the same authors claim that:
“Despite the rapid proliferation of genetic and genomic databases around
the world, and the associated burgeoning of national and international
guidance documents suggesting various ways of governing them,
consensus over the most appropriate ethical norms and legal rules is still
a very long way off.”202
5.9.
In particular, the ‘ownership’ of genetic material remains controversial for
medical researchers and lawyers alike. Ethical guidance documents may
“disguise the extent to which people use the term ‘ownership’ in very different,
indeed, contradictory ways.”203 Within the realm of forensic bioinformation, it is
usually asserted that all information is ‘owned’ by the police and inevitably, this
claim to ownership has been important in the historical development of relevant
governance structures. For instance, when applying to have personal
information removed, it is to the Chief Constable of the force that obtained the
bioinformation that one must apply, reflecting the force ‘ownership’ of the data
and samples from which the data were derived. This mirrors the fact that
information on both the PNC and IDENT1, ‘belongs’ to Chief police officers, and
is administered on their behalf by the NPIA. Whether this concept of ownership
of such large bioinformation collections continues to be appropriate was most
recently questioned by the Home Affairs Select Committee, who were told that:
“DNA profiles are owned by whichever police force entered them on the
database, with the Chief Constable acting as the data controller... a
strength of the current system was that data were owned by the individual
police forces rather than a central organisation. They felt this guaranteed
that greater care was taken in the recovery and recording of DNA material
and profiles.”204
5.10.
This division of responsibility and implicit assumption about ownership are
given statutory force through the Crime and Security Act 2010. Should there be
an opportunity for more considered legislation, the ownership issue might be
revisited and the hybrid arrangements in the Act - by which, chief officers of
police may make deletion decisions at their discretion in the light of principles
set out in the Act, but must comply with guidance issued by the NDNAD
Strategy Board, might be replaced with mandatory principles set out in a single
statutory code and with an independent tribunal to determine appeals.
201
Capron et al, ‘Ethical Norms and the International Governance of Genetic Databases and Biobanks:
Findings from an International Study.’ (2009) 19 (2) Kennedy Institute of Ethics Journal, 101-124, p.102.
202
ibid., p.101.
203
ibid., p.107.
204
Home Affairs Committee, ‘The National DNA Database’ Eighth Report of 2009-2010, March 2010, HC
222-1. para. 39.
76
5.11.
While there have been recent changes to the governance of the NDNAD, these
have been narrowly focussed and predicated upon the need to take the
NDNAD out of the Custodianship of the Forensic Science Service, (once they
became a government owned company) and then out of the Home Office. In
April 2007, the NPIA took responsibility for the delivery and operational
oversight of the NDNAD, ensuring it is operating to set standards as well as
oversight of the laboratories and suppliers in terms of delivery of service,
maintenance and accreditation.205 The NPIA are also responsible for all
IDENT1 services, overseeing the delivery (by Northrop Grumman) of NAFIS
and LIVESCAN systems by the private contractor.
5.12.
The NDNAD Strategy Board, chaired by the ACPO lead on DNA, is responsible
for the overall strategic management of the NDNAD. The NPIA, as Custodian of
the database, is responsible for:
•
Providing the IT infrastructure;
•
Setting the requirements and monitoring the supplier laboratories;
•
Delivering the NDNAD services to police forces;
•
Ensuring the integrity of the management of the NDNAD and the data;
•
Developing the database in line with police requirements;
•
Providing management information.
The Ethics Group acts independently, while the Forensic Regulator as their
sponsor, is tasked with ensuring their independence and advising the Minister
on the Group’s findings and recommendations.
Issues and Prospects
5.13.
The governance of the NDNAD in many respects sets an international and
national benchmark for transparency (e.g. with the amount of information now
provided in the annual report and the NPIA website). Indeed, the governance
arrangements for DNA have already been subject to more public deliberation
than those for other police information, including IDENT1 and the Police
National Computer (PNC). Nevertheless, NDNAD governance has developed
piecemeal as it seeks to catch up with the UK’s pioneering use of forensic DNA.
The Magee Report of 2008, looking at the broader area of criminality
information and public protection, pointed out that:
“there is no overarching architecture for criminality information and no
individual or organisation that could reasonably be held responsible for its
absence. Each of the many organisations in the public protection network
has its own accountabilities but none is accountable for the whole.”206
205
NPIA Website, accessed 1 December 2009.
Sir Ian Magee, ‘The Review of Criminality Information’ (The Magee Report) (July 2008) Home Office,
London. p.7.
206
77
5.14.
The NPIA may have been more forward thinking with arrangements for the new
Police National Database, having sought from the outset to create mechanisms
to ensure its transparency. This body is also considering the governance of,
and access to, the PNC, this being a recommendation of the Magee Report.207
Indeed, Magee’s first recommendation stated that all organisations dealing with
personal information needed to clarify their governance arrangements, in
particular, settling: “where ownership and accountability lie.”208
5.15.
There are however, concerns over the membership, oversight and transparency
of the workings of the NDNAD Strategy Board. Efforts of Human Genetic
Commission (HGC) members to publicly record concerns had not worked as
intended and their informed ‘lay’ presence depends upon the continued
cooperation and, indeed, existence of the HGC. A case can be made for further
widening lay membership in order to maximise public involvement and
engagement in order to secure confidence in the transparency of the working of
the Board as well as to ensure consideration of a range of public concerns.
5.16.
In the foreword to the 2009 NDNAD Annual Report, the Chairman of the
Strategy Board conceded that the governance of the NDNAD needs to be more
‘broadly based’ and should “recognise the individuals and organisations that
can input and contribute to the effective operation of the NDNAD.” To this end,
he proposes extending membership of the Board to include “the DNA Ethics
Group, Human Genetics Commission (HGC), NPIA, Forensic Science
Regulator and representation from Northern Ireland and Scotland and the
Information Commissioner Office (ICO) as an observer.” Whether this is
‘broadly based’ remains open to question, with members of the HGC already on
the Board (as ‘lay’ members) and the Ethics group already playing a role. Other
agencies – like the Regulator and NPIA are already included within the existing
networks of forensic influence. It is also important to note that the ICO
representative is included as an ‘observer’ only. However, he also stated
encouragingly that “a collaborative approach by all the relevant agencies is the
most effective way to provide oversight and direction for the NDNAD.”209
5.17.
A comparison with the National DNA Index System (run by the FBI) may be
instructive. US National and State DNA Boards include a range of police,
government, academic and civil society actors on their oversight bodies. The
National DNA Index System is also monitored and audited by the Department
of Justice Office of the Inspector General. There would be value in further
considering the composition and remit of these bodies and those which exist in
other jurisdictions. The relationship between these kinds of oversight bodies,
207
Recommendation 24: By Spring 2009, ACPO working with NPIA and stakeholders, should clarify the
governance of the PNC and develop a clear and agreed approach in the light of the issues this report
identifies as to who in which organisations should have what access to the PNC.” The Magee Report p.72.
208
Thomas, R. & M. Walport ‘Data Sharing Review’ (July 2008).p.2.
209
All quotes from: Chairman’s Introduction, NDNAD Annual Report 2007-09, p.2.
78
regulators, and the various technical working groups that seek to establish
quality standards in the US and elsewhere is also a subject of interest and
should be investigated further. What models might be found for acceptable
governance frameworks in, for example, Canadian, French or Dutch DNA
database legislation? Finally, Ireland is to establish a DNA database in 2010,
which will be overseen by an independent oversight committee, with
unrestricted access to the database. The Committee, to be chaired by a judge
(or former judge) of the High Court or Circuit Court, will report annually to the
Justice Minister. There are severe penalties in place (5 years imprisonment and
a Euro 50,000 fine) for unlawful disclosure of data from this database, similar to
laws in place in the EU and applicable in England and Wales.210
5.18.
As long ago as 2001, the House of Lords Science and Technology Committee
called for the establishment of an independent body: “to oversee the workings
of the National DNA Database, to put beyond doubt that individual’s data are
being properly used and protected.”211 Similarly, in their 2002 Report ‘Inside
Information’, the Human Genetics Commission voiced concerns about oversight
of the NDNAD, recommending: “at the very least, the Home Office and
Association of Chief Police Officers establish an independent body, which
would include lay membership, to oversee the work of the National DNA
Database custodian and the profile suppliers.” They claimed that the
Government had indicated it would consider this, and “conduct a review aimed
at improving the security and efficiency of the National DNA Database”.212 The
HGC repeated their call in 2009: “we recommend that an independent body be
established to oversee the management and use of the NDNAD, and that this
body should conduct its business in an open and transparent way to the fullest
extent that the operational sensitivities of policing will allow.”213
5.19.
The consultation paper, ‘Keeping the Right People on the DNA Database’,
envisaged supplementary arrangements to the NDNAD governance structure,
but the scope and timing of these were vague. The Forensic Regulator will still
have no direct oversight of the NDNAD, but will be involved in establishing and
enforcing relevant scientific quality standards with respect to the NDNAD and
DNA profiling. The latest government proposals in response to the consultation
continue to suggest minimal changes to governance structures, with few
details, except for a requirement for the Home Secretary to lay the NDNAD
Strategy Board’s annual report, subject to any redactions, before Parliament,
and for them to take responsibility for the guidance issued to Chief Constables
over removal requests. These changes also look to be limited to the
governance of the NDNAD, with no explicit mention of IDENT1.
210
‘Gardai get new powers to collect DNA from bodies’, The Irish Independent, 21 December 2009.
House of Lords Science and Technology Select Committee, Fourth Report ‘Human Genetic Databases'
(March 2001). Para.1.27.
212
Human Genetics Commission, ‘Inside Information’ (May 2002, London). p.23.
213
Human Genetics Commission, ‘Nothing to Hide, Nothing to Fear,’ (November 2009, London) p.7.
211
79
5.20.
With regard to fingerprints and IDENT1, there remain significant governance
questions. There is no annual report published for IDENT1, in contrast with the
NDNAD. There is also no equivalent ‘IDENT1 Strategy Board’ that is apparent.
With the demise of the National Fingerprint Board, and the absence of any
replacement, the NPIA (under the guise of Forensics21) now provides oversight
of fingerprint issues nationally. Arrangements for the governance of fingerprints
reflect their dual purpose: the validation of identity (e.g. to expose alias use
during police enquiries or after arrest); and the establishment of identity of
those individuals whose fingerprints are recovered from scenes of crime. Both
of these processes are managed internally within the police service.
5.21.
The NPIA set the operating model for the use of fingerprints, and also run the
Police Database Board, (which covers all police databases), where major
stakeholders are represented (though there are no external lay members). The
NPIA have undertaken reviews of quality and policy and made
recommendations in reports submitted to Chief Officers. There has also been
comparative work undertaken between forces to ensure improvement and
compliance where appropriate. These reports are submitted to the Home Office
who monitor the cost benefits of their subsidy for LIVESCAN units. However
there is some remaining confusion over the governance of fingerprint work.
Whilst there continue to be regional meetings, practitioners need guidance on
quality and policy issues surrounding fingerprinting. The international exchange
of fingerprints is an issue of particular significance and seems likely to require
the further development of current governance arrangements.
5.22.
Whilst there is evidence of dissatisfaction with current arrangements for the
governance of forensic bioinformation, it is also difficult to see any consensus
on what mechanisms are needed to assure transparency and accountability of
the uses and users of forensic bioinformation. Any mechanisms have to
account for commercial considerations and interests in commercial
confidentiality at the same time as making publicly available as much
information consistent with concerns with security and confidentiality. There is
also a need for appropriate levels of consistency in policy and practice in the
management of different kinds of forensic bioinformation databases.
5.23.
The Marper judgment presents a major opportunity to review current
arrangements. Parliament could seek to set out ‘principles of governance’ either
for all bioinformation collections or all forensic science collections – including
both DNA and fingerprints. This requires a forward-looking perspective to
provide flexibility to deal with - as yet - either unanticipated futures or
developments not yet validated, in the science and its application. It may be
helpful to consider the relevance of Gibbons et al (2007) eight ‘desiderata’ for
an ideal medical biobank governance framework. In this account, a framework
governance structure would:
80
•
•
be firmly grounded in principle and normatively justifiable;
reflect the characteristics and address the issues identified as being
important by scientists, researchers and clinicians;
promote and facilitate valuable, lawful, and ethical genetic research;
provide appropriate protection for individual rights and public interests;
contain mechanisms to resolve conflicts between competing rights,
values or interests;
be straightforward, accessible, and clearly drafted;
have coherence, clarity and internal consistency; and
be readily adaptable to reflect future reforms, technological
advancements or changing needs.214
•
•
•
•
•
•
5.24.
In addition, there is a need for an ‘ethically robust’ approach to forensic science
practice, which should be the starting point of a governance structure. As Irwin
comments when considering science and ‘values’:
“Although scientific knowledge does not have a moral dimension in itself,
science is conducted by individuals who certainly possess morality and
values and these should be applied to their work. Scientists should
‘declare’ these values, engage with the values of the public, (and in so
doing become far more likely to command public support).”215
Such an ethically robust approach to governance, which began with
attentiveness to human rights principles would require identification of relevant
rights and expectations – privacy, consent, transparency, accountability,
proportionality, etc.
5.25.
It should also be borne in mind, that while the UK has presently not ratified the
European Charter of Fundamental Rights, it became law for most EU countries
on 1 December 2009. This Charter includes respect for private and family life
(Article 7) but also includes Article 8, the Protection of Personal Data (see Data
Protection later) and Article 41, the Right to Good Administration, which states:
•
•
Every person has the right to have his or her affairs handled impartially,
fairly and within a reasonable time...
This right includes:
o the right of every person to be heard, before any individual measure
which would affect him or her adversely is taken;
o the right of every person to have access to his or her file, while
respecting the legitimate interests of confidentiality and of
professional and business secrecy;
o the obligation of the administration to give reasons for its decisions.
214
op.cit. n.185. p.171.
Irwin, A. ‘The Politics of Talk: Coming to Terms with the ‘New’ Scientific Governance’, Social Studies of
Science, (2006) 36 (2) 299-320, p307.
215
81
Of course this Charter remains new and ‘untested’, and there remain difficulties
in specifying the operational meaning of key terms. It cannot be applied in the
UK and there is no agreement on whether this should be a consideration, or
whether governance would be better shaped by reference to existing human
rights protections.
5.26.
‘Principles’ need to be built into legislation (though readiness to rhetorically
evoke such principles is often not matched by rigorous attempts to define them
in the context of forensic science, the justice system or even the ECtHR). At
present, the incremental approach to bioinformation, with no primary legislation
effectively setting out the basis of the NDNAD or IDENT1, has meant that:
“We now have multiple pieces of legislation which need to be fitted
together in order to understand exactly what is going on… what is missing
is independent, accountable and powerful oversight; a fundamental
reappraisal of the basis of the National DNA Database; a suitable
framework for its development, it’s management and it’s governance –
which is not actually in law at the moment – clarity of purpose and also
articulation of the values that actually underpin this, which are lost in a
morass of laws…”.216
5.27.
In ‘Inside Information’, the HGC started with an overarching principle of ‘respect
for persons’ from which they derived a number of secondary principles,
including the principles of privacy; consent; confidentiality and nondiscrimination.217 Such ‘principles’ could then include (inter alia):
•
•
•
•
•
5.28.
consent (for victims, witnesses and volunteers), or when the law requires
from suspects or convicted offenders;
proportionality (recognised as particularly difficult to define in legislation);
transparency (public confidence/ trust);
accountability;
inclusiveness (including independent lay membership of governing
bodies).
There may be alternative or additional principles that need to be considered,
and the changing conceptions of issues such as ‘privacy’ in the 21st century
also need to be taken into account. How are such principles best incorporated
into legislation (i.e. in relatively inflexible primary legislation, in more easily
amended secondary legislation or to be interpreted and developed as a ‘living
document’)? Should measures for ensuring conformity of practice and
governance in line with such principles only take place at a national level when
police forces are organised locally?
216
Prof. Graeme Laurie, Evidence to the House of Lords Select Committee on the Constitution,
‘Surveillance: Citizens and the State’, 2nd Report of Session 2008-09, (TSO, London). para. 209.
217
Human Genetics Commission, ‘Inside Information’ (May 2002), p.14.
82
5.29.
Instead of a fragmented and disjointed legislative and jurisprudential
framework, there is a need for a statutory basis for biometric governance which
would be based upon these fundamental principles, and attentive to human
rights and ethical considerations. This could take the form of primary legislation
with extensive powers to modify the details of the scheme in line with the
evolution of science and practice via secondary legislation. However, whilst
some argue for a single piece of legislation providing the framework for the
governance of all collections of forensic bioinformation, others believe that
greater consideration has to be given to the intended uses of such information
prior to constructing an appropriate legislative framework.
5.30.
Any system of governance has to be capable of distinguishing between
different objects of interest as well as the differing uses to which the information
can be put. Whilst one-to-one identification may seem the underlying ambition
of profile comparison and fingerprint matching, the additional uses of DNA
samples and profiles (e.g. for familial searching and for ethnic inferencing) may
create special problems that need to be separately addressed in any regulatory
framework or governance arrangements. A governance strategy also needs to
address issues arising from the regulation of a mixed economy within forensic
science.
5.31.
It can be argued that decisions concerning the content and use of databases
and their regulation are inextricably linked. It is useful then to think first about
DNA and fingerprint practice – the collection from whom, comparison with what
records for what investigative purposes, etc. – and only then decide on the
appropriate framework within which regulation should be located. The Human
Genetics Commission have called for a ‘definition of purpose’ and stress that
there is a need for a clear purpose will then precipitate clarity on other
questions such as retention, governance etc. Legislation and operational
practice can then follow, for example, legislation should clearly delineate
‘boundaries’ for the use and protection of bioinformation/ forensic science e.g. if
used for ‘counter-terrorism’ purposes.
5.32.
So before legislation is drafted, a number of questions need to be addressed in
order to ensure clarity, and that the legislation is providing the powers and
governance that are necessary for ‘success’. Such questions would include:
•
•
•
•
What are the range of uses to be made of the data (bioinformation) and
what safeguards are required to protect its integrity and keep it secure?
What is it that makes data ‘sensitive’ when the state/ police hold it?
Is bioinformation more ‘sensitive’ than any other police information?
Is DNA data more ‘sensitive’ than fingerprint data?
83
•
•
•
Can requirements for good governance of bioinformation stand apart from
the requirements for good governance of forensic science in general?
Is any police information ‘more’ sensitive than other types of public
protection information (i.e. immigration; vetting and barring information
etc.) – in which case do we need a ‘public protection information’
governance framework?
Are new issues raised by:
o
the increasing effort to ‘join-up’ information databases;
o
the needs of counter-terrorism;
o
the increasing use of such data collections for the identification of
bodies, especially in Disaster Victim Identification work?
5.33.
Greater consideration of the type of information, and the uses to which it can be
put are essential for shaping any new governance structure for this domain of
public life. As Magee points out: “effective governance requires an appropriate
balance of ownership, process and control… the key requirement is for a
governance mechanism that is aligned to the purpose it serves.”218 There may
be a need for multiple frameworks of governance, particularly if the new
governance structures are intended to have broad application – for example, to
all forensic bioinformation types, to all forensic science, or even all personal
information managed by the police, or information for public protection.
5.34.
If bioinformation were to be regarded as primarily a source of information for the
police, then it might come to be managed under MOPI (Code of Practice on the
Management of Police Information, 2005).219 This Code of Practice was
introduced in direct response to criticism in the Bichard Report about the
handling of information concerning Ian Huntley prior to the murder of two young
girls. This would clearly have an impact on the overall governance structure,
and integration of different forms of information and management of MOPI,
together with the ultimate scrutiny of police information by the Information
Commissioner’s Office.
5.35.
Magee has recently called for the creation of a ‘public protection network’ which
would encompass all agencies that have a public protection and crime
prevention remit, and hold and process personal data to fulfil that role.220 This
218
Magee ‘The Review of Criminality Information’ (Magee Report, July 2008) Home Office, London. p.23
MOPI is described as ‘a statutory code’. It has been is promulgated by the Home Secretary under
the Police Acts of 1996 and 1997 (as amended by the Police Reform Act 2002). This legislation sets out
the scope of the code and requires that any such code be laid before Parliament after promulgation and
may contain requirements about consultation, but not that the code itself requires consent or may be
annulled by Parliament as if it were secondary legislation. ‘Chief Officers [of police] are required under
the Act to “have regard” to any such codes’ (Forensic Science Regulator (2008) Manual of Regulation Part One: Policy and Principles, p19 (London, Home Office).
220
This would then include the police and all associated departments under the police umbrella (such as
SOCA/ counter terrorism groups etc.); ‘barring and vetting’ agencies such as the Criminal Records Bureau;
immigration agencies including the UK Borders Agency, and the Independent Safeguarding Authority etc..
219
84
network would then include all relevant agencies and their data, although the
size and complexity of the network would create particular challenges:
“On the one hand we need agile, empowered organisations that can
respond swiftly to specific and local needs. On the other we need
mechanisms that will ensure the safe and appropriate capture, sharing
and use of criminality information between these organisations –
including between nations – and which will command public
confidence.”221
Magee outlines those details requiring explicit agreement and recording across
the protection network that would make clear ownership and decision making
rights, including:
•
•
•
•
•
•
•
“Capture: what information do we seek, for what purpose, and how do we
gather it?
Storage: where do we store it, for how long, and under what security?
Access: how can it be accessed, who has the right to do so, and do they
have the ability to change the information?
Sharing: what interconnections should and do exist between repositories
of information, what is the nature of those connections? E.g. does one
master copy of the record exist, or is the information broadcast to multiple
repositories? How rapidly is information available to those who need it?
Analysis: is information available in a format suitable for its intended use
and when it is needed?
Action: do decision-makers act on the basis of the information available,
and do their actions result in successful outcomes?
Management: do managers understand the performance of the PPN, the
contribution of this information to this and are they able to communicate it
to staff?”222
The challenge of governing a large and dispersed group of agencies and
individuals would be met by: “focused, independent and long-running
accountability…”.223 This would not be a task for the Information Commissioner
alone as decisions about the effective use of the data for public protection
would be beyond the (present) scope of his role.
221
op.cit. n.218. p.23.
op.cit. n.218. p.28.
223
op.cit. n.218. p.33.
222
85
Scientific and Organisational Integrity
Quality Standards
5.36.
Much effort in recent years has gone into establishing and sustaining consistent
quality standards in DNA profiling. The strong emphasis on quality assurance
for this technology contrasts with other forensic disciplines. The role of the UK
Forensic Regulator, with the responsibility of overseeing the quality of all
forensic science, was created in 2007. The Regulator’s Manual (still in draft
form and requiring detailed appendices) sets out requirements for all forensic
science services, in order to maximise the effectiveness of the criminal justice
system and maintain public confidence in the quality and reliability of the
forensic science provided to police forces and others. To do this, the Regulator
establishes and monitors quality standards and ensures accreditation of
suppliers of forensic services, including quality standards which apply to
national forensic science intelligence databases. The NDNAD is mentioned
specifically in this regard, although IDENT1 is not.224 Section 12 of the
Regulator’s Manual states that the Regulator has a role in overseeing quality
standards applicable to national forensic science intelligence databases. The
December 2009 Report from the Regulator states that he is to release quality
standards for the NDNAD and National Ballistics Intelligence Service databases
for consultation in early 2010.
5.37.
The Interim Regulator undertook two reviews of the application of forensic DNA
technology during the first year of the Regulator’s office. However, neither have
been made public because of ‘commercial confidentiality’. A third review, of
LCN DNA was undertaken by the new Regulator during 2008 and was
published on the Home Office website.225 In 2010, the Regulator is undertaking
reviews of familial searching as well as setting standards for new common
frequency databases to be used in future DNA profiling systems.226
5.38.
The Forensic Regulator oversees accreditation (via UKAS) of all laboratories
that supply DNA profiling services to the police. Scenes of crime examination
are included within his accreditation schedule. However there remain questions
over whether quality regulation reforms are being applied equally to all aspects
of bioinformation collection, retention, and usage. The regulator does not for
example, play a similar role in accrediting fingerprint bureaux, nor police
custody suites, where fingerprints and DNA samples from suspects are
acquired. Seven suppliers currently provide profiles to the NDNAD,227 each of
whom must be accredited to ISO17025 & Custodian Standards for accreditation
224
The Forensic Science Regulator, Business Plan 2008/09 – 2010/2011, para 3. available at
http://police.homeoffice.gov.uk/publications/operational-policing/Forensic_Science_Regulator_3.html .
225
ibid., para 11.
226
The Forensic Regulator, Report, (December 2009), p.9.
227
The FSS Ltd; LGC Ltd; Orchid Cellmark; SPSA (Scottish laboratories); FSNI (Northern Ireland);
FDS (Forensic DNA Services); Eurofins (EFS & Medigenomix) with another coming on stream in 2009.
86
(LAB 32) and approved by the National DNA Database Strategy Board. They
need to demonstrate that they have in place anti-contamination measures
(including elimination databases of staff and manufacturers, and un-sourced
contaminant profiles). Their profiles must meet the minimum load criteria and
meet duplication requirements. They must also take part in closed proficiency
testing and validate their process and system changes.
5.39.
On the one hand, the introduction of a regulator was presented as creating a
generic standard for forensic science providers in the UK – based on ISO
standards and ‘a light touch’ in steering providers, but there remain concerns
about perceived lack of ‘teeth’ and gaps in regulation with a fear that
accreditation may prove to be superficial. Civil remedies for breaches of
regulatory standards may be too little, too late and cannot provide an answer to
mistakes already made within the criminal process. It is a matter of regret that
the UK, as a world leader in privatising its forensic science providers within an
adversarial system, had not properly set-up a regulator system before the
introduction of a mixed economy. The emerging shape of the forensic science
marketplace remains difficult to discern. The work of regulating the provision of
goods and services within this marketplace has only recently begun and
deserves careful examination.
5.40.
The December 2009 Regulator’s Report states that the Home Office has now
established a ‘Forensic Science Strategy Group’, including the Regulator and
Chaired by a Home Office senior civil servant, which will meet monthly. This
group is intended to supplement the work of the Regulator, whose remit does
not extend to the broad range of risks to the supply and use of forensic science,
although the Regulator will also be forming an additional sub-group to ‘identify
and manage this broader array of risks’.228 The Regulator is also extending his
work to provide guidance on the interpretation and presentation of forensic
evidence to prosecuting authorities and courts, with a new specialist group
established for this purpose. There are clear moves then to extend the
Regulator’s work into areas that he had previously been reluctant to enter.
There will also be interplay between the Home Office forensic science strategy
group and the work of the Regulator. The constitution and remit of the Home
Office group is as yet unknown.
5.41.
The requirement to have quality standards and accreditation has now become
part of an EU Council framework decision (15905/09). With international
exchange increasing, it is vital that quality standards in forensic science can be
assured across Europe: “The adoption of common quality standards for forensic
science across Europe has become ever more important as the international
exchange of DNA profiles offers greater potential to increase public safety in a
228
The Forensic Regulator, Report, (December 2009), p.3.
87
significantly more mobile society.”229 Now, ‘laboratory activities’ carried out by
forensic science providers must be accredited so that public authorities in each
country can recognise results generated by a laboratory in another EU country.
The national accreditation body of each member state will then ensure
compliance with the relevant international ISO standard. However, there may
still remain gaps in oversight with regard to forensic bioinformation, with
fingerprint bureaux not included under ‘laboratories’, only fingerprint
development laboratories. This distinction permits police fingerprint bureaux to
continue operating without requiring ISO accreditation.
Data Protection and Security
5.42.
Concerns over the use of personal information are not new, though they may
be attracting greater attention. The European Commission found that in Europe:
“public unease about the use of personal information is widespread and
has remained consistent for almost twenty years. Some 64% of EU
respondents – and as many as 77% of UK respondents – expressed
concerns about whether organisations holding their personal data handle
it appropriately.”230
The security of data, particularly personal data, has gained prominence however,
since a series of high profile losses of sensitive information, mostly by public
sector agencies, as highlighted in the Thomas & Walport report:
“Repeated losses of sensitive personal information in both the public and
private sectors demonstrate the weakness of many organisations in
managing how data are shared. The advent of large computer databases
has allowed the loss of massive datasets in ways that were simply
impossible with paper records.”231
5.43.
The Trustguide Report (2006) found a feeling of ‘vulnerability’, where people
reported a lack of ‘control’ over the data collected about them, and this was
most acutely felt around their ‘identity’:
“Technology creates many new challenges and causes us to redefine that
which we may have taken for granted in the physical world and one field
where this is most apparent is the creation and protection of one of our
most valued and most used attributes – our identity. ... Attendees reported
that as more data is gathered and stored electronically, particularly in
centrally controlled databases, they feel more vulnerable. Most of this
vulnerability is focused on a lack of control over who is collecting their
229
Foreword, CC Neyroud, NDNAD Annual Report 2007-09, p.4.
Eurobarometer: Data Protection in the European Union – Citizens Perceptions (February 2008). –
available at http://ec.europa.eu/public_opinion/archives/flash_arch_en.htm .
231
Thomas, R. & M. Walport ‘Data Sharing Review’ (July 2008).p.i .
230
88
data, who might have access to it, how their data may be used now or in
the future and the potential for function creep..”232
5.44.
The Coleman Report (2008) examined the risk of fraud, accidents (including
loss of data); cyber-attacks and threats etc. to government information
holdings.233 There is clear public concern over the processing and sharing of
data by government agencies as well as how securely they hold them. Victim
Support state that they believe the public are unconvinced that DNA data can
be stored securely: “…the Government should take steps to assure the public
that data on the [DNA] database is secure, and that the database contains only
that information which is necessary to adequately protect the public.”234
5.45.
The Information Commissioner dealt with 94 breaches of the Data Protection
Act in the five months preceding April 2008 alone, indicating that these are far
from isolated incidents. Control of data becomes a greater issue with increases
in the amount of protected data and the number of users of the data. Many
breaches, some which may occur daily, can be relatively insignificant, with little
impact and no sanctions necessary. However, some breaches have been
significant, with serious implications and high profile media attention has led to
a plethora of reports, inquiries, and reviews, including the ‘Data Handling
Review’ (Hannigan Report), which culminated in a new Security Policy
Framework which provides the guidance on security and risk management for
all government (and associated) bodies, including the police. The SPF contains
70 minimum security requirements that must be adhered to by all government
bodies and agencies, including suppliers of services.
5.46.
The NDNAD Annual Report (2007 – 2009) states that “many of the minimum
standards...were already met or exceeded by the security measures in place for
the NDNAD.”235 However, a comprehensive security review resulted in new
working practices and increased monitoring of activity, leading to improvements
including: reinforced security training for staff; revised incident management
scheme which includes reporting to the ICO; additional specific measures to
protect personal information; stricter controls and measures for the use of
removable media; security clauses in all contracts with external providers;
increased audit and management accountability. The NDNAD has also been
brought within the NPIA main data centre and the use of fax machines has
been phased out, with all communications now undertaken via secure
government email with strict audits and controls and overlying encryption when
required. The NDNAD Annual Report goes on to affirm that:
232
Lacohee, H., S. Crane, and A. Phippen, ‘Trustguide: Final Report’, (Trustguide, October 2006).p 84.
Protecting Government Information – Independent Review of Government Information Assurance (The
Coleman Report), June 2008, Chapter 3.
234
Victim Support, ‘Response to Keeping the Right People on the DNA Database’ (July 2009), p.1
235
NDNAD Annual Report 2007-2009, p.36.
233
89
“Security continues to be given the highest priority within the NDNAD
Delivery Unit and by the NPIA senior management who are committed to
a process of continuous improvement for the NDNAD and to an ongoing
programme of technical and procedural compliance audits to ensure that
the high standards which have been set are embedded and being
implemented within day operational practices by all operational staff.”236
5.47.
The generous accessibility of the PNC, in addition to the practice of sending
demographic details attached to DNA samples to laboratories for analysis,
demonstrate that there may still be ample opportunity for improper access or
lapses in security. The Crown Prosecution Service, in the spotlight for serious
data losses, have recently reviewed and altered policies on their security,
including the use of portable media and encryption. Work on further reducing
risk and strengthening “information risk governance, covering both personal
data and other sensitive information” is ongoing.237 Other law enforcement
agencies that receive or use forensic bioinformation may also be forced to
undergo such reviews to ensure their data security.
5.48.
Relevant to discussion of data governance is consideration of what kind of data
protection laws exist in relation to DNA profiles and fingerprints generated for
criminal justice purposes only. In some jurisdictions, data protection regimes
differentiate between the collection and retention of personal data for criminal
justice and other purposes, but however these regimes operate it is necessary
to understand their application to DNA profiles and to fingerprints before there
can be agreement on a governance model for forensic bioinformation.
5.49.
The EU Directive 95/46/EC on Data Protection,238 which is currently under
review, came into force in October 1998, being given effect in domestic law by
the Data Protection Act 1998. The EU directive focuses on the protection of
individuals with regard to the processing of personal data and the free
movement of such data and applies to DNA data unless specific DNA
legislation determines otherwise. Article 6 of the EU Directive states that
personal data must be: adequate, relevant and not excessive in relation to the
purposes for which they are collected and /or further processed; and kept in a
form which permits identification of the data subjects for no longer than is
necessary for the purposes for which the data were collected or for which they
are further processed. The preamble to the Directive states:
236
ibid., p.37.
The Cabinet Office, ‘Data Handling Procedures in Government: Final Report’ (The Hannigan Report),
June 2008, p.29.
238
See: Status of implementation of Directive 95/46 on the Protection of Individuals with regard to the
Processing of Personal Data, available at
www.europa.eu.int/comm/justice_home/fsj/privacy/law/implementation_en.htm .
237
90
28) Whereas any processing of personal data must be lawful and fair to
the individuals concerned; whereas in particular, the data must be
adequate, relevant and not excessive in relation to the purposes for which
they are processed; whereas such purposes must be explicit and
legitimate and must be determined at the time of collection of the data;
whereas the purposes of processing further to collection shall not be
incompatible with the purposes as they were originally specified.”
Not all UK commentators feel that Section 64 of PACE is sufficiently exact to
meet the requirements of this Directive.
5.50.
The Data Protection Act 1998 is based upon eight data protection ‘principles’
derived from this EU Directive, that can be applied to the use of all personal
data, including the third principle that states that data must be adequate,
relevant, and not excessive in relation to the purpose for which they were
obtained, while the fifth principle states that data should not be kept for longer
than is necessary for the purpose(s) for which it was collected. The seventh
principle states that “appropriate technical and organisational measures shall
be taken against any unauthorised or unlawful processing of personal data and
against accidental loss or destruction of, or damage to, personal data.”
5.51.
In 2001, the EU established a European Data Protection Supervisor to oversee
the implementation of data-protection standards within EU institutions.
Regulation (EC) 45/2001 also introduced strict rules on data processing and
rights for citizens to access their data. Article 20 however, permits law
enforcement authorities significant exemptions from the data protection regime,
including the processing of forensic data. This limits the role of the EDPS with
regard to forensic data, although he has been highly critical of the Prüm Treaty.
5.52.
In the new European Charter in Fundamental Rights (which the UK has not
signed), Article 8 concerns the protection of personal data, stating that:
•
•
•
•
Everyone has the right to the protection of their personal data;
Such data must be processed fairly for specified purposes and on the basis
of the consent of the person concerned or some other legitimate basis laid
down by law;
Everyone has the right of access to data which has been collected
concerning him or her, and the right to have it rectified;
Compliance with these rules shall be subject to control by an independent
authority.
91
5.53.
It is probably the case that such requirements would be easily met by proper
adherence to the Data Protection Act 1998. However, application to all data –
whether for policing or other purposes – would mean that data protection would
encompass the use of forensic bioinformation across the EU and could have
significant implications, and may require the greater involvement of the
Information Commissioner than has hitherto been the case in the UK (even
though he now sits on the NDNAD Strategy Board as an ‘observer’).
5.54.
The original EU Directive may also be in need of updating:
“Fuelled in part by technological, commercial and social developments
since its adoption in 1995, voices in some quarters are increasingly
questioning whether the Directive, and by inference the UK’s DPA, is still
fit for purpose. Some are calling for the Directive to be reviewed.”239
Moves to strengthen data protection across the EU may dramatically impact
upon domestic data protection. For example, the European Council Framework
Decision of 24 June 2008 on the protection of personal data processed for
police and judicial cooperation in criminal matters emphasises the need for timelimits for erasure and review. Yet in the UK, following the Bichard Inquiry,
changes to the law including: the retention of fingerprints and DNA; the
admissibility at trial of bad character evidence; obligations to provide information
to the CPS and the courts, etc. resulted in ACPO reviewing their ‘weeding’
policies (removing ‘old’ convictions from the PNC). Though still maintaining the
‘step down’ system (where some old convictions are made only available to the
police), they moved to a complete system of retention whereby no convictions
are deleted or ‘weeded’ from the PNC except in exceptional circumstances (i.e.
if the conviction was wrongly obtained). 240
5.55.
The ‘Five Chief Constables’ case241 was prompted by the disclosure of old
convictions relating to five individuals, to potential employers as the ‘stepping
down’ procedure did not prevent disclosure in such circumstances. The
Information Tribunal ruled that this policy of permanent retention of all
convictions was unlawful under the Data Protection Act 1998, and this was
appealed by five Chief Constables in the Supreme Court. The Appeal
concerned whether two of the principles of the DPA were being complied with:
DPP 3 which states that excessive data must not be retained and DPP 5, that
data must not be retained longer than necessary. The police used the ability to
re-open ‘cold cases’ with DNA advances as just one justification for permanent
retention of conviction data.
239
Thomas, R. & M. Walport ‘Data Sharing Review’ (July 2008).p.22.
The areas of bad character evidence, the CPS and courts’ requirements were covered in the EU
Framework decision 2009/315/JHA and updated provisions are in the Coroners and Justice Act 2009.
241
Chief Constables of Humberside Police; Staffordshire Police, Northumbria Police, West Midlands Police,
and Greater Manchester Police, and the Information Commissioner, and the Secretary of State for the Home
Department [2009] EWCA CA Civ 1079.
240
92
5.56.
Lord Justice Waller stated that compliance with the first Data Protection
Principle (that data must be processed fairly and lawfully) did not require
constraints be placed upon the purposes for which data is retained (as long as
they are lawful), but that it is important “that people know what the data is being
retained for and so that the Information Commissioner and data subjects can
test the principles under the DPA by reference to the purposes identified.”242
Further, Lord Justice Carnwath ruled that with respect to the PNC, Article 8 (5)
of the EU Directive recognised the importance of maintaining a ‘complete
register of criminal convictions’ provided that it is “under the control of official
authority” and there existed suitable safeguards provided under national law,
thus: “The power to maintain “a complete register”, to my mind, logically
encompasses power to maintain records of all convictions, without regard, for
example, to age or relative seriousness.”243 It was pointed out that with regard
to the S & Marper ruling, “the nature of the information was quite different.” and
the case “is no authority for the proposition that a record of the mere fact of a
conviction engages Article 8.”244
5.57.
Questions of the effectiveness of data protection legislation with regard to
forensic data then raise uncertainties about the powers of Data Protection
Commission staff. In the ‘Five Chief Constables’ case, it was ruled that the ICO
was meant simply to ‘review’ the police policy on data retention, not replace the
police judgment with his own. While the ICO has recently been given the power
to impose financial penalties, questions remain surrounding the proper use and
extent of his powers. There is also a need for better understanding of how
these powers are actually exercised in the course of routine and exceptional
uses of forensic bioinformation. The Dutch Data Protection Authority has the
right to audit the DNA database and exercises this right regularly. The ICO has
been involved in discussions with ACPO in respect of some uses of the NDNAD
(in particular familial searching). The resulting ‘Memorandum of Understanding’
details safeguards that have been agreed and the ICO has approved the use of
familial searching when: “restricted to the most serious cases and intrusion into
the private lives of individuals is minimised.”245 The precise nature of relevant
safeguards are unknown, however as the MOU is not a public document.
5.58.
To date, the ICO has had limited involvement with the NDNAD, prior to the
commencement of his observer role on the Strategy Board. It is unclear
whether the ICO has any involvement or oversight of other forensic intelligence
databases. However, the Information Commissioner’s response to the
Government consultation expressed surprise at the lack of mention of the DPA,
given that the issue centred upon the fair and lawful use of personal data. As
he and others point out, the obtaining and retention of forensic bioinformation
clearly engages the First Principle of the DPA, while the indefinite retention
242
ibid., para.31.
ibid.. para.72.
244
op.cit. n.241. para.80.
245
op.cit. n.218. p.89
243
93
engages the requirements of the Third and Fifth Principles:
“The fifth Data Protection Principle requires that personal data processed
for any purpose or purposes shall not be kept for longer than is necessary
for that purpose or those purposes. For DNA profiles on those who have
not been convicted the retention period needs to be necessary for the
purpose of policing. ‘Necessary’ in this context, means more than
useful.”246
Yet no attempt was made throughout the consultation document representing
the Government response to Marper, to “address any of the obvious data
protection compliance issues.”247
Scientific Research and Technological Development
5.59.
The use of genetic collections for research is well established in the medical
field, and there is a wealth of material available on how research is to be
approved etc., though the network of ethical committees are often labyrinthine.
The House of Lords Science and Technology Select Committee affirmed the
need for strong oversight of research using genetic databases in 2001, or the
promise of genetic research would not be realised:
“...robust systems with strong oversight mechanisms were needed to
ensure that research on human genetic databases was carried out to the
highest ethical standards, in both private and public sectors....”248
“arrangements for the handling and analysis of personal data held in
human genetic databases should be carried out to the highest ethical
standards, with proper attention to the rights of the individual to privacy
and confidentiality of their personal medical and other data. It is essential
that there is high public confidence in this activity, otherwise some or
many of the benefits to be gained from the advances in genetics will not
be realised.”249
5.60.
The maintenance of high scientific standards is dependent largely upon proper
forensic research. However, the role of Research Councils and major charities
in funding forensic science research remains unclear. Efforts by the NPIA, the
Association of Chief Police Officers (ACPO) and others to encourage university
research should be supported in the hope that the wider scientific community
becomes willing to participate in this work. Such university research will require
246
Information Commissioners Office, Response to ‘Keeping the Right People on the DNA Database’,
August 2009, p.7.
247
ibid., p.4.
248
House of Lords Science and Technology Select Committee, Fourth Report ‘Human Genetic Databases'
(March 2001). Para.7.34.
249
ibid., para.7.46.
94
support in terms of access to data. It may be useful to consider what kind of
partnerships may be possible between NPIA and other actors in order to secure
research and development funds. The UK forensic field is not proactive in
assessing risks and measuring ‘quality’ however. Moreover, unlike the USA’s
National Institute of Standards and Technology (NIST), there is no longer a
government/publicly funded organisation in the UK to set relevant standards for
the forensic community.
5.61.
Arrangements for the use of the NDNAD or cellular materials retained by
providers for research have become more transparent in recent years. The
Ethics Group made it one of their main priorities to prepare proposals for the
Strategy Board that would create a clear and effective governance framework
for this. At present, those making research requests direct these to the
Custodian, who then provide the details to the Strategy Board. Each proposal is
assessed on its merits and further advice sought from the Ethics Group. The
NDNAD annual report states that requests are not agreed unless they have
‘clear operational benefits to the police’, but details in the public domain remain
scant as ‘commercial confidentiality limits what details can be provided about
specific research proposals.’250 Figures are provided with a short description
outlining the general purpose of the research. Since 1995, there have been 46
requests for access to samples or data with 26 have been approved, 18 have
been rejected and two remain under consideration (as at 31st March 2009).251
For further ‘operational’ work to be done using the NDNAD (e.g. Y-STR work)
written authorisation is required from the police force ACPO officer, the CPS
caseworker, as well as the Chair of the NDNAD Strategy Board.252
5.62.
Despite the use of so-called ‘junk DNA’, as the Police Foundation have claimed,
this process may be insufficient and requires clarity and transparency:
“DNA samples are an intimate window into a person’s make-up and
reveal a wide range of genetic information including details relating to
their personal health and family relationships. The Police Foundation
would welcome detailed guidelines on how the DNA will be used and
assurances as to the future use of profiles or planned research as well as
confirmation as to which bodies will be allowed access to the data.”253
It is clear however, that the use in research of DNA profiles or retained samples
by third parties is a matter of public concern. The HGC Report ‘Inside
Information’ reported that:
250
NDNAD Annual Report 2007-2009 p.41.
ibid., p.38.
252
ibid., p.42.
253
Police Foundation response to ‘Keeping the Right People on the DNA Database’, (2009), p.4.
251
95
“Two thirds of those in our People’s Panel survey felt any access was
inappropriate. Others felt that if there is to be any access.... for research
purposes then this should (a) only by for the purpose of crime detection
and prevention and (b) the normal conventions of medical research
should apply with regard to informed consent and confidentiality.” 254
5.63.
Research with actual or potential forensic science applications may often be
carried out ‘on the back of’ more generic scientific research conducted in
universities and elsewhere. However, following the changed status of the
Forensic Science Service, it is not clear which bodies may have a role in
searching for or disseminating the results of scientific work with potential
forensic applications. There remain significant and distinctive problems
surrounding IPR in forensic science research, and ways need to be found of
balancing the need for commercial confidentiality with the interests of the court
in transparent and reproducible scientific work.
Public Confidence and Trust
5.64.
A key criterion for effective policing is public consent. As Flanagan stated in his
2008 review of policing:
“Policing is far too important to be left to the police alone. It is a public
service and one that can only be effectively carried out with the support
and consent of the public. Using and developing this engagement with the
public is one of the most important challenges in modern policing and it is
a challenge that must be met at all levels.”255
Indeed, Flanagan states that the public: “must always be the single most
important aspect of policing” as it is only through engagement with the public
can the police understand their priorities, and be accountable.256 Public opinion
is therefore an important aspect of policing by consent, but more widely, is vital
to any policy decision-making within a representative parliamentary democracy.
5.65.
For public consent to policing to be maintained, they must have confidence in
the police and their governance. Public confidence features prominently in
Public Service Agreement 24 concerning the effectiveness, transparency and
responsiveness of the criminal justice system as a whole. Openness and
transparency are crucial to the gaining and maintenance of public confidence
and trust in the operation of all organisations. It is crucial not just for the
operation of the criminal justice system, but for government as a whole, that
254
Human Genetics Commission, ‘Inside Information’, para.9.50.
Flanagan Report ‘The Review of Policing – Final Report’ (February 2008) Home Office, p.5 available at
www.police.homeoffice.gov.uk/publications .
256
ibid., p.7.
255
96
there is trust in public institutions, indeed, “Trust is the basis for human rights
and democracy.”257
5.66.
As has been stated on many occasions, the use of forensic bioinformation
during investigations and prosecutions requires the confidence of the public for
it to be effective in securing convictions. The public may also expect the
government to exercise their duty of care and utilise technologies available to
them in order to protect the public.258 However, the public also need to trust that
government will use technologies wisely and not overreach their powers. With
the storing of sensitive data, the public need to have confidence that the
databases are secure and being used legitimately. As the Fraser Report states:
“The DNA database stores private genetic information and continued public
confidence in the legitimate use and adequate safeguarding of this information
relies on accountability and transparency of information.”259
5.67.
The House of Lords Select Committee on the Constitution previously claimed
that policy-making in the area of surveillance has not been up to standard, with
too few details and a lack of specificity in legislative Bills, meaning that
Parliament was unable to effectively scrutinise proposals.260 The Committee
also stated that: “The openness of organisations, both about their personal
information and surveillance plans and practices, and about ways in which the
public can be more effectively involved in understanding and shaping them, is
important.”261 The Committee argue that:
“If trust in relationships between the citizen and the state is to be
maintained, public understanding of surveillance and the way in which
personal data are processed must involve organisational transparency,
starting at an early stage in the Government’s policy proposals.” 262
5.68.
The Thomas-Walport Report also portrayed transparency as a prerequisite for
public trust, emphasising that public bodies must provide clearer and better
information to the public about data sharing.263 That standard-setting bodies
must operate in an open and transparent way is a ‘basic requirement for public
trust’.264 The Royal Commission on Environmental Pollution considered that
‘transparent’ meant: “there must be full publicity for their existence, their terms
257
O’Neill, Onora, ‘The Reith Lectures 2002 – A Question of Trust.’ Lecture Two, available at:
www.bbc.co.uk/radio4/reith2002
258
Royal Academy of Engineering, Dilemmas of Privacy and Surveillance (RAE, London, March 2007) p.44.
259
Fraser, J. (2008) Acquisition and Retention of DNA and Fingerprint Data in Scotland p.14 as published in
September 2008 at www.scotland.gov.uk/Consultations/Current .
260
House of Lords Select Committee on the Constitution, ‘Surveillance: Citizens and the State’, 2nd Report of
Session 2008-09, February 2009, The Stationary Office, London. para. 357.
261
ibid., para.433.
262
ibid., para.434.
263
Thomas, R. & M. Walport ‘Data Sharing Review’ (July 2008). para 8.14.
264
Royal Commission on Environmental Pollution, (1998), ‘Setting Environmental Standards’, 21st Report,
(London, The Stationery Office).
97
of reference, the decisions they take and the reasons for them…”, while
‘openness meant that: “there must be adequate opportunities for those outside
an institution...to contribute fully to the decision-making procedure.”265
‘Transparency’ however can be a catch-all term that can mislead, with the
Thomas-Walport Report leading to provisions in the 2009 Coroners and Justice
Bill, later dropped, that included almost carte-blanche data-sharing between
government departments and bodies, a move that was widely rejected.
5.69.
The Government has made a commitment to enhanced transparency, outlining
‘key principles’ to which they will give due regard when reviewing policy:
•
•
•
Are robust safeguards in place to protect information and individual
liberties?
Are our plans and actions proportionate to the damage and the threat
they are seeking to prevent?
Are we being as transparent as possible? Are citizens being given the
right amount of choice?266
All government departments are required to publish ‘Information Charters’ which
outline the information they keep, how it is used and contained, with a
presumption of openness. However, “there will always be some information and
some uses of it (e.g. in the area of national security and law enforcement
arenas) where transparency must rightly be limited…”267 The government has
however, demonstrated this commitment to openness and transparency in other
areas, for example in 2008 the National Security Strategy was published,268
committing government departments to publicly available standards, and in
2009, the new Security Policy Framework, which sets out universal mandatory
standards for security of information in government, has largely been placed in
the public domain for the first time; “allowing greater access, increasing
awareness, transparency, and sharing good practice.”269
5.70.
However, the development of IDENT1 and the establishment of the NDNAD
have until recently been considered operational policing matters and decisions
have been taken internally, without public consultation or external involvement.
A similar situation as has occurred in other countries setting up their DNA
databases, resulting in difficulties with public confidence:
265
ibid., p.124.
Government Response to the House of Lords Select Committee on the Constitution’s Report:
‘Surveillance: Citizens and the State’, 13 May 2009, Cm7616, The Stationery Office, London. p.1.
267
The Cabinet Office, ‘Data Handling Procedures in Government: Final Report’. June 2008, p.23.
268
Cabinet Office Homepage, ‘Understanding the Security Policy Framework’ available at:
http://www.cabinetoffice.gov.uk/spf/faqs.aspx accessed on 23 December 2009.
269
Sir Gus O’Donnell, ‘Foreword’ in ‘HMG Security Policy Framework’ (October 2009), p.5
266
98
“Clearly there has been little opportunity for civic engagement and
biological citizenship with respect to the establishment and operation of
forensic DNA databases in Australia... there has been a technocratic
policy approach....A strong companion of technocratic systems in public
distrust.”270
5.71.
Rose & Novas’s concept of ‘biological citizenship’,271 stresses the importance of
democratic decision-making with regard to bioethical issues. In recent years the
Government have at least espoused a new more ‘open’ form of decision
making, with the public engaged in science policymaking. Tony Blair claimed in
2002 that: “the benefits of science will only be exploited through a renewed
contract between science and society, based on a proper understanding of
what science is trying to achieve.”272 Indeed, in 2000, the House of Lords Select
Committee on Science and Technology recommended that “direct dialogue with
the public should move from being an optional add-on to science based policymaking and to the activities of research organisations and learned institutions,
and should become a normal and integral part of the process.”273
5.72.
However, a commitment to social consensus on science policymaking assumes
the location of an elusive ‘public’ with a homogenous ‘public opinion’.274 The
House of Lords Select Committee on the Constitution have previously stated
that: “Assertions about what ‘the public’ feel or want concerning surveillance are
not conclusive, although they do often go unchallenged.”275 Within the realms of
forensic bioinformation, this can clearly be seen to be a false expectation, as
public opinion concerning forensic bioinformation varies widely. Accordingly it is
legitimate to ask how seriously public engagement is taken in this area of
scientific development “in particular, what is the relationship between this broad
rhetoric and institutional practice?”276
5.73.
It is not obvious that the consultation: “Keeping the Right People on the DNA
Database”, seriously affected the Government’s eventual proposals. The range
of responses tended to support the argument that even when the public
respond negatively to developments, or disagree with policy suggestions, the
more centralised control over risk management, and ‘professionalisation’ of
bodies such as the police/ Forensic Regulator, to make decisions, could mean
270
Hindmarsh, R. (2008) ‘Australian biocivic concerns and governance of forensic DNA technologies:
confronting technocracy’. New Genetics and Society, 27(3) 267-284, p.278.
271
Rose, N. & C. Novas (2005) ‘Biological Citizenship’, in: A.Ong, & S.Collier (eds) Global Assemblages:
Technology, Politics and Ethics as Anthropological Problems. Oxford, Blackwell, 439-463.
272
Tony Blair, ‘Science Matters’, 10 April 2002 available at: www.number-10.uk/output/Page1715.asp .
273
House of Lords Select Committee on Science and Technology, (2000) ‘Science and Society’, (London,
The Stationery Office), p.43.
274
see: Irwin, A. ‘The Politics of Talk: Coming to Terms with the ‘New’ Scientific Governance’ Social Studies
of Science, (2006) 36 (2) 299-320, p.303.
275
House of Lords Select Committee on the Constitution, ‘Surveillance: Citizens and the State’, 2nd Report of
Session 2008-09, February 2009, The Stationary Office, London. para. 399.
276
op.cit. n.274., p310.
99
that: “the possibilities for public challenge to the dominant institutional
framework of risk management will be constrained.”277 As Hindmarsh explains;
“... for effective participatory transitions or civic trajectories that are not
simply add-on components... inclusive participatory approaches need to
address the overarching institutional context of the socio-technical system
under investigation...”278
5.74.
There have been attempts to engage with the public around issues of DNA etc.,
with the HGC Citizen’s inquiry279 and the Trustguide Report.280 Both held focus
groups around UK drawn from general public, discussing issues of forensic
DNA use and retention, and public trust in technology and data protection. The
HGC inquiry led to 29 core recommendations for the HGC to consider in their
final report. Recommendations from citizens included a nationwide public
awareness campaign and substantive proposals about DNA retention, rules on
collecting samples, the governance of the NDNAD and other issues. The HGC
argued persuasively that there is a need for in-depth engagement with public at
this level, and less reliance on instant ‘opinion polls’ which do not give people
time to think about or absorb information required to understand issues.
5.75.
The Trustguide Report reached similar conclusions on the need to engage at a
deeper level with the public. They found that many citizens did not believe the
government’s reasons for needing to keep biometric data (though not
specifically relating to forensic purposes). Focus group members also tended
not to trust the Government’s reported reasons for greater surveillance and
data collection and felt that DNA collection by the government (not the police)
was the most unacceptable form of data collection. There was apprehension
about ‘function creep’ with the retention of biometric data in particular,281
apprehension that the European Court of Human Rights considers ‘legitimate’:
“The Court maintains its view that an individual's concern about the
possible future use of private information retained by the authorities is
legitimate and relevant to a determination of the issue of whether there
has been an interference. Indeed, bearing in mind the rapid pace of
developments in the field of genetics and information technology, the
Court cannot discount the possibility that in the future the private-life
interests bound up with genetic information may be adversely affected in
novel ways or in a manner which cannot be anticipated with precision
today.282
277
op.cit. n.274., p307.
op.cit. n.270., p.280.
279
Human Genetics Commission, ‘A Citizens’ Inquiry into the Forensic Use of DNA and the National DNA
Database: Citizens Report’. July 2008.
280
Lacohee, H., S. Crane, and A. Phippen, ‘Trustguide: Final Report’, (Trustguide, October 2006).
281
supra.
282
S & Marper v UK, [2008] para.71.
278
100
5.76.
Bigger political issues (particularly the creation of a national ID card) may be
causing a shift in public opinion regarding the capture and retention of personal
information. In addition, the use of predictive profiling (mostly in the US), is
propagating the idea that certain amount of data could, and will be used against
the provider. A series of records that could lead to refusal of rights/ surveillance
etc. could make people increasingly nervous about linking of databases and the
use of databases.283 The Royal Academy of Engineers argues that ‘reciprocity’
is essential to establishing and maintaining trust in both public and private
sector data collection schemes:
“Data collection and use systems should be designed so that there is
reciprocity between data subjects and owners of the system. This
includes transparency about the kinds of data collected and the uses
intended for it; and data subjects having the right to receive clear
explanations and justifications for data requests.”284
However, with regard to forensic bioinformation databases, there cannot
realistically be such a degree of reciprocity at the level of the individual.
Although some of these requirements should be expected – such as clear
information upon bioinformation collection. It may be however that demands
could be made for this level of reciprocity at the level of the ‘public’.
5.77.
The American Public Health Association Code of Ethics stresses that the
effectiveness of institutions is reliant upon public trust and reciprocity, and that
trust can be dependent upon actions on the part of the institution, including:
“...communication; truth telling; transparency; (i.e. not concealing information);
accountability; reliability; and reciprocity. One critical form of reciprocity and
communication is listening to as well as speaking with the community.”285 The
Standing Committee for Youth Justice have warned however, that the issue of
‘public confidence’ in the NDNAD is complex. Some of the reasons for public
distrust involve:
“lack of trust in sampling and retention procedures, concerns about how
this highly personal data may be used, and the public recognition of the
lack of tangible evidence informing how decisions around the NDNAD are
made. Evidence of the disproportionate use of the database to record
information about certain groups in the population, such as individuals
from black and ethnic minority backgrounds, is also of serious
concern.”286
283
There are concerns that linking datasets diminishes privacy as information on individuals becomes
centralised and more accessible. A debate in medical research is whether more protections need to be in
place to prevent biobanks being accessed by the police – akin to the certificates of confidentiality in the USA.
284
Royal Academy of Engineering, ‘Dilemmas of Privacy and Surveillance’ (RAE, London, March 2007) p.9.
285
American Public Health Association, ‘Code of Ethics’, available at: www.apha.org/codeofethics/ethics.htm
286
Standing Committee for Youth Justice, Response to ‘Keeping the Right People on the DNA Database’,
August 2009, para.12.
101
5.78.
Underlying such concerns is the wider question of public understanding of the
police uses of forensic bioinformation (including the commissioning of research
on samples and profiles collected by the police in support of criminal
investigations). Thomas and Walport stress that the utility of databases is
lessened if the public do not trust the organisations handling the data:
“Only when people better understand what happens to their personal
information will they invest more trust in the organisations that process it.
And only when levels of trust are suitably high will organisations be able
to take full advantage of the potential benefits offered by the use of
personal information....”287
Such issues underlie calls for an ‘independent body’ to administer the NDNAD,
and several responses to the Government consultation called for the retention of
profiles to be subject to the scrutiny of an independent body reporting directly to
Parliament, bypassing even the Executive and Ministers.
5.79.
There remain serious shortcomings in the ways in which ‘forensic’ matters are
communicated to the general public. Given the importance of public confidence
to the success of policing, more work needs to be done on developing public
communication and consultation. This issue has most recently been taken up
by the UK Statistics Authority, in their current examination of why crime
statistics are not trusted or believed by the public, with a preliminary focus
being on the communication of such statistics to the public.288 There are further
complications caused by the misrepresentation of forensic science by media
outlets. Where there is irresponsible reporting, there is a false impression of
what forensic science can achieve and how complex issues like error rates
should be understood. As O’Neill states, transparency is only a ‘good’ if the
information can be made sense of, and simply demanding more information
may not increase trust:
“increasing transparency can produce a flood of unsorted information and
misinformation that provides little but confusion unless it can be sorted
and assessed. It may add to uncertainty rather than to trust. And unless
the individuals and institutions who sort, process and assess information
are themselves already trusted, there is little reason to think that
transparency and openness are going to increase trust. Transparency
can encourage people to be less honest, so increasing deception and
reducing reasons for trust: those who know that everything they say or
write is to be made public, may massage the truth.”289
287
Thomas, R. & M. Walport ‘Data Sharing Review’ (July 2008).p.56.
UK Statistics Authority, ‘Overcoming barriers to trust in crime statistics’, (December 2009).
289
O’Neill, Onora, ‘The Reith Lectures 2002 – A Question of Trust.’ Lecture Four, available at:
www.bbc.co.uk/radio4/reith2002
288
102
It is essential that information is from a reliable source that can be checked:
“global transparency and complete openness are not the best ways to
build or restore trust. We place and refuse trust not because we have
torrents of information (more is not always better), but because we can
trace specific bits of information and specific undertakings to particular
sources on whose veracity and reliability we can run some checks.”290
5.80.
If provided with accurate, reliable, and ‘checkable’ information, the public can
then be suitably ‘informed’ to consent: “informed consent can provide a basis
for trust provided that those who are to consent are not offered a flood of
uncheckable information, but rather information on whose accuracy they can
check and assess for themselves. This is demanding.”291 Both the Royal
Academy of Engineers and the Human Genetics Commission have called for
greater transparency and provision of information to the public:
“data collection and use systems should be designed so that there is
reciprocity between data subjects and owners of the system. This
includes transparency about the kinds of data collected and the uses
intended for it; and data subjects having the right to receive clear
explanations and justifications for data requests.”292
“A condition of public support and informed debate is that there should be
sufficient, reliable information available. Openness (access to information)
and transparency with regard to how the information is produced, as well
as the amenability of information to non-specialist understanding and
appropriate support for the public’s ability to understand and interpret the
information, are all important.”293
5.81.
There is then a need for a proper communications strategy to dispel myths, and
inform and educate ministers; judges; lawyers; and the public, emanating from
reliable sources. This should include key processes/ technologies etc. with
clear and agreed upon definitions and explanations so as to ensure clarity. For
example, all parties will need to agree on what ‘deletion’ or ‘retention’ means in
practice, what is ‘familial searching’ etc. (to be understandable by public). Only
once the public and stakeholders are all properly informed, and speaking in the
same language, will progress be possible toward effective governance that
retains public confidence. It may be useful to consider how civil society groups
may be informed of strategic developments in forensic science in general and
the uses of forensic bioinformation in particular at a relatively early stage – and
whether they may have a stronger voice in discussions surrounding these
developments. As Irwin states, such ‘openness’: “is not intended to block
290
ibid., Lecture Two.
ibid.
292
Royal Academy of Engineering, ‘Dilemmas of Privacy and Surveillance’ (RAE, London, March 2007) p.9.
293
Human Genetics Commission, ‘Nothing to Hide, Nothing to Fear?’ (2009) Para 6.5.
291
103
scientific progress, but instead create a more open and reflective culture where
new scientific possibilities can be fully realised.”294
5.82.
Magee stresses that positive attitudes to biometric data for use in public
protection are vital: “Ministers need to lead a public debate to help improve
public understanding and confidence.”295 There is also need for ongoing
engagement with statisticians to work on how forensic science may be clearly
explained to juries and legal professionals with minimum risk of confusion.
There also needs to be absolute clarity about the potential uses of forensic
bioinformation to prevent unauthorised use or ‘mission creep’. This can occur
most often where there is confusion or initial lack of clarity over the intended
primary functions of the resource. Not that all ‘mission creep’ should be
considered negatively:
‘Function creep is often beneficial and has been a means to real progress
in many areas of human endeavour. However, where the potential exists
for function creep to lead to unwelcome consequences, there is a need
for appropriate safeguards. Two safeguards can be applied: the clear and
precise definition of the proper function; and effective regulation of use
(the second being ineffective without the first).’296
What may be required is further public and parliamentary debate on the
legitimate uses of forensic bioinformation, for example, should the NDNAD be
used to find or identify ‘missing’ persons?
Integration
5.83.
As both the Flanagan review and Magee report have stressed, policing takes
place within a network of actors and agencies: “Policing is not simply the
preserve of the police. Modern policing is carried out in partnership with a wide
range of local agencies, from councils to primary care trusts to schools”.297 The
use of forensic bioinformation, like any police information, needs to be
integrated successfully into such networks, to ensure its effective use. There is
also a need to demonstrate that forensic bioinformation is being effectively
integrated both within policing domestically; and international systems for crime
control; surveillance (intelligence gathering); and movement of persons.
5.84.
Despite the importance of such integration, there is no clear, or explicit
guidance on how forensic bioinformation is successfully, (or should be)
integrated into wider policies and practices in policing and criminal justice. How
should such information be integrated with other sources/ types of information
294
op.cit. n.274, p308.
Magee ‘The Review of Criminality Information’ (Magee Report, July 2008) Home Office, London,p.88.
296
Human Genetics Commission ‘Nothing to Hide, Nothing to Fear’ (London, 2009), para 5.19
297
Flanagan Report ‘The Review of Policing – Final Report’ (February 2008) Home Office, p.7.
295
104
regularly relied upon by the police, courts, and other law enforcement agencies?
The Home Office/ACPO strategy of 2004,298 replacing the DNA Expansion
Programme, made a commitment to a more integrated approach to forensic
science, particularly in improving the effectiveness of fingerprints. Yet it is
difficult to find evidence of any strategy being implemented in practice. The use
of fingerprints in particular is becoming increasingly widespread, across several
agencies, some that rely upon their own databases, others utilising IDENT1 (i.e.
UK Borders Agency et al). ‘Sharing’ of IDENT1 is expected to increase, with the
NDNAD the sole preserve of the police, although we have noted in Chapter 4,
the emergence of DNA as a border or immigration control measure and UKBA’s
interest in whether it might assist them to determine asylum applications.
5.85.
The Bichard Report of 2004 was critical of the ability of police forces to share
information on individuals and recommended that the police nationally improve
the management and sharing of information and intelligence at both national
and local levels. The IMPACT programme responded to this recommendation,
creating a statutory Code of Practice on the Management of Police Information
(MoPI) in 2005, and the IMPACT Nominal Index (INI), which will allow
authorised users in one force to quickly identify which other forces hold
information on a person of interest. These will both be supported by the Police
National Database (PND) which will provide a national intelligence sharing
system. There is a draft Code of Practice on the operation of the PND, which
aims to “promote consistent and lawful use of the PND across the Police
Service.” All forces must comply with the MoPI Code of Practice by 2010, in
order to assist with the rollout of the PND, delayed until 2011.299
5.86.
The Magee Report praised the IMPACT programme stating that “the framework
drawn up by the police, the MoPI in 2006, is a good example of efforts
to….implement a structure to facilitate the sharing of criminality information.”300
While praising these efforts, Magee recalls that there has been significant
investment (over £2bn) in IT schemes across the criminal justice sector since
2001, with some benefits for individual agencies, but connections between
these systems so that they can ‘communicate’ have been an ‘afterthought’.301
The IMPACT programme has also been like many government IT programmes,
in that it has had troubles “with delays, funding problems, cost overruns and
delivering fewer business benefits than originally envisaged.”302 There have
also been issues regarding accountability, which has not been a strong feature
of major government IT programmes.
298
Home Office (2004) Confident Communities in a Secure Britain: the Home Office Strategic Plan 2004-08
(London: The Stationery Office) p.73
299
NPIA Website accessed 16 December 2009
300
Magee ‘The Review of Criminality Information’ (Magee Report, July 2008) Home Office, London p.41
301
ibid., p.70
302
ibid., p.69
105
5.87.
Recent changes in the organisation of key government departments will impact
on policy and practice in forensic science support to policing. Key to these will
be the emerging role of NPIA. Existing networks of influence involving the
Home Office, ACPO, and as a government owned company, the Forensic
Science Service (FSS) and others are likely to be modified by the presence of
the NPIA as an agency with a role in both policy making and practice
improvement. It will be important for NPIA to effectively integrate the work and
interests of a number of scientific, operational and other communities. The bulk
of the NPIA budget is spent on the national delivery of services; accordingly the
sum available to support the development of strategy and the assessment of
potential innovations remains small (and most likely shrinking). Policing and
priorities in developing scientific applications need to be subject to ongoing
ethical scrutiny, although it is not clear what body exists – or should exist – to
assist in such scrutiny. It will be important for the NPIA to effectively integrate
the work and interests of a number of scientific, operational and other
communities as it seeks to deliver the several work streams of Forensics21.
5.88.
The number of databases and the organisation of the police and other relevant
agencies has leant itself to duplication, with SOCA being a prime example.
SOCA inherited over 350 databases from predecessors upon its creation (they
are hoping to reduce this to between 50 and 60).303 The attempt to create an
explosion on a Delta Airways flight on Christmas Day 2009 also highlighted the
issue of potentially counter-productive duplicity of databases with numerous,
vast ‘watch-lists’ in existence around the world with minimal communication.304
The proliferation of data has limited efficient and effective exchange of
information between databases, with technological issues often overwhelming.
Just domestically, with 43 police forces across the country, this is exacerbated
by sheer volume of data, with the police of England and Wales estimated to
hold over 70 million operational records across 350 different systems.305
5.89.
The UK experience has shown that the expansion of DNA retention enabled by
legislation in 2001 and 2003 also created the necessity for keeping ‘criminal’
records for those not convicted of a criminal offence. Changes in the DNA and
fingerprint retention regime will require the consideration of the necessity for the
parallel removal of ‘criminal’ records of those whose DNA profiles and
fingerprints are to be destroyed although there is no intention (presently
expressed) to return to the ‘weeding’ that was previous ACPO policy. The move
to retain all arrest information indefinitely has been supported by the ‘Five Chief
Constables’ ruling which found indefinite retention of all PNC records lawful.
This emphasises the need for more widespread consideration of the retention
of all criminal justice information and what should be the relationship between
bioinformation databases and the criminal justice process and public policy.
303
ibid., p.71
‘World terror watch lists flawed’, Washington Post (Online), 9 January 2010.
305
Magee ‘The Review of Criminality Information’ (Magee Report, July 2008) Home Office, London p.42.
304
106
5.90.
Magee called for a ‘Public Protection Network’ (PPN) encompassing an array of
agencies, moving from the idea of ‘police’ data; ‘prisons’ data; ‘immigration’
data etc. and promoting the sharing of ‘criminality’ information under the
umbrella of ‘public protection’ data.306 He calls upon the Ministry of Justice to
lead in the area of criminality information, with its portfolio of prisons, probation,
courts and criminal justice IT, making it ideally placed for this role.307 With a
PPN, Magee believes that there could be strong oversight and governance,
permitting the productive exchange of information with connectivity across the
various agencies and government departments who deal with ‘criminality’
information. With strong governance, the public could have confidence in the
PPN, and believe “…that action is taken is proportionate to the risks being
addressed, that there are sufficient checks and balances in place and that
governance arrangements will ensure high standards.308 This could go some
way to meeting the request by the Joint Committee on Human Rights, that the
government show: “…that any proposal for data sharing is both justifiable and
proportionate, and that appropriate safeguards are in place to ensure that
personal data is not disclosed arbitrarily but only in circumstances where it is
proportionate to do so.”309
5.91.
Magee argued that it is critical that the public have confidence that the PPN is
“…collaborating to meet the new challenges posed by advances in criminality
information and the problems and opportunities presented by international
information.”310 As the ‘Data Sharing Review’ by Thomas and Walport (2008)
highlighted, there are risks in sharing and also not sharing information: “There
are symmetrical risks associated with data sharing – in some circumstances it
may cause harm to share data, but in other circumstances harm may be
caused by a failure to share data.”311 They also stressed the need for good
governance: “It is equally important that such decisions [about sharing personal
data] are taken in the context of good mechanisms of governance including
transparency, audit and accountability.”312
Conclusion
5.92.
Public trust is an essential precondition for the effective use of forensic
bioinformation. The government need trust to enable ‘consensus’ legislation.
The police need trust in order to utilise the technologies and only trust can allay
suspicions of ‘Big Brother’ futures. With trust in the institutions responsible for
collecting, using, and governing forensic bioinformation, individuals and
306
ibid., p.90
ibid., p.70
308
ibid., p.5.
309
Joint Committee on Human Rights, ‘Data Protection and Human Rights Report’ (Fourteenth Report of
Session 2007-08) para.14.
310
Magee ‘The Review of Criminality Information’ (Magee Report, July 2008) Home Office, London, p.5.
311
Thomas, R. & M. Walport ‘Data Sharing Review’ (July 2008).p.i.
312
ibid., p.ii.
307
107
communities can gain the benefits of these technologies yet still know that
respect for human rights and the democratic process remain unchallenged.
These are not simply matters of technology and science:
“If we are to design and develop trusted technologies we need to
understand the complex inter-relationship between trust, confidence,
control and security. The first issue that arises here is the need to
understand that this is not simply a technological problem that can be
solved in isolation...”313
5.93.
Securing and maintaining trust in any institutional arrangements requires clear
lines of accountability and the possibility of appropriate levels of independent
oversight. Sufficient information must be available to enable relevant publics to
give support and consent. This information needs to emanate from reliable
sources, and be ‘checkable’, and therefore available to external researchers.
“...public confidence is reflected in government policy through legislation
and budget. Privacy debates and lingering civil liberty concerns can erode
public confidence, and replacement of misinformation with factual DNA
information is essential.”314
5.94.
Trust in the operation of forensic databases is especially sensitive to the
provision of security and adequate data protection: ‘While it may take years of
effective governance to establish institutional trust, it can be wiped out very
quickly, however fairly or unfairly, by high profile mistakes or accidents.”315
There cannot be room for failure as this will be followed by a catastrophic loss
of confidence in those who manage such data as well as in the management
procedures themselves. Clarity of purpose and aims secured through proven
quality standards and the oversight of such standards will also contribute to the
healthy regard necessary for these technological innovations to be used for the
achievement of the public good.
313
Lacohee, H., S. Crane, and A. Phippen, ‘Trustguide: Final Report’, (Trustguide, October 2006).p84.
INTERPOL. ‘Handbook on DNA Data Exchange and Practice’ 2nd edn. (2009) available at
www.interpol.int p.49.
315
Royal Academy of Engineering, ‘Dilemmas of Privacy and Surveillance’ (RAE, London, March 2007)
p.44.
314
108
Summary and A Proposal
Beyond S & Marper
6.1.
England & Wales is clearly at a crossroads, with significant decisions to be
made regarding the direction now to be taken about the collection, retention
and use of forensic bioinformation. Recent debate remains dominated by the
exchange of bold claims made for the usefulness of forensic bioinformation
(e.g. “The NDNAD continues to provide the police with the most effective tool
for the prevention and detection of crime since the development of fingerprint
analysis over 100 years ago.”316) rather than by the careful analysis of robust
data. A further limitation on the ability to conduct the much demanded
‘informed debate’ is that the potential for, and limitations of, the use of
bioinformation globally in law enforcement and related activities have been
inadequately discerned.
6.2.
After years of developments in forensic bioinformation occurring away from
public gaze, with minimal parliamentary debate or media attention, the ruling at
the European Court of Human Rights in S & Marper has thrown a spotlight on
forensic bioinformation databasing. While the NDNAD was increasingly
featuring in studies of ‘surveillance’,317 and also had been the subject of reports
by the Nuffield Council on Bioethics, the Human Genetics Commission, and
numerous papers by Genewatch, Liberty, Justice et al, it has now received
renewed and highly focused government attention. Since the ruling in
December 2008 there have been two government Bills containing proposals on
forensic bioinformation preceded by two consultation papers. The Home Affairs
Select Committee also conducted a brief inquiry into the National DNA
Database (although they did not mention IDENT1 in their inquiry).
6.3.
This period of consultation and deliberation, has ceased for now, with the
legislative response of the UK Government apparently dealt with in the ‘washup’ period of Parliament before the general election, with the passing of the
Crime and Security Act 2010. However, it is clear that the ECtHR judgment will
have ongoing ramifications, with its requirement to radically reshape the
forensic bioinformation regime of England & Wales, and will also serve to shape
emerging regimes elsewhere in Europe. However, the debate has focussed
upon the ‘retention’ of forensic bioinformation, since the court in Strasbourg
ruled that UK police powers to retain bioinformation breached human rights.318
There is perhaps then a lost opportunity if the debate cannot move beyond this
one narrow issue of retention (and the retention of the bioinformation of
‘innocent’ individuals at that).
316
Foreword, CC Neyroud, NDNAD Annual Report 2007-09, p.4.
i.e. Home Affairs Committee 5th Report of Session 2007-08 ‘A Surveillance Society? HC 58-1 (The
Stationery Office, London, June 2008) ; House of Lords Select Committee on the Constitution: ‘Surveillance:
Citizens and the State’ HL Paper 18-1 (The Stationery Office, London, February 2009).
318
In particular, the right to privacy found in Article 8.
317
109
6.4.
The European Court of Human Rights, in reaching their unanimous decision,
were scathing of the UK’s ‘indiscriminate and blanket regime’ of retention,
stating that the government needed (and had failed) to provide ‘weighty
reasons’ for their policies and practices. They also stated that the UK bore a
‘special responsibility’ as a country at the vanguard of forensic bioinformation
use. Most other EU countries have not followed England, Wales and N. Ireland
in implementing such an extensive retention regime. The initial response of the
UK Government to the judgment was met with derision (not assisted by the
scandalously poor quality ‘evidence base’) and their initial proposals for
changes were dropped from the Policing and Crime Act 2009. Amended
proposals were brought forward in primary legislation - the Crime and Security
Bill - in the dying days of parliament before breaking for a general election.
6.5.
The issue dominating debate and government efforts since the Marper
judgment is the ‘blanket and indiscriminate’ retention regime for DNA and
fingerprints. In both the ruling and subsequent debate, comparison with the
Scottish retention regime has been a constant feature. This might reflect views
about a procedural difference (retention in the absence of conviction only, in
effect, after case by case scrutiny by a procurator fiscal in Scotland,)319 as well
as the variation between outcomes following arrest that trigger retention in
England and Wales compared with Scotland. The government are clear in their
intention to avoid individual case by case decisions, except possibly in ‘terrorist’
cases where a Chief Constable can authorise ongoing retention ‘in national
security interests’ for two years, which is then renewable.
6.6.
The Government have attempted what they have designated a ‘scientific’
approach to retention in order to demonstrate the ‘utility’ of retaining
bioinformation in crime detection, rather than an ethical, or legalistic, or human
rights approach (for instance, relying on ethical, legal, or human rights based
tenets or arguments). The difficulty is that the evidence produced for this
‘scientific’ approach has failed expert scrutiny and so has not been able to
provide a base upon which to build proposals.
6.7.
The Council of Europe Committee of Ministers (which monitors the responses
to ECtHR judgments and sees that they have effect), raised a series of
questions regarding the governments’ proposals in the Policing and Crime
Bill.320 In their assessment of the provisions set out in the Bill, they asked
whether the proposed retention regime was proportionate and struck a fair
balance between public and private interests. They specifically pointed to the
proposal to retain DNA from arrestees for non-serious offences, stating that this
did not conform to the requirement of proportionality. They also stated that the
319
See Section 18A (1) of the Criminal Procedure (Scotland) Act 1995 (as amended): ‘that criminal
proceedings ….were instituted…..’.
320
The response to S & Marper vs UK was considered at the 1065th Meeting, September 2009.
110
provisions for children and adults were too similar and did not respond to the
requirements of the judgment in S & Marper. Similarly, the continuation of the
mechanism for the destruction of profiles did not respond to the Court’s wish for
an independent review mechanism and was an ineffective response. The
Committee also criticised the ‘evidence’ used to inform the proposals, stating:
“Given the UK’s claimed ‘pioneer role’, reliance only on academic studies,
two of which do not relate to the United Kingdom and an approach to
those studies which appears to rest on the principle that unconvicted
individuals will commit criminal offences, do not appear sufficient to justify
retention periods which do not appear to be in conformity with the Court’s
judgment.”
6.8.
In W v The Netherlands,321 the ECtHR held that retention of DNA for convicted
persons was acceptable where it was retained “for a prescribed period of time
dependent on the length of the statutory maximum sentence that can be
imposed for the type of offence committed.” This was in direct contrast to the
approach of the UK government, which “does not consider retention on the
basis of any link with the maximum sentence but rather on the possibility of
future offending.” Indefinite retention of DNA from convicted offenders does not
itself have unqualified support, with many questioning retention in cases which
are very minor, or in which DNA plays no part. The Equalities and Human
Rights Commission have called into question the legality and usefulness of
retention in some cases, i.e. a conviction for a minor offence (perhaps single
offence) of insurance fraud.
6.9.
In a subsequent meeting in December 2009, the Committee of Ministers
referred to the revised proposals in the Crime and Security Bill, but still noted
that “a number of important questions remain as to how the revised proposals
take into account certain factors held by the European Court to be of relevance
for assessing the proportionality of the interference with private life here at
issue, most importantly the gravity of the offence with which the individual was
originally suspected, and the interests in deriving from the presumption of
innocence.” They also queried where further proposals were in relation to “the
institution of an independent review of the justification for retention in individual
cases.”322
6.10.
It should be possible to develop a retention regime that is Article 8 compliant.
Some technological protections may afford a more ‘privacy friendly’ regime and
such (albeit partial) solutions should be considered. It must also be borne in
mind that any changes resulting from the Marper judgment will impact on the
private organisations that hold samples on behalf of Chief Constables (the
321
322
W vs The Netherlands (application No. 20689/08, decision of 20/01/2009)
1072nd (DH) Meeting, December 2009. Decisions Adopted at the Meeting. Section 4.2.
111
forensic service providers) and methods will have to be put in place to ensure
compliance with agreed understandings. The regulations regarding the uses
made by private organisations of such samples, along with their holding of
identifying information need to be reconsidered (along with the bioinformation
data held by other agencies such as the Crown Prosecution Service (CPS). It is
also important that any revised retention regime keeps in mind the significance
of DNA and fingerprint records for the successful investigation of crime,
especially those crimes – like violent offences and serious sexual assaults - in
which biological evidence can be central to an investigation and prosecution.
Consideration should be given to collecting and retaining the best possible data
on uses of the NDNAD after any legislative changes. Only in this way will it be
possible to measure the effect of the changes and monitor satisfaction with the
resulting regimes in case there is demand for further changes, or indeed future
challenges to the legality of forensic bioinformation use.
6.11.
The issue of requests for removal has been the focus of the Home Affairs
inquiry, prompted by Damian Green MP’s arguments that a ‘postcode’ lottery
exists depending upon which force ‘owns’ the DNA and their rates of acceding
to requests. The NDNAD Ethics Group has repeatedly drawn attention to their
dissatisfaction with the ‘removal’ process, preferring: “an independent, statutory
appeals procedure in order to provide an effective remedy”.323 Without such a
process, they argue that there is a risk that the process is; “inconsistent and
discriminatory…In particular there is no element of independence that is at the
heart of a requirement to provide an effective remedy under the ECHR”.324
Liberty go further in questioning: “why a person must make out ‘exceptional
circumstances’ before deletion can be considered. This test bears no
relationship to the tests of necessity and proportionality found in the Human
Rights Act or to the Data Protection Principles….”325
6.12.
There is still then no satisfactory consensus on how best to approach questions
on the use and retention of forensic bioinformation. The preference in England
and Wales has been to begin by focussing on police powers and on questions
of ‘utility’, a preference which has generated piecemeal legislation largely
uninformed by more general considerations. An alternative is to begin by
considering principled questions of human rights. The European judges have
reminded us of the necessity of justifying the police retention of bioinformation
which in many circumstances can be seen as a breach of privacy, but
acknowledged there were differences between the three categories of personal
data that would require careful scrutiny although their ruling did not consider in
detail the differences between DNA profiles, samples and fingerprints.
323
NDNAD Ethics Group Response to “Keeping the Right People on the DNA Database” August 2009, p.7.
ibid. p.8.
325
Liberty, Response to “Keeping the Right People on the DNA Database” August 2009, p.29.
324
112
6.13.
In determining whose information (including bioinformation) should be retained
when the activity giving rise to an arrest has not been subject to judicial
deliberation, it may be necessary to improve the quality of information about the
circumstances of any arrest and charge as well as the reasons for failing to
proceed with a case, or failing to obtain a successful prosecution of a suspect
offender. The use of arrest as the ‘trigger’ for taking bioinformation could be the
critical issue (rather than subsequent retention). Indeed, in their response to the
government’s consultation, the NDNAD Ethics Group states that:
“The use of arrest as the trigger for retention of those unconvicted
encompasses many people who are not subsequently either charged or
proceeded against…. We seriously question whether this is a sufficient
basis for justifying retention. We therefore recommend that further
research needs to take account of these points in order to reach a better
understanding of the proportionality of appropriate ‘triggers’326
6.14.
Soothill and Francis explain why the arrest ‘trigger’ may be inappropriate, but
more importantly, why ‘arrest hazard rates’ are not satisfactory indicators of
offending behaviour, and therefore a base on which to make predictions of
future offending (and therefore justify retention):
“While police arrests are not whimsical, they come at the
beginning and not the end of the criminal justice process. Some people
are disproportionately at risk of being taken into questioning by the police
and being arrested. In contrast, a conviction is the outcome of the
evidence being tested in court. In fact, arrests are useful indicators of
police action, but not of guilt. Re-arrests are dangerous indicators and
making arrests the pivotal criterion encourages the notion that we are
moving towards becoming a police state.”327
6.15.
326
327
It may be true that, particularly when dealing with young offenders, a swift and
appropriate sanction may be preferable to waiting for a court disposal, indeed
significant efforts have been made to increase non-court disposals to avoid
lengthy delays and this may have beneficial impacts on both the court system
and the offender, suggesting that dependence upon a court to decide upon
‘guilt’ may not always be ideal. Better knowledge of offending patterns as well
as potential ‘dangerousness’ needs to be subject to consideration if something
more discerning than ‘blanket retention’ is to be acceptable. In the absence of
such data, ‘blanket retention’ may continue to be seen as providing the best
opportunities for the identification of offenders through fingerprints and DNA
profiles, and may support calls for a ‘universal’ database.
op.cit., n.326. p.8.
Soothill, K. & Francis, B. ‘Keeping the DNA Link’ (17 July 2009), New Law Journal, 1021-1022: 1021.
113
6.16.
New considerations arise when the collection and retention of forensic
bioinformation is viewed from the perspective of crime prevention rather than
criminal prosecution. The traditional repertoire of checks and balances function
largely in relation to prosecuting accused persons within the context of criminal
trials. However, the modality of crime prevention is informed by a different set
of considerations about checks and balances, and this is not always realised.
Without such consideration there is a real danger of function creep, and it is
easy to lose transparency in the proprietary – and legality - of such uses.
6.17.
In recent debates there has been a tendency to attribute the sustained drop in
crime since 1997 in part, to the use of forensic DNA. This in part has intimated
that the DNA database has some ‘deterrent’ effect which has led to a decrease
in offending. However, there is no evidence of the deterrent value of the
NDNAD. While it is feasible, if difficult, to assess the utility of DNA in
investigations and prosecutions, measuring any ‘deterrent’ effect would be
extraordinarily difficult, if not impossible. Arguments therefore that the retention
of DNA can ‘deter’ must necessarily always fail without any possibility of
evidence to support such claims.
6.18.
Indeed, sociological theories of the power of ‘labelling’328 may be relevant to
debates around bioinformation retention, which would lead to diametrically
opposed conclusions regarding the ‘deterrent’ effect of retention. Labelling
theory would suggest that by attaching to an individual labels such as ‘future
suspect’ or ‘potential criminal’, as is arguably achieved by the Government’s
current policy of assessing the ‘risk’ posed by arrestees as warranting retention
for six years, you actually increase the likelihood that the individual will ‘live up’
(or in this case ‘down’) to the label. According individuals (particularly young
people) status as a future offender creates the perfect conditions for these
individuals to fulfil these prophecies (the ‘self-fulfilling prophecy’ a well
established phenomenon in psychology).
6.19.
It could be argued, with the support of such theories, that by treating people as
‘future’ criminals (or ‘pre-suspects’329), you might make it more likely that this
will become true. Newspaper reports have already highlighted an instance
where the presence of a DNA entry on the PNC, brought to a police officers’
attention through mobile checking of a driver’s identity while on patrol, led to
differential treatment and increased suspicion.330 It cannot be discounted that
328
There is an extensive literature on this approach to criminology that addresses the issues, inter alia, of
the subjectivity of the processes that are sometimes used to identify deviant behaviour, and the
consequences of this for both society and the labelled individual. Key texts include, Becker H.S. (1963)
Outsiders, Cohen, S.(1972) Folk Devils and Moral Panics and Plummer K (1979), ‘Misunderstanding
Labelling Perspectives’ in Downes, D. And Rock, P. (eds.) Deviant Interpretations.
329
Lynch, M., Cole, S., McNally, R. And Jordan, K. (2008) Truth Machine: The Contentious History of DNA
Fingerprinting (Chicago and London, The University of Chicago Press), p. 152.
330
‘DNA Database – Head to Head’, BBC Website, available at:
http://news.bbc.co.uk/2/hi/uk_news/8354740.stm accessed 25 April 2010.
114
the behaviour of others alters when interacting with an individual who has their
details retained on the PNC, retention that will be necessary due to the
requirements of the bioinformation retention regime. It is already clear that such
retention of details on the PNC has an adverse impact on visa applications, and
potential employment.
6.20.
It is necessary to distinguish between efforts to regulate the retention of DNA
profiles and fingerprints taken from known individuals and those collected
during the search of crime scenes. It generally seems to be the case that there
is no restriction on the retention of the latter kind of information regardless of
the apprehension or prosecution of criminal suspects, albeit that in England and
Wales, these data are removed from databases (a situation that should also be
addressed but currently does not feature in debates).
6.21.
There are also significant linguistic issues at stake when questions of forensic
bioinformation (and other kinds of police-relevant information) retention are
raised. There are problems in translating key terms across jurisdictions, and
there are also issues surrounding the uses of polyvalent terms like ‘innocent’
when political arguments take place. Decisions about the proper balance
between liberty and public protection need to take into account the differing
meanings attributed to words like ‘innocent’ and ‘serious crime’ for example, by
a range of police and other actors. This linguistic and definitional aspect is
particularly relevant as the definition of a ‘serious crime’ can be subjective.
Whilst some may relate this term to offences of murder, rape, robbery etc, a
‘minor’ domestic violence case may be reported after years of abuse and may
be the precursor to more serious domestic violence if not treated as a ‘serious
offence’ from the outset.
6.22.
The comment by ECtHR about the ‘blanket and indiscriminate’ nature of the
retention regime need not necessarily result in the rule base approach
advocated in Keeping the Right People on The DNA Database, which proposed
that DNA profiles should be retained automatically in a narrower range of
circumstances and for a shorter time period than at present. This has been
presented as similar to Scots law with simply a wider range of circumstances
that might give rise to retention (and a longer retention period). Crucially this
misses the really distinguishing feature of the Scottish arrangements: the case
by case scrutiny by a Procurator Fiscal and the scope for judicial supervision.
This analysis is reinforced by the information about a more selective sampling
policy followed by the police themselves. During the initial seminar participants
were reminded that the Nuffield Council on Bioethics had been advised during
the preparation of its report that the scrutiny of individual decisions was
unaffordable. Further discussion elicited comments that because in Scotland a
smaller number of profiles are processed, it is easier to make case by case
decisions about retention. The comparatively lower levels of recorded crime,
115
higher clear-up rates and the more generous number of police officers in
Scotland (per 100,000 of population) supports this observation.
6.23.
During the Commons Committee stage of the Crime and Security Bill, ACPO
stated that retention proposals based on the Scottish arrangements should be
rejected for England and Wales for reasons of additional cost. Instead, ACPO
argued for simple retention rules based on offence categories with deletion via
computer algorithm, enabling the whole process to be managed by algorithm.
The Scottish system, with its emphasis on an examination of individual
circumstances, ‘was rejected because it would take up an inordinate amount of
police resources’.331
6.24.
It was also suggested during the third seminar that Scotland might be able to
establish a central clearing house for such decisions, but that to put such a
system in place in England and Wales would be more difficult. This was not
followed up at the time (indeed it was a final remark at the end of a long day). It
is possible, however, that this was not raised as a financial or logistical issue,
but one relating to the notional ownership of samples or profiles. Certainly the
legislation hurriedly pressed upon Parliament prior to dissolution appears to
offer a hybrid of uncertain central influence over decision making by individual
chief constables in respect of the DNA profiles.
6.25.
These arguments fail to address the point that the real lesson from Scotland
may be that it offers a model of individuals within the criminal justice system
exercising professional judgment on a case by case basis. Such an approach
would meet concerns about both excessive retention and the problem of
formulating rules to predict potential future harm. For example, if a person was
arrested but it soon appeared that it was a case of mistaken identity or to
ensure that bioinformation was collected from a potential witness for elimination
purposes, it should be a simple matter for those handling the case to have the
profile deleted or retained temporarily as a volunteer sample. The individual
concerned should not have to initiate a deletion process and might reasonably
expect that no trace of the sample having been taken remains on the PNC. In
contrast, even when no charges were laid or proceedings were discontinued,
information about an individual may be retained for good reasons. The latter
circumstances were recognised as an important lesson from the failure to take
action that might well have prevented the double murder at Soham and is
reflected in the individual case management of information set out in the MOPI
code. Neither a hybrid of central and local decision making, nor deletion by
algorithm are acceptable substitutes for the individualised decision making at all
stages of the criminal justice system and local procuratorial or judicial review
available to some six million fellow UK citizens.
331
Public Bill Committee, 26 January 20010, Q93 and 97-100.
116
6.26.
If decisions are taken at the right time it is difficult to see how an approach
should lead to higher costs. Ironically, the exercise of greater discretion by
individual police officers was the main element in evidence given to the Public
Bill Committee by the Government’s ‘Police Bureaucracy Czar’ and the
Chairman of the Police Federation when testifying about another aspect of the
legislation now enacted as the Crime and Security Act 2010. In contrast,
ACPO’s stance appears to demonstrate insufficient concern for both legitimate
concerns about privacy and the deskilling of the police work. On the latter issue,
the Chairman of the Police Federation commented:
‘Although we are very good at following the quantitative, measuring
approaches, what matters to the public is not the quantity of what we do,
but the quality. It is no accident that we have inspectors and
superintendents in the police service. They actually used to inspect and
superintend what those under their control did. We have lost that to some
degree. The quality of what officers are doing seems to be less important
to those in command than the quantity, and we have to get away from
that.’332
6.27.
Finally, there is some evidence to support the contention made in this report
that a margin of appreciation based on no more than a survey of legislation is a
possibly more fragile basis for reaching views on proportionality than is
suggested in Marper. Scotland may have achieved a balanced approach to
retention issues that still need to be settled south of the border. This appears to
have been as much a chance outcome, as one of good professional and
political judgment. It is far from clear, however, from reading recent debates at
Westminster that what we would argue is the defining characteristic of Scots
law in this area, the exercise of individual discretion, has been fully understood.
Evidence and Assessment
6.28.
Whilst forensic science in general is a major police consumable,333 it is also
highly vulnerable to arbitrary cuts during budgetary crises.334 The changing
regime for funding expenditure on forensic science has already forced the
police to question whether existing or expanded forensic spending represents
good value, and there is not always support amongst senior levels within the
police service for the maintenance of the priority accorded to forensic science in
332
Public Bill Committee, 26 January 20010, Q 15.
Accounting for approximately half the ICT spend and slightly less than on transport: Home Office
Police Science and Technology Strategy 2008-2009 (London: Home Office, 2004) p26.
334
Often due to a lack of scientific awareness among senior police officers, see: Blakey, D.
(2000).Under the Microscope, (London: Home Office), Blakey, D. (2002) Under the Microscope
Refocused (London: Home Office) and Coleman, D. (2004) ‘Beyond DNA in the UK – The Police
Perspective’, in Townsley, M. and Laycock, G. (eds) Forensic Science Conference proceedings:
Beyond DNA in the UK – Integration and Harmonisation (London: Home Office), pp 9-10.
333
117
recent years compared with other policing budget items.335 Other agencies,
such as the Crown Prosecution Service, also face significant decisions
regarding their future requirements for the provision of forensic bioinformation,
decisions that need to be informed by a sound understanding of the
implications of ‘staged reporting’ arrangements for the production and utilisation
of forensic science results. Such decisions necessarily impact upon the whole
legal system with regard to the direction of limited financial resources.
6.29.
Expenditure on DNA and fingerprints like all other activities within the criminal
justice system should, wherever possible, be assessed for its effectiveness in
line with other budget items and subject to cost-benefit analysis, even though
such assessments are notoriously difficult to carry out.336 Failing this the police
and the NPIA should at least demonstrate that the collection and use of forensic
bioinformation is being organised optimally, that they do not divert resources
from budgetary items that are proven to be more beneficial, and that police
internal forensic organisation and procurement of external forensic services are
cost-effective. It is necessary then accurately to assess the opportunity costs of
expenditure on forensic bioinformation, its impact upon the state’s ability to offer
justice to all citizens, and its implications for the wider resourcing of the criminal
justice system.
6.30.
In summary, arguments over fundamental issues concerning the collection,
retention and use of forensic bioinformation as well as disagreements about
how key problems are to be resolved, continue to be poorly supported by a
credible evidence base on which to ground important decisions. Instead, after
an unconvincing consultation exercise, the current UK government appears to
have abandoned the idea, discussed when we began our work, of publishing a
forensic science White Paper. It has recently conceded that there is a need for
new primary legislation dealing with forensic bioinformation collection, retention
and database governance, which resulted in provisions in the Crime and
Security Act 2010.
Internationalisation & Exchange
6.31.
With increasing mechanisms, and imperatives to exchange law enforcement
data, internationally and particularly within the EU, and the extent to which
decisions about this have to comply with rules and policies negotiated with
other governments or have been influenced by bodies outside this country, an
335
Peter Neyroud, NPIA Director, 31st March 2008, in evidence to Home Affairs Select Committee
inquiry into the ‘Surveillance Society’. Available at:
http://www.parliament.uk/parliamentary_committees/home_affairs_committee.cfm .
336
The most recent work assessing value for money and the ability of economic analysis to inform
decision making within policing was highly cautious about the application of such techniques (see
Stokedale, Whitehead, & Gresham (1999) Police Research Series Paper 103 ‘Applying Economic
Evaluation to Policing Activity’ (London: Home Office), since then there has been a major increase in
data to which such techniques might be applied.
118
examination of the UK situation can no longer be undertaken without
consideration of the international context in which policing and efforts to
maintain ‘security’ are now played out. The Secretary General of Interpol
argued that DNA profiling is ‘a discovery that has benefited mostly the
wealthiest of countries’ and yet there was a clear need for greater international
cooperation using this forensic technique.337 Despite this exhortation, little has
been published about the scale of international cooperation or the effectiveness
of recent and emerging measures taken to share police information, forensic or
otherwise, across national boundaries.338 There are also pressing, but as yet
unresolved scientific, and policy issues, brought into stark relief by the Prüm
Treaty, which automates the exchange of some forensic bioinformation.
6.32.
International cooperation involving the exchange of forensic bioinformation is in
its infancy. While information is scarce, the amount of cooperation appears to
be occasional even when investigating serious crimes. Within the European
Union this might change as a result of the Prüm Treaty, although the
introduction of technologically advanced searching systems and databases
alone will not necessarily result in major increases in activity. The Council of the
European Union has already sought to restrict the extent of future searching
under Prüm, presumably to prevent analysts being overwhelmed by search
requests and the need to verify results. This state of affairs may not be in the
interests of justice. There is a consistent, but almost anecdotal view held by
experienced investigators about the value of routinely exchanging
bioinformation for dealing with transnational crime. In the absence of statistical
analysis and independent case evaluation however, it is difficult to reach an
informed view about the optimal scale and arrangements for such cooperation.
6.33.
Currently there are many jurisdictional differences in practice both within
common law jurisdictions and between common and civil law jurisdictions. In
some jurisdictions (e.g. US and Canada) legislation operates at both State and
Federal level. Whilst complexities arise from this, there are many jurisdictions in
which such levels exist and it seems unlikely that relevant local agencies in
such jurisdictions will easily surrender sovereignty to national bodies. However,
the interoperability of NDNAD and the Scottish database, together with the
ability of the forensic services of the two administrations and Scottish criminal
justice colleagues to manage different retention regimes on a large scale (as
indicated in Table 4.1) is clear evidence that forensic cross-jurisdictional
cooperation can be successfully managed on a large scale despite legal
differences. While this report identifies some gaps and lack of consistency in
the data and suggests where research is urgently needed, the scale and detail
of the interoperability analysis indicates that the NDNAD has set an example of
greater openness nationally and internationally for criminal justice databases,
337
Noble, R.K. (2007) Opening Remarks at 5th International DNA users’ conference for investigative
officers, 14 November 2007 (Lyon, Interpol) published at www.interpol.int .
338
Smith, D. (2007) Uncorrected Transcript of Oral Evidence to House of Commons Home Affairs
Committee, 23 January 2007, published at www.publications.parliament.uk .
119
including the exchange of information about fingerprints, and possibly genetic
collections in the genetic medical research and health fields.
6.34.
Any consideration of the multinational exchange of forensic bioinformation is
unlikely to give rise to novel technical issues. In some countries the size of
AFIS databases and the use of different multiplex systems or differences within
the same jurisdiction present the same problems as those that have to be
resolved before international cooperation can be successful. It helps to identify
with greater clarity some of key strategic issues facing the forensic community.
This is of little value, however, in the absence of institutions with the resources
and authority to foster greater national coordination.
6.35.
The key issue is higher risk of erroneous or missed identifications arising from
essentially the same type of problems in respect of both fingerprints and DNA:
the recovery from the crime scene of incomplete (both), degraded (DNA),
damaged (fingerprints) or distorted (fingerprints) material. Experts working in
both disciplines face similar questions about contamination or poor recovery,
transmission or storage techniques, but the main risk of missing a correct
match or declaring an incorrect identification stems from specific reasons
inherent to each discipline. The cognitive issues arising in respect of the
identification of similarities and differences in patterns by fingerprint examiners
are undoubtedly intensified by the scale and speed of modern database
algorithmic searches, and the ability of individual examiners to understand and
manipulate the operations of different proprietary systems.
6.36.
For UK DNA international casework the chief obstacle now and increasingly in
the future are the different multiplex systems used across the globe, the known
technological obsolescence of SGM+ and the considerable scientific difficulties
in moving to a sufficient number of overlapping loci with the other widely used
systems developed in the USA and China. These problems will undoubtedly
increase, as the Government has admitted, because of its decision to no longer
retain DNA samples. The initial trials involving Germany and the Netherlands
under Prüm have demonstrated that with many countries obtaining valid results
from cooperation depends on the ability to reanalyse cellular material taken
from a known individual.
Governance and Accountability
6.37.
Irrespective of views on the Marper judgment, there is a clear consensus that
the advances made technologically in the forensic use of DNA in England and
Wales have not been balanced by equal attention to the governance of this
emerging field of activity. This can be contrasted unfavourably with the
arrangements in Canada and some European countries. While it remains
120
questionable whether direct and periodic legislative scrutiny, along the lines of
the Canadian model, is the best and most informed approach for ensuring
public trust in the nature of future trajectories in the forensic use of
bioinformation, it remains perhaps preferable to foregoing any scrutiny at all.
Self-regulation is problematic and self-regulatory systems are prone to failure,
pointing to the need for stronger safeguards to be in place.
6.38.
There remain questions over how to regulate the taking of bioinformation by the
police (not considered in Marper or in ensuing debates). It is not mandated and
often occurs in cases where apparently a complete irrelevance, while it is not
taken in other instances, such caprice adding to public mistrust. Sensitivities
surrounding retention are exacerbated by the difficulties encountered by the
public in finding out their rights when having/ had DNA taken, in particular,
getting data removed, a point forcefully made by the recent Home Affairs
Committee Report on the NDNAD.
6.39.
A proper response to Marper must also include an effort to define key terms
(‘transparency’, ‘accountability’, ‘safeguards’, ‘proportionality’ etc) in a way that
establishes clear general principles for the governance of forensic
bioinformation in the UK. Innovations in current practice can then be
interrogated and positioned properly by reference to those principles. There is
also a need to think more generally about ‘data about individuals’ rather than
simply ‘bioinformation’. This means consideration of what kinds of forms such
data can take, where such data should be stored, how they should be
accessed, by whom and for what purposes. There needs to be recognition of
different levels of sensitivity and significance and not develop regimes which
constrain less intrusive forms and uses simply because some other forms and
uses may be especially contentious.
6.40.
Whatever governance arrangements are put in place, it is crucial that public
confidence and trust remain the touchstones for ‘success’. The UK Statistics
Authority, in their interim report on crime statistics, suspect that there is a wide
mistrust of official information, and argue that the way that information is
reported is critical to public perceptions and confidence:
“Trust is compromised if official statistics are not produced in a way that is
professionally independent, and seen to be independent…. But trust can
also be undermined by confusion and by the misuse of the statistics,
whether the misuse occurs inside or outside the Home Office. It is also
possible that the recent controversies about the use of research evidence
(in respect of drug classifications and the DNA Database) may have an
impact on trust in statistics.”339
339
UK Statistics Authority, ‘Overcoming barriers to trust in crime statistics’, (December 2009), p.11
121
6.41.
Trust in the use of forensic bioinformation has been hindered by the lack of, or
flawed research, which is then misrepresented or used to mislead. This has
sometimes exaggerated the impact of the NDNAD and does little to explain the
value of forensic bioinformation (including the value of fingerprints). The use of
emotive anecdotal cases and statements invoking ‘public protection’ with
almost meaningless numerical data, serves only to confuse, and removes from
consideration the impact on individuals and on human rights, which is the
‘balance’ that Marper demanded. There has as yet been scant attempt to
develop a human rights based approach to forensic bioinformation in the UK,
with strong governance at its core. If public confidence is to be maintained in
the future of forensic bioinformation, the issues outlined in this report, including
the production of robust research data, the creation of effective governance
regimes, undertaken within a framework that takes seriously the international
context, could be considered essential starting points.
A Proposal
6.42.
The formulation of legislation for the governance and operational use of
bioinformation has not been central to this study. Nevertheless, we closely
observed the passage of the Crime and Security Act 2010 and this has
informed our work. There are significant doubts about the durability of this
legislation. A considerable weight of expert opinion suggests that the retention
regime therein would not withstand judicial scrutiny. In view of this, we conclude
this report with a brief consideration of the objectives that might underpin
revised legislation in the new Parliament and provide an illustration (not a
recommendation) of what a reformed system might look like.
6.43.
Various problems inhibited the emergence of a coherent, comprehensive and
effective response to Marper. The political and institutional considerations are
dealt with in our summary account of the passage of the Act (paragraphs 1.12 –
1.20) and our concerns about the possible misinterpretation of the lessons from
Scotland are set out at paragraphs 6.22 – 25. This final section approaches the
issue differently by suggesting that the starting point to further legislation should
be a clearer and more convincing demonstration of the objectives and technical
arrangements that might ensure the success of new legislation. Underlying this
is our belief that further discussions about the response to Marper need to
dispel the various myths woven in this context about measures to protect rights
being inimical to a criminal justice system capable of detecting and dealing with
offending behaviour.
6.44.
New legislation, if developed with integrity and open-mindedness could achieve
a greater degree of public confidence (from both the rights and utility
perspectives). It should contain clear principles with precise arrangements to be
put in place and tested incrementally. This process might be initiated by the
122
creation of an independent statutory governance body, whose first task would
be to implement satisfactory arrangements for the management, and when
required, the destruction of tissue samples taken from known individuals and
the DNA profiles analysed as a result of such actions. We suggest that there
are four principles that should underpin the preparation of new legislation:
•
•
•
•
6.45.
Every effort should be made to ensure that the new legislation is likely to
withstand judicial challenge on ‘blanket and indiscriminate’ grounds and
that, for this purpose, proper consideration should be given, inter alia, to
whether this requires an approach based on a case by case scrutiny of
retention decisions in the absence of a conviction or where a conviction
has been quashed on appeal.
The legislation and its implementation, including risk assessment and
optional appraisal studies, should be sufficiently robust and
comprehensive to minimise the risk of later changes both in response to
the UK suffering serial judicial reversals on the retention issue or
recognition within the UK that the present system is inefficient in both
operation, incapable of ensuring a uniform quality in the use and
management of all forms of forensic bioinformation and police information
generally, and lacks sufficient transparency to ensure public confidence in
terms of both rights and utility.
The new arrangements should provide the discretion necessary for
investigators to retain and share forensic bioinformation and police
information when there is a professional need, and that the exercise of
such discretion will be subject, in line with the MOPI code, to regular and
effective professional review and audit and, for issues affecting the
retention of bioinformation, any challenges before an independent tribunal.
The general operation of the new arrangements and individual decisions
should be informed and regulated by the independent statutory oversight
of all forms of forensic bioinformation and, possibly in due course, police
information generally. This should be based on effective audit
arrangements, adequate data collection and publication, the
commissioning of research and ultimate accountability directly to
Parliament.
We have formed many views on the future of forensic bioinformation during this
project but we single out the following as some of the immediately relevant:
•
The potential shortcomings of the Crime and Security Act 20010 are
illustrated by the apparent lack of consideration given to the fact that
convictions may be overturned on appeal. For this reason alone, it
appears improbable that the recently enacted arrangements will not be
123
repealed or amended in the new Parliament irrespective of any further
judgment in Strasbourg.
•
The great advantage of the Scottish case by case approach is that it could
never be described as ‘blanket and indiscriminate’. Retention in all cases
reflects a considered individual judgment with independent oversight.
•
The Government estimated that the cost of changing the information held
on NDNAD and NAFIS to implement the reforms in the Crime and
Security Act 2010, excluding sample destruction, in excess of £50m.340
Little attention has been given to the possibility that if the new Act is
judicially reversed, further costly changes may be needed. This may help
to put the refusal - based on assertions about affordability – to consider a
case by case approach, in a more accurate economic perspective.
•
There are likely to be major advantages in being able to detect crime if
retention decisions were to be included within a case based and MOPI
regulated process. This would integrate the management of retained
bioinformation with other police information about individuals who had not
been convicted of an offence. By comprehensively combining information
it would assist officers when making decisions to retain or destroy
bioinformation, both initially and periodically in the reviews required under
the MOPI code. Such decisions would be unconstrained by rigid rules
about the suspicions of any specific types of offence that may have
resulted in arrest. Instead individual officers would be allowed to make a
professional decision about an individual’s behaviour following an arrest
through to conviction, acquittal or the overturning of a conviction on
appeal. Such a system would recognise the professional skills of police
officers and emphasise the trust that society has in their personal integrity
rather than the inferior management of important but highly variable
information by machine.
•
If the police are already MOPI compliant, other than for the cost of
introducing such changes, there should not be a major increase in cost. In
preparing an illustration of how the system might operate we fail to see
how this cannot be incorporated within the decision making and recording
processes happening now at key stages in an investigation: arrest, the
laying of charges, the initiation and continuation or discontinuation of
proceedings, the verdict and appeal decisions. In the longer term a
system that integrated all forensic bioinformation and PNC management
with MOPI would presumably be more cost effective, as well as more
efficient, than the present dispersed arrangements for the governance of
DNA profiles, fingerprints and other MOPI information. This would be
consistent with recommendations in key studies from Bichard Report
340
Home Office (2009), Impact Assessment of the implementation of the S & Marper ECtHR judgment
regarding DNA profiles, DNA samples and fingerprints retention published at
www.crimereduction.homeoffice.gov.uk/crimeandsecuritybill/ .
124
(2004) to the Flanagan Review of Policing (2008), which have been
concerned with avoiding ‘gaps in intelligence sharing’ and duplication.341
•
•
6.46.
Even under the Act of 2010 can the system really operate chiefly by the
application of an algorithm? Conviction would require a management
decision to convert a temporary period of retention to a permanent one.
Similarly acquittals, if they are to be taken into account, would require a
further intervention. If the system has to be designed to respond to such
instructions, other decisions could be notified to NDNAD management, as
part of the routine paperwork at two other key decision points: if it is
decided not to charge the arrestee or to discontinue proceedings.
The burden of paperwork could be reduced by default deletion and
destruction procedures. For example, if the police have not informed the
NDNAD and IDENT1 database management that proceedings have been
instituted within a set period after the notification that charges have been
laid, the information would be automatically deleted. This would also
provide a safeguard against the failure discovered by HMIC to comply with
earlier deletion legislation because the discontinuation of proceedings and
acquittals had not been notified by the police to the database operator.342
A further safeguard against a reoccurrence of this problem would be
provided in this respect by holding any information relating to unconvicted
or acquitted individuals on a separate MOPI database.
While the authors cannot claim any expertise in systems design, Figure 6.1 sets
out an option for how such arrangements might be introduced. The existing
NDNAD Ethics Committee might form the nucleus of the proposed audit and
ethics board and members of the NDNAD Strategy Board would become part of
the new executive body. It will be noted that we consider that the audit and
ethics body should be responsible for initiating research, as well as approving
external research requests and that its statutory audit functions should extend
to record keeping, database operations and the storage of samples by all
accredited forensic suppliers. This body should produce an annual report for
Parliament with sufficient data to describe the routine operations of the new
system, the basis volumes of business and results achieved, and the volume of
and results obtained from international cooperation.
341
Muir, R. (2009) Arrested Development: Unlocking change in the police service (London, IPPR), pp. 33-34.
In the 1990s there were major problems in removing profiles that should have been deleted under the
current law from the then quite small NDNAD. As a result, a large number of samples and profiles –
estimated at 50,000 but acknowledged to be perhaps higher - were being held on the NDNAD unlawfully.
See Williams, R., and Johnson, P. (2008) Genetic Policing (Cullompton: Willan), p. 84.
342
125
Figure 6.1: An option for arrangements for case by case decisions about retention, indicating
mandatory and discretion decision points within the criminal justice system. Mandatory
information requiring action by NDNAD operational staff is in solid green lines and optional police
decisions requiring action by NDNAD in dotted green lines. In the absence of action by the police
to order retention, in the absence of a conviction report, the profile will be deleted from the
database within a specified time of the original arrest or the latest discretionary retention decision.
126