Facilitating witness co-operation in organised crime cases: an

Facilitating witness co-operation in

organised crime cases: an international

review

Nicholas Fyfe

James Sheptycki

Online Report 27/05

Contents

Executive summary iii

1. Introduction 1

Research Aim 1

Methodology 1

Structure of the report 2

2. Legislative provisions to protect witness in organised crime cases 3

The US Federal Witness Security Program (WITSEC ) 3

Witness protection in Europe 5

Witness protection legislation in other countries 12

3. Accomplice testimony, plea bargaining and witness immunity 16

The US experience 17

The use of 'crown witnesses' in Europe 18

Historical background 19

The Italian experience 19

The German experience 21

The Dutch experience 22

The Irish experience: Northern Ireland 23

The Irish experience: The Republic of Ireland 24

Witness compellability 25

4. Discussion 27

'Costs' and 'effectiveness' 27

Legitimacy and accountability 29

Conclusion 33

References 35

Appendix 40

List of tables

2.1 A summary of legislation relating to witness protection in Europe 6

2.2 A summary of legislation relating to witness protection in countries 14

outside Europe and the US

3.1 Summary of measures used with respect to co-operating, criminal witnesses 16

3.2 Arguments for and against the compellability of witnesses 26

Executive summary

Witness intimidation and the need to draw upon 'accomplice testimony' from co-operating

witnesses are some of the key difficulties involved in the investigation and successful

prosecution of organised crime. To address these difficulties, countries have introduced

innovative approaches in legislation and policy to facilitate witness co-operation in organised

crime investigations. Three approaches in particular have commanded the attention of policy

makers: legislative provisions to protect witnesses; plea bargaining and immunity in relation to

witnesses giving evidence; and, the compellability of witnesses. Based on a wide-ranging

review of published and unpublished sources, this report examines the use of these different

approaches for tackling organised crime.

The longest established witness protection programme is the US Federal Witness Security

Program (WITSEC) established by the 1970 Organised Crime Control Act. The main form of

protection offered by this legislation is the secret and permanent relocation of witnesses and

their families to places of safety and, if necessary, a change of their identities. Although

initially viewed as highly effective, several important problems emerged with WITSEC soon

after it was established. These included: no clear definition of the boundaries of protection,

resulting in the type of assistance being resolved on a case-by-case basis; and little

consideration being given to the unintended victims of witness relocation, such as the

communities into which witnesses were moved given that most of the protected witnesses

had criminal convictions, and therefore posed a potential threat in the places to which they

were moved. As a result of these difficulties WITSEC has undergone a series of reforms.

Over the last ten to fifteen years, US-style witness protection programmes have become a

key feature of the European criminal justice landscape, although variations in the nature and

extent of organised crime as well as differences in legal systems, mean that the precise form

of witness protection arrangements varies from country to country. Some countries have

specific legislation governing the operation of their witness protection programmes while

others, like the UK and Netherlands, have none. Some countries, including the UK, view

witness protection as largely a police function while others give a key role to the judiciary and

government ministries. In some countries, like Belgium and Italy, there is one national or

federal witness protection programme; other countries, including the UK and Germany, have

several regional or local programmes. Despite these differences there are clear attempts to

harmonise aspects of witness protection arrangements across Europe.

The relevance of plea bargaining and immunity in cases of organised crime centres on the

role played by 'accomplice testimony' provided by criminal informants who become so-called

'crown witnesses', individuals to whom a promise has been made about the future handling of

any criminal proceedings against him/her in exchange for witness co-operation. There are

important variations between countries in the use of these measures. Japan and France, for

example, have no system of plea bargaining or witness immunity, while the United States has

well-established practices in these areas.

Although accepted as key tools in the prosecution of organised crime, the use of plea

bargaining and witness immunity statutes in the United States has proved controversial.

"Buying" testimony with promises of sentencing leniency or immunity is seen by critics as

wrong because it presents too great a risk of perjury and thus of wrongful convictions.

Creating an environment in which providing 'substantial assistance' is the main way informant

witnesses get reduced sentences also risks the so-called 'co-operation paradox' whereby

"kingpins" receive lower sentences than their underlings because the "kingpins" have more

information to exchange for a 'substantial assistance' recommendation.

In Europe, several countries have relied on the use of crown witnesses to tackle organised

crime. In Italy, for example, these crown witnesses, (or 'pentiti') benefit from reward

regulations (or 'premali') so that a judge may decide not to sentence or to impose a less

severe sentence on these informants. To ensure the truthfulness of statements, a review

procedure exists which allows the sentence to be raised if a crown witness had made false

statements. Nevertheless, the use of crown witnesses is also highly controversial, with

concerns over the role that criminal informants may play in the instigation of criminal activity,

their questionable motivations for providing information, and their credibility as witnesses.

While there is little research evidence on the use of witness compellability in organised crime

investigations, studies of the use of this measure in the context of domestic abuse have

indicated that the arguments for and against are not clear-cut. On the one hand,

compellability may reduce the risk of witness intimidation occurring as the responsibility for

testimony has (somewhat) shifted from the witness to the prosecution. However, a compelled

witness may become hostile to the prosecution and only give evidence which assists the

defence or refuse to give evidence and risk being punished for contempt. Moreover, the

experience in other legal contexts is that compellability is only effective when it is backed up

by other measures, and especially adequate protection and support for witnesses. Whilst,

superficially, it might seem tenuous to apply evidence from domestic violence cases to a

consideration of witness compellability in organised crime, there are certain similarities which

make such a comparison instructive. In particular, familial and inter-personal relationships will

typically link witness and defendant in both types of case and they may in turn have a critical

bearing on witness co-operation.

The evidence base relating to the 'costs' and 'effectiveness' of the measures considered in

the report is very weak. In relation to the financial costs of witness protection programmes,

WITSEC in the United States had a budget of about $24million in 2000 and had over 16,000

participants; in Italy, the Central Protection Service had a budget of 200 billion Lire in 2000 (c.

£72 million), and had over 5000 participants. Anecdotal evidence of the 'effectiveness' of

these and similar programmes is generally positive, indicating that few, if any, protected

witnesses have become the victims of attack, while the limited amounts of conviction data

relating to cases involving protected witness testimony indicate high numbers of defendants

being found guilty.

Viewed in broader terms, however, assessments of 'costs' and 'effectiveness' are less

encouraging. While some have argued that witness protection and immunity for criminal

informants provide an 'exit strategy' for career criminals and help them establish a place in

legitimate society, data from the United States indicates that 20 per cent of protected

witnesses are arrested within two years of being put on the WITSEC programme.

Many of the measures designed to facilitate witness co-operation raise issues regarding their

perceived and actual legitimacy. Witness immunity statutes in the United States, for example,

give wide discretion to prosecutors in deciding whom to prosecute and whom to grant

immunity to which risks undermining the principle of equal protection under the law. Witness

immunity might also be seen as a 'bribe' on the part of governments and could be viewed by

critics as encouraging potential informants to give false testimony. Specific concerns exist in

Europe about the legitimacy of measures to assist witnesses in the context of the European

Convention of Human Rights (ECHR) concerning the guarantee of a fair trial (Article 6). To

date, the use of crown witnesses and court room measures like screens and video-links for

vulnerable and intimidated witnesses have been accepted as ECHR-compliant.

There are significant variations between countries in the types of measures they have

introduced to facilitate witness co-operation in organised crime investigations, partly reflecting

differences in the scale and nature of the organised crime and partly linked to differences in

legal traditions and environments. The evidence-base on which to judge the efficacy of these

measures, however, is very weak and there remains a real need for empirical research

evaluating different approaches. Moreover, given that the overwhelming majority of witnesses

on protection programmes and those involved in prosecutions as 'crown witnesses' are

themselves criminals, the potential for scandal is ever-present and needs to be guarded

against. If these techniques are to be utilised, therefore, they should ideally be employed in

the context of enhanced evidence as to their effectiveness, and under the umbrella of

transparent legislative guidelines as to their appropriate use.

1. Introduction

One characteristic of organized crime is that the most culpable and

dangerous individuals rarely do the dirty work. Although the organizations'

leaders are ultimately responsible for its crimes, they typically deal through

intermediaries and limit their own participation to behind-the-scenes control

and guidance. (Jeffries and Gleeson, 1995)

Organised crime's oldest and most effective tool against criminal prosecution

is killing the witnesses against them. (Schur, 1988)

These two reflections on the nature of organised crime highlight the difficulties surrounding

the investigation and successful prosecution of this type of criminal activity. To address these

difficulties, criminal justice agencies typically need to use 'accomplice testimony' in the form of

evidence from so-called 'co-operating' or 'crown' witnesses and to be able to offer protection

to these witnesses given the risks of intimidation that they face. Against this background, an

increasing number of countries have introduced innovative approaches in legislation and

policy to facilitate witness co-operation in organised crime investigations. Three approaches

in particular have commanded the attention of policy makers:

• legislative provisions to protect witnesses in organised crime investigations;

• plea bargaining and immunity in relation to witnesses giving evidence in organised

crime cases; and

• the compellability of witnesses in organised crime cases.

Research aim

This research forms part of the Crime Reduction and Community Safety Group (CRCSG)

Organised Crime Research Programme. The aim of this report is to examine the use of these

different approaches for tackling organised crime. Given, however, that legislation, policy

development and independent evaluation in this field is, in many countries, in its infancy, the

discussion of the three approaches will be uneven in its depth and coverage.

The review will focus particularly on issues to do with:

• legislative provisions to protect witnesses;

• compellability of witnesses; and

• plea bargaining and immunity.

Methodology

This report is based on a wide-ranging review of published and unpublished sources

including academic research papers and policy documents. In addition, specific

requests for information were submitted to researchers and law enforcement

agencies in Europe, North America and Australia. In order to identify relevant

published academic and non-academic materials, searches of a number of databases

were carried out (see Appendix for details).

Several important issues emerged from this data collection exercise:

• Despite the relatively prominent discussion of organised crime cases and the

experiences of witnesses in the news media, social scientific and legal research in the

areas of witness protection, plea bargaining and immunity, and witness compellability

in the context of organised crime is largely in its infancy and in some cases appears

to be non-existent.

• Witness protection appears to have gained the greatest research attention, most of it

focused on the experience of the United States which has the longest established

programme for protecting witnesses in organised crime cases (see Chapter 2). By

contrast, no academic references relating to the use of witness compellability in

organised crime cases could be found, although some work on compellability in

relation to domestic violence cases does exist (see Chapter 4). The absence of a

large research-based literature on witness co-operation and organised crime most

likely reflects a combination of influences, but particularly the newness of and secrecy

surrounding many of the measures that have been introduced.

• The limited and uneven information available on measures used to enhance witness

co-operation in organised crime investigations has important implications for this

report. Using legislation and policy papers it was possible to sketch out the broad

contours of what measures are used where, but given the absence of independent,

evaluative research, it was much more difficult to assess what measures were

effective in facilitating witness co-operation in organised crime cases and why.

Structure of the report

To date, legislative provisions to protect witnesses appears to have attracted the highest

levels of policy and research attention and Chapter 2 of this report focuses on this area. It

reviews witness protection legislation in the US and Europe but also includes consideration of

similar legislation elsewhere in the world, highlighting similarities and differences in the

structure and operation of witness protection programmes. In the field of plea bargaining,

witness immunity and accomplice testimony there was limited policy material available for

scrutiny but some research findings that have assessed these measures in the US and

Europe are reviewed in Chapter 3. The chapter also examines witness compellability. The

only available research evidence, however, focuses on the use of this tactic in domestic

violence rather than organised crime cases; consequently, discussion of this is limited by the

available material. Finally, Chapter 4 brings together the research reviewed in the report to

raise three broader issues about the use of these measures in organised crime investigations

concerning the balance between 'costs' and 'effectiveness', accountability and legitimacy.

2. Legislative provisions to protect witnesses in

organised crime cases

While a range of issues have been identified as problematic in investigating and prosecuting

organised crime groups, the lack of effective co-operation from witnesses in organised crime

investigations and court cases has been identified as an area of special concern. Effective co-

operation from witnesses is widely recognised as playing a key role in the successful

investigation and prosecution of all types of crime (e.g. Gabor ef. al., 1987; Silke, 2001).

Facilitating witness co-operation in cases involving organised crime groups has been

particularly problematic. A perception exists that poor witness co-operation in these cases

significantly hampers the successful investigation and prosecution of organised crime groups

in England and Wales.

A number of foreign jurisdictions have developed detailed legislation provisions to facilitate

witness co-operation in organised crime cases. The following chapter provides a review of

those measures which are aimed at providing witness protection in such cases. The first

system to be considered is the US Federal Witness Security Program (WITSEC) as this is the

most well known and well established of the current systems.

The US Federal Witness Security Program (WITSEC)

Described as 'the paradigm program' (Roberts-Smith, 2000) on which many other countries'

witness protection programmes are modelled, the US Federal Witness Security Program

(WITSEC) was established by the 1970 Organised Crime Control Act. The background to the

Act was the limited success in the 1960s of the Justice Department's attempts to tackle

Italian-American organised crime and, in particular, the problems created by the Mafia code of

omerta, the code of silence which meant that it was proving difficult to get members of 'the

mob' to testify (Earley and Schur, 2002). By providing a high level of security to mob

witnesses, including secret relocation and a change of identity, WITSEC became the key to

breaking omerta.

The Organised Crime Control Act states that 'The Attorney General shall provide for the care

and protection of witnesses in whatever manner is deemed most useful under the special

circumstances of each case'. Key provisions of the Act include:

• specifying that for a witness to qualify for protection under the Act they must be a witness

in a specific case in progress and there must be evidence that it would be in the federal

interest for the Justice Department to protect the witness;

• giving the Attorney General the authority to provide for the security of actual and potential

government witnesses and their families in organised crime cases;

• allowing the protection of a witness and their family to continue for as long as these

people are, in the Attorney General's judgement, in danger; and,

• giving the Attorney General authority to determine what kind of facilities are to be

provided for a witness and their family such as housing, welfare and health care.

In the wake of a number of high-profile successful cases, some commentators heralded

WITSEC as one of 'the most effective law enforcement tools for organised crime' (Montanino,

1990, p.502). Several important problems, however, began to emerge with WITSEC soon

after it was established. These included:

• No clear definition of the boundaries of protection, resulting in the type of assistance

being resolved on a case-by-case basis.

• Little consideration given to the unintended victims of witness relocation. These

included individuals and organisations unable to recover unpaid debts from witnesses

and their dependants enrolled in the Program; non-relocated parents of children taken

into protection by a partner and denied access to their children; communities into

which witnesses were moved given that most of the protected witnesses had criminal

convictions, and therefore posed a potential threat in the places to which they were

moved.

• Under-staffing and under-training of the Marshals Service, leading to breaches in

security, delays in dealing with the documentation needed for new identities, and an

inability to respond effectively to the welfare needs of witnesses and their families

trying to adjust to their new lives.

Against this background, Congress reviewed the structure and operation of WITSEC,

focusing on the need to recognise and place a higher value on the rights and interests of the

public when those rights and interests conflicted with the purpose of WITSEC (Lawson,

1992). In 1984 these issues were addressed in the 'Protection of Witnesses' chapter of the

Comprehensive Crime Control Act. Its main provisions include:

• broadening the Attorney General's authority to offer relocation or other protection to a

witness, their immediate family or a person closely associated with the witness if that

person is also in danger, in official proceedings concerning organised criminal activity,

drug trafficking offences, or other serious federal or state felonies where the witness's

testimony is likely to lead to retaliation by violence or threats of violence;

• defining what the Attorney General can provide for a protected witness in terms of

establishing a new identity, housing, subsistence payments, help in finding employment,

and any services needed to help the witness become self-sufficient;

• requiring a written assessment from the Attorney General when selecting witnesses for

the Program, evaluating whether the need for a person's testimony outweighs the risk of

danger to the public, including potential harm to innocent victims;

• giving statutory authority to a Memorandum of Understanding to be signed by relocated

witnesses in which they must agree not to commit any crime and to comply with legal

obligations and civil judgements. The Memorandum also outlines the procedures to be

followed in case of a breach of the Memorandum and for the filing and resolution of the

grievances of witnesses provided with protection;

• the Act allows those trying to recover debts from relocated witnesses to use the courts as

a means to enforce judgements against relocated individuals, eliminating the previously

laissez-faire approach to relocated persons' debts;

• in relation to parent-child relationships, the Attorney General must give an assurance that

relocation will not infringe any of the non-relocated parent's custody or visitation rights;

• authorising the Attorney General to pay compensation from a victims compensation fund

if the victim of a crime committed by a protected witness seeks compensation; and,

• the Office of Enforcement Operations is required to submit a quarterly report to the

Deputy Attorney General detailing the results of the testimony provided by relocated

witnesses.

At one level, these new provisions clearly represent an advance on the original legislation. At

another level, however, even with these improvements to the original legislation, the

government's policy of witness protection appeared to place harm to the public from

organised crime above harm to the public from protecting witnesses. Policy makers, it was

claimed, were prepared to accept a level of unintended victimisation that might result from a

policy of relocating individuals of whom the overwhelming majority were criminals (Lawson,

1992). One suggestion has been for stricter standards in the selection of witnesses for

WITSEC by means of an independent review board which would focus more on the likelihood

that witnesses would be able to make the necessary behavioural adjustments to relocation

and a new identity than on simply the importance of their evidence. In addition, several areas

of the US have established alternative forms of witness protection programmes. Examples

include the Short Term Protection Program established in Washington in 1991 in response to

localised criminal and street gang organisations and the Illinois Gang Crime Witness

Protection Program established by State legislation in 1996 (Illinois Criminal Justice

Information Authority, 1997). The latter is administered by the Illinois State Police and is

aimed at facilitating the prosecution of gang crimes by providing additional assistance and

resources for the protection and relocation of victims and witnesses.

Witness protection in Europe

Despite its origins in a very particular law enforcement context, WITSEC has provided an

important 'model' for other countries developing their own witness protection programmes.

Over the last ten to fifteen years in Europe, US-style witness protection programmes have

become a key feature of the criminal justice landscape and are now viewed by many law

enforcement agencies as a crucial tool in cases involving organised crime and terrorism.

Nevertheless, variations in the nature and extent of organised crime, as well as differences in

legal systems, mean that the precise form of witness protection arrangements varies from

country to country. Table 1.1 provides a summary of the key features of legislation relating to

witness protection in European states.

Some countries such Belgium and Germany have specific legislation governing the operation

of their witness protection programmes while others, like the UK and Netherlands, have none.

Despite these differences, however, the eligibility criteria for witness protection measures are

broadly similar. The witness needs to be giving evidence in relation to the most serious

crimes and those who are close to the witness who might be endangered are also eligible for

protection. The forms of protection available are also quite similar (regardless of whether

there is specific legislation) and normally involve the relocation of a witness and his/her close

family, the possibility of formally changing their identity and help with social and economic

assimilation in the communities to which they are moved.

Table 2.1: A summary of legislation relating to witness protection in Europe

Some countries, including the UK, view witness protection largely as a police function (Fyfe,

2001), while others give a key role to the judiciary and government ministries. In Belgium, for

example, a Witness Protection Board, comprising public prosecutors, senior police officers

and members of the ministries of Justice and the Interior, take the decision about who is

protected. In Italy a Central Commission takes the decisions, chaired by the Under-Secretary

of State at the Ministry of the Interior. By contrast, in the UK, decisions about inclusion on

protection programmes are taken by a senior police officer.

In some countries, like Belgium and Italy, there is one national or federal witness protection

programme; other countries, including the UK and Germany, have several regional or local

programmes.

In some countries, most notably Italy, the nature and scale of organised crime means that

there are thousands of participants on witness protection programmes while in others there

are probably no more than a few hundred. In the Netherlands and Germany, for example, the

number of participants on witness protection programmes is estimated to be between one and

three people per million inhabitants (in Germany about 650 people a year are taken on to

witness protection programmes); in Italy there are currently about 91 people per million

inhabitants, equivalent to over 5,000 people of whom about 1,000 are witnesses and 4,000

are their close relatives (Heijden, 2001; Italian Ministry of the Interior, 2001).

Despite these differences there are clear attempts to harmonise aspects of witness protection

arrangements across Europe (Council of the European Union, 1995; 1997). A European

Liaison Network, co-ordinated by Europol and comprising the heads of specialist witness

protection units, was established in 2000 and has begun a process of providing guidance and

sharing best practice. As a Europol document explains, the main goal of the Network is 'to

create a useful, common platform for future co-operation and to give those Member States in

which the implementation of witness protection is still underway the great chance to avoid

waste of time in 'reinventing the wheel' again' (Europol, 1999). One example of this is the

production by the European Liaison Network of guidelines on the 'Common criteria for taking

a witness into a Protection Programme' (Europol, 2003). A further important dimension to

witness protection arrangements in Europe is the presence of the United Nations War Crimes

Tribunal in the Netherlands. The Victims and Witnesses Section of the Tribunal plays an

important role in providing witness protection services for those testifying in cases involving

serious violations of humanitarian law, such as the crimes against humanity in the former

Yugoslavia (McCusker, 2002; Klipp, 1996; and Hampson, 1998).

Witness protection legislation in other countries

In addition to Europe, the US model of witness protection has also informed the development

of witness protection programmes in other parts of the world. Table 2.2 summarises some of

the key features of the legislation in Australia, Canada, Jamaica, Japan and South Africa.

With the exception of Japan, where witness protection arrangements are limited to what

happens at court, this legislation reveals many similarities with WITSEC and the various

European witness protection programmes. In particular, several interesting issues emerge

from this legislation. In contrast to most of Europe and the US, but like the UK, decisions

about who is protected in Australia and Canada are taken by senior police officers. In

Jamaica the Ministry of National Security and Justice take these decisions while in South

Africa it is the Director of the Office for Witness Protection who is appointed by the Minister of

Justice.

In terms of the structure of witness protection arrangements Australia and Canada - like the

US - operate federal witness protection programmes, but this does not preclude individual

states within these countries using legislation to set up their own witness protection schemes.

In Australia until the late 1980s, for example, witness protection was the responsibility of

individual police authorities. These were largely ad hoc arrangements and suggestions were

made for a national witness protection scheme to be run by a new independent agency. This

was rejected in favour of mandating the Australian Federal Police to operate a National

Witness Protection Program. Most, though not all, Australian states have also passed

legislation creating state-level witness protection programmes run by the state police.

Finally, with regard to accountability, both Australia and Canada require their federal

protection programmes to submit annual reports on their activities and expenditure to the

national parliament.

Table 2.2: A summary of legislation relating to witness protection in countries outside Europe and the US

3. Accomplice testimony, plea bargaining and

witness immunity

The relevance of plea bargaining and immunity in cases of organised crime centres largely on

the role played by 'accomplice testimony' provided by criminal informants who become so-

called 'crown witnesses', 'government witnesses' or 'co-operating witnesses'; individuals to

whom a promise has been made about the future handling of any criminal proceedings

against him/her in exchange for witness co-operation. There are, of course, important

variations between countries in the use of these measures. Japan and France, for example,

have no system of plea bargaining or witness immunity, while the United States has well

established practices in these areas. Table 3.1 summarises some of these differences for

countries for which evidence was available. The following sections examine some of these

cross-national differences in approach to co-operating witnesses in more detail.

Table 3.1 Summary of measures used with respect to co-operating, criminal witnesses

The US experience

Although state and local level prosecutors may be involved in the prosecution of organised

crime, this section focuses on the federal level given that several features of federal law

combine to give federal prosecutors certain advantages over their state and local

counterparts in the investigation and prosecution of organised crime cases (Jeffries and

Gleeson, 1995). One of these advantages is the Racketeer Influence Corrupt Organizations

Act (discussed by Levi and Smith, 2002) but other advantages include the use of accomplice

testimony, plea bargaining, and witness immunity.

Accomplice testimony

Given that one of the key characteristics of organised crime is that the leaders of criminal

organisations typically operate through intermediaries, their guilt can rarely be proved by the

testimony of victims, eyewitnesses or forensic evidence. Successful prosecution of organised

crime leaders therefore normally requires the use of accomplice testimony. In federal courts,

a defendant (unlike a defendant in many state courts) can be convicted on the uncorroborated

testimony of an accomplice, although juries are cautioned to use care in evaluating the

testimony of accomplices. Indeed, because juries tend to be sceptical of such testimony,

federal prosecutors rarely rely solely on accomplices. Nevertheless, the fact that federal

courts are more permissive than state courts in allowing uncorroborated testimony is viewed

as a crucial advantage in the federal prosecution of organised crime because such testimony

can provide a richly detailed context to a case (e.g., that a criminal organisation met at a

particular location and that the witness was in a position to know about the types of criminal

acts at issue) that can help make the case against a defendant compelling.

Federal Sentencing Guidelines and plea bargaining

The Federal Sentencing Guidelines introduced in 1987 established a mechanism to prescribe

sentences for federal crimes with judicial discretion narrowly restricted by the creation of

sentencing 'ranges'. Although originally produced to address the absence of uniformity in the

sentences of federal defendants, the guidelines have become 'a valuable tool, used by

federal law enforcement authorities to turn targets and defendants into accomplice witnesses'

(Jeffries and Gleeson, 1995). Before the guidelines it was difficult to 'turn' members of an

organised criminal group because investigators were unable to make a 'persuasive plea offer'

given the uncertainties about the reward an accomplice witness might receive for co-

operating. With the Guidelines, however, a 'prospective co-operator' can be shown exactly

where they will fall within the 'sentencing chart' if found guilty and that the only way out of the

position on the chart is to co-operate with the government. The prosecution can then offer the

individual the opportunity to plead guilty, typically to the most serious offence s/he has

committed, and to testify for the government in exchange for a 'substantial assistance motion'

(described by Bowman, 1999, as 'the most powerful prosecutorial tool in modern federal

criminal law'). The court is also provided with a report on the nature and value of the co-

operation rendered. This arrangement allows the prosecutor to argue to the jury that any

leniency the witness receives will come from the court and therefore counter any suggestions

by the defence that the accomplice's willingness to testify in exchange for leniency makes

him/her an unreliable witness. As Jeffries and Gleeson argue, 'By creating a system in which

the prosecutor alone can make an accomplice witness eligible for leniency, but in which the

terms of that leniency are determined afterward by the court, Congress has vastly

strengthened the hands of federal prosecutors in dealing with organized crime'.

According to Bowman (1999) the Sentencing Guidelines have radically altered the plea

bargaining environment in federal courts. For most defendants, virtually the only ground on

which a departure from the severe sentences set out in the guidelines can be based is

'substantial assistance' to the government. Indeed, according to Bowman, not only have the

Guidelines encouraged federal prosecutors to bargain for testimony, they have also changed

the way criminal defendants think about co-operating (he quotes one defendant who claimed

that "Before [the guidelines], nobody wanted a snitch jacket. Now everybody rats, man").

However, several concerns exist in the US about plea bargaining and 'substantial assistance'

recommendations by federal prosecutors. In particular, critics of current practices argue that

"buying" testimony is wrong because "testimony" purchased with promises of sentencing

leniency present too great a risk of perjury and thus of wrongful convictions. Furthermore,

creating an environment in which providing 'substantial assistance' is the main way informant

witnesses get reduced sentences may generate a range of 'unfair results'. It risks the so-

called 'co-operation paradox' whereby "kingpins" receive lower sentences than their

underlings because the "kingpins" have more information to exchange for a 'substantial

assistance' recommendation.

In response to these concerns, however, Bowman (1999) sets out a number of counter

arguments. 'The true justification for exchanging leniency for co-operation', he contends, 'is a

utilitarian argument from necessity - without accomplice information and testimony, many

serious crimes, particularly complex crimes involving groups could not be solved or punished'.

This is echoed by Ewing (2000) who argue that there are strong public policy interests in

allowing bargaining for testimony:

Many prosecutions depend heavily on the testimony of co-conspirators and

accomplices as evidence against the defendant. Prohibiting prosecutors from

entering into plea bargains in exchange for testimony would make it much

more difficult to prosecute crimes, such as drug trafficking conspiracies,

where there are few potential witnesses who were not involved in the criminal

activities at some level. This reality provides a strong public policy

justification for continuing to permit prosecutors to make plea agreements

that exchange leniency for testimony.

The US has both federal and state level witness immunity statutes which allow a witness with

a valid Fifth Amendment privilege (i.e. the privilege that no person 'shall be compelled in any

criminal case to be a witness against himself) to be compelled to testify by a court if the

prosecution determines that the testimony is necessary and in the public interest. If the

prosecutor makes such a determination, the witness is granted limited immunity which

protects them from the use of their testimony as evidence against them in a criminal

prosecution. The rationale given by the proponents of immunity statutes is this: 'the Fifth

Amendment protects only compelled testimony that tends to incriminate, and therefore if a

person is granted immunity from prosecution for his incriminating statements, the Fifth

Amendment does not apply and that person may thus be compelled to testify' (Menza, 1999).

The use of "crown witnesses' in Europe

The use of criminal informants in pursuit of organised and serious crime is now well known

across Europe (Tak, 1997). At present in many countries in Western Europe, police

intelligence systems have developed a considerable capacity to identify and recruit criminal

informants in the illicit or underground economies that feed organised crime. These

capacities have been developed in the context of perceived failure to control these markets by

traditional law enforcement means. Informers who have been deployed in the criminal milieu

may contribute useful information that directs other aspects of police investigations (such as

suggestions about where to target other surveillance efforts through the use of tools such as

telephone intercepts and financial monitoring). However, there is a current of opinion that the

ability of these criminal informers to become 'crown witnesses' in the context of criminal trials

has the greatest potential to impact upon organised crime.

Historical background

The use of 'crown witnesses' has deep historical roots. According to Radzinowicz (1956) the

legal figure of 'approvement' is an ancient idea in common law. As he explained it, under the

rubric of 'approvement',

an offender was allowed to confess both to the particular crime and to all the

other treasons or felonies in which he had been either an agent or principal.

He could then offer to name an accomplice and to prove his guilt. When the

Court admitted his offer, which it had the power to refuse, the accused was

put on trial: if he were acquitted, the approver was sentenced to death; but if

convicted, the approver was pardoned. (ibid, p. 44)

The use of the pardon for crown witnesses became a major instrument for bringing criminals

to justice (ibid. p. 55). However, giving 'impunity for accomplices' was acknowledged to be

fraught with moral ambiguity and challenged on effectiveness grounds, not least 'because the

majority who were pardoned for having 'turned the King's evidence' did not, on release, lead

an honest life but instead having got rid of the leader of the gang, stepped into his place' (ibid,

p. 53-54).

In the 19

century, continental jurists in the classical tradition of Beccaria cautioned that those

who are guilty of crimes cannot claim full credibility. When a witness makes a statement

against an accomplice, and by doing so becomes eligible for a reduction of sentence, pardon

or other privilege that the criminal justice system may give, reliance on any such statement is

always questionable (Mittermaier, 1851). According to Peter Tak (1997) it was subsequent to

the establishment of the modern police that the use of the crown witness 'nearly disappeared

from the scene of criminal proceedings' (p. 4). One might speculate that the emphasis of the

'new police' on prevention, and especially in its British manifestation, the emphasis on high

visibility police patrol, tended to de-emphasise the darker police arts of managing informants.

Whatever the reasons for the hiatus in the use of serious criminal informants, it was only

during the 1970s that the practice experienced a revival when, under the influence of political

activism, unorthodox investigation methods and means to clear up crime were in demand.

Since then the use of pentiti-type witnesses in cases of serious and organised crime, and, of

course, terrorism has come increasingly to the fore in numerous European countries.

The Italian experience

The Italian anti-Mafia trials that have taken place depended significantly on pentiti (crown

witness) testimony. This testimony was facilitated by premiali (reward regulations). Such

regulations pertain to four kinds of crime: terrorism, kidnapping with ransom aims, production

of and dealing in drugs, and organised (Mafia) crime. Premiali regulations for all four types of

crime are established along similar lines. The judge may decide not to sentence or to impose

a less severe sentence in light of co-operation given to police. The conditions for the

application of premiali regulations vary, but the main requirement is that the co-operation

must be aimed at preventing or thwarting criminal offences or at supplying information which

is crucial to the progress of an investigation (Tak, 1997, pp. 18-19).

The use of this approach in Italy began with the efforts against the Red Brigades, but mutated

in the wake of the arrest and trial of Tommaso Buscetta in the United States in 1984 (Greer,

1995). Information provided by Buscetta, plus the evidence of one Salvatore Contorno led to

the first 'maxi-trial' which involved 475 defendants which ended in December 1987, 22

months after it began with 338 convictions. By early 1989 only 60 of those convicted at the

maxi-trial were still in prison,

In contrast to the lack of success against organised crime in Italy, enforcement efforts against

terrorist cells bent on the 'subversion of the democratic order by violent means' met with

evident success; at least for a time. According to Greer, (2001, p. 131-133) an important part

of the explanation for the diminution of political/ideological terror in Italy after its inception in

the late 1970s, has to do with the narrow social base that such terrorism sprung from. This

stands in notable contrast to the more broadly rooted Mafia-related crime (and which, given

its propensity to corrupt officials in the highest offices of the state and judiciary, can also be

said to have a political dimension). Greer (2001, pp. 131-133) charted a series of temporary

legal innovations that were made in the wake of the activities of the Red Brigades. These

included:

• The crime of'terrorism or subversion of the democratic order'(Law N. 191,21 March

1978). The law contained statutes providing for new intelligence gathering provisions

(including telephone surveillance) and extra police powers to detain suspects and

conduct interviews without the presence of a lawyer.

• The Cossiga Act (Law N.15 of Feb. 15 1980) which made it an offence to join, promote,

organise, or direct an association aimed at the subversion of the democratic order by

violent means. Article 5 gave immunity from punishment for those who collaborated with

the authorities in order to prevent crimes against the state, and who provided substantial

evidence regarding conspiracies, the modus operandi of their crimes, and identification of

conspirators.

• A temporary penf/f/ statute (Law N. 304 of May 29, 1982) offering immunity from

prosecution and/or reductions of penalties to persons involved in terrorist activities

providing that they had dissociated themselves from such activities, made full confessions

and helped to reduce the impact of their wrong doing.

• An act which provided formal conditions for the reduction of penalties (Law N. 34 of Feb.

18, 1987) requiring dissociation, confession, repudiation and reformation, but not active

co-operation in the judicial process.

• No legal instruments came into force (or, apparently, were considered) to provide pentiti

with formal witness protection (Tak, 1997, p. 19).

Greer characterised the pentiti programme in the combat against terrorism in Italy as a

success. This judgement was arrived at not only because of the number of arrests and

successful prosecutions, but also because of the apparent diminution of such activity. It is

therefore perhaps worth noting that in 1999, the reborn Red Brigades claimed responsibility

for killing another political consultant, Professor Sergio D'Antona, in broad daylight on a street

in Rome. While not on the same scale as in the 1970s, the rekindling of political violence was

apparent to all, claiming the life of at least one police officer.

Greer's analysis of the contribution of pentiti strategies and associated witness protection to

effective control of organised crime was less positive. In his estimation, the pentiti process

was 'dogged by controversy and marked by peaks and troughs' (2001, p. 127). The

controversies surrounding the credibility of witnesses and their motives, as well as a 'series of

familiar political crises' cast doubt on the witness protection programme which had, by the

end of the 1990s, more than a thousand pentiti under its wing. He was not able to provide

any quantitative data that could show the impact of these crime control measures on

organised crime and corruption in Italy. Nevertheless, in summing up his analysis, Greer

observed that the new millennium has 'dawned in Italy, with the Mafia weakened, but not

defeated, and with the credibility of the pentiti as a legitimate and viable method of dealing

with it in serious doubt' (ibid. p. 128).

According to Tak (1997)

As the years have passed, it has become clear that the figure of the crown

witness could be crucial to the fight against other forms of organised crime, if

parallel legal and administrative measures were developed to do full justice

to the role of the crown witness inside and outside the criminal procedure.

The latest crown witness legislation is, therefore, part of a package of

measures which have considerably strengthened their position in the criminal

procedure. These measures also offer more certainty about the credibility of

statements, provide the necessary protection against revenge attacks and

produce broader acceptance of this figure by the judiciary and the public at

large. These measures were urgently required when the reliability of crown

witnesses in Mafia trials was questioned. (p. 19)

This analysis provided a close view of the legal innovations surrounding the pentiti process,

but did not seek to provide any measure of impact. Tak's overview reveals considerable

trouble in co-ordinating the multi-agency efforts required in managing witness protection (p.

22). It would, perhaps do to note that this echoes Norris and Dunningham's analysis of inter-

agency conflicts in the management of informants in the UK (2000). Italian Minister of

Interior, Massimo Brutti, acknowledged the evident difficulties of managing witness protection

in the Italian context in 2001. Brutti confirmed a series of botched witness protection cases

reported in La Republica - some of which culminated in murder of protected witnesses -

saying that the cases were representative but would not recur under a law passed in January

of that year. He stated at that time that "Now there is a structure for witnesses and one for

judicial collaborators. I do not deny there was a lot of confusion" (Carroll, 2001). Table 1.1

summaries this revised structure.

The German experience

Like the Italian case, the use of state's witnesses {staatszeugen) in Germany - sometimes

known as crown witnesses {Kronzeuge) - evolved out of earlier experiments in trials of those

accused of terrorism (Greer, 1995). The prosecution case against the so-called Baader-

Meinhof gang hinged upon crown witness testimony provided by an associate of the group,

one Gerhard Muller. However, it was somewhat later, in the mid-1980s, that this type of

evidence became systematically used in the trial process, and then in the context of Neo-nazi

groups (Kolinsky, 1988). These were viewed as somewhat exceptional circumstances and

when it was first mooted that these legal innovations be further adapted to address organised

crime, legal scholars, the Bar and the judiciary declared themselves against (Tak, 1997, p.

12). Pursuant to the Verbrechensbekamfungesgesetz of October 28, 1994 (BGB1 I p. 3186)

provision was made for this type of witness testimony in cases involving participants in

organised crime (Tak, 1997 p. 13). As Tak notes:

The Explanatory Memorandum to the regulation goes into its extraordinary

nature. The regulation violates the principle of legality which in Germany is

the basis of prosecution, as well as the principle of the rule of law. More

particularly, it touches on the principle of equal treatment and endangers the

purpose of criminal administration of justice, namely the irrefutability of legal

order (Unverbruchlinchkeit der Rechtsordnung). All this is, however, taken

for granted with regard to strictly defined forms of organised crime because

they constitute a very serious threat to society. (Tak, 1997, p. 13)

Other scholars echoed this assessment criticising the new laws on the grounds that, very

often, the evidence seemed to lack credibility, there was no corroboration requirement, the

approach was of doubtful efficacy and the legitimacy of the criminal justice system could be

damaged by the appearance of a deal being struck with criminals (Vercher, 1992, p. 284).

Given these grave doubts, the legal framework that has developed for witness protection,

especially as it involves the use of undercover police methods and criminal informants, is

highly circumscribed.

The most comprehensive digest of relevant legislation for Germany

that is currently available was penned by the former Head of Division of Judicial System,

Federal Ministry of Justice (Hilger, 2001). Drawing on this source, a number of points can be

cited that are of particular relevance here. Section 68 of the Code of Criminal Procedure

(StPO) provides for the anonymity of witnesses in cases where their safety is in question. But,

exemption from stating one's identity does not release one from the duty to state in the main

trial (when asked) in what capacity the observations were made. This is aimed specifically at

persons acting in an undercover capacity.

i. The fact that prosecuting authorities employed undercover methods may come to light

during the trial, and if there is a compelling wish to avoid this 'such witnesses will have

to be dispensed with' (Hilger, 2001, p. 100). Section 223 allows such witnesses to be

questioned by the appointed judge via a video link. Section 247 allows that the accused

can be removed from the courtroom 'if the fear exists that a co-accused or a witness will

not tell the truth if questioned in the presence of the accused (ibid. p. 100). In addition, it

also allows for proceedings to be held in camera, and specifically states that the court

has a duty to protect witnesses. Section 96 allows that 'the court may not require an

authority to submit files or information if the highest service authority states that the

disclosure . . . would be disadvantageous to the state' (ibid. p. 101) and that where it is

necessary for the protection of the witness, their anonymity may be preserved, other

individuals (police, public prosecutor, judge) who have questioned the witness during

the investigation to provide evidence on a hearsay basis (ibid. p. 101). Sections 110b

and 110d allow for the new/secret identities of undercover officers in perpetuity. Witness

Protection Act (Zeugenschutzgesetz) (Sections 58a, 168e and 255a), allows for

evidence to be provided via video-link or videotape. In its decision to allow for this the

court is required to balance (i) the duty to effectively detect and control crime in a

manner consistent with the principles of justice; (ii) respect the interest of the accused,

in particular their right to ask comprehensive questions of the witness; and, (iii) the duty

to protect witnesses.

The Dutch experience

The Dutch experience of undercover policing methods was substantially coloured by the

expose provided by the Parliamentary Committee of Enquiry chaired by Maarten van Traa

(den Boer, 1997). Caution about the use of these methods, and efforts to ensure that future

use be circumscribed by the letter of the law need to be read in this light (Tak, 1997).

The Van Traa enquiry examined the methods used by the police against drug traffickers. It

shed light on several cases in which either the police, or informers employed by police,

infiltrated drug rings and, in so doing, committed crimes just as serious as those of the

criminals they were mixing with. According to the findings, the Haarlem police had taken

'unacceptable' initiatives in order to infiltrate international drug trafficking networks. In

The anxiety about the use of informers and police undercover methods in Germany during this period was probably

heightened because of various scandals. In the mid-1990s the German press reported that a secret agent and an

interpreter with the BND had been involved in two cases of entrapment, one concerning Russian plutonium and the

other Colombian cocaine. The plutonium scam was shown to be a set-up which enabled the federal government to

mount an international propaganda campaign about the Russian Mafia peril and the need to bolster the authority of

President Boris Yeltsin at the height of the war in Chechnya. Above all, it enabled the BND to justify the large secret

service budget even though the Cold War had ended. The Colombian cocaine plot was intended to trap traffickers

bringing 330 kg of the drug into Germany. In the end it came to nothing, but nobody knows whether the cocaine did

get into the country or not. The war between the Bavarian police and the federal secret services was at the bottom of

the leaks to the press, especially to the monthly magazine Stern. The magazine reported that the mole Rafael "Rafa"

Ferreiras, unhappy with the amount he was paid for the plutonium scam, pulled out of the cocaine deal.

particular two 'supercops', Klaas Langendoen and Joost van Vondel, whom the press

nicknamed the 'Royal Couple', were controlling monitored deliveries in order to facilitate the

infiltration of criminal organisations using criminal informants. These individuals imported

large amounts of drugs, putatively in order to gain the confidence of organised criminals and

identify people at the top of the criminal world, a practice known as the 'Delta Method'. These

criminal informants were successful because police closed their eyes to the CID-sponsored

operations. Thus protected, the police undercover operation rivalled the criminal very

underworld they were supposed to infiltrate and undermine both in terms of volume and

profits (EIPA, 1996). However, it seems also that many of the informers, and at least some of

the police involved, were personally profiting from the drug smuggling operation.

The scandals that ensued had an international dimension, since they involved police agencies

outside of the Netherlands, including German agents of the Bundeskriminalampt (BKA) and

US agents from the Drugs Enforcement Administration (DEA), who avoided giving evidence

to the Van Traa enquiry under cover of diplomatic immunity. In the face of the refusal of

America's DEA and Germany's BKAto provide information on the level of their involvement in

the operation, the Dutch parliament dropped this aspect of the enquiry.

The so-called IRT scandal created a strong awareness of how badly things could go wrong

with undercover police operations, and how dangerous the use of criminal informants could

be. Subsequent legal innovations in this field have thus been quite cautious and closely

circumscribed.

The Irish experience: Northern Ireland

The experience of Northern Ireland is arguably more fraught than any of the national

experiences discussed previously in this report. The use of the 'supergrass' - that is 'state's'

witnesses' - were introduced there in the 1980s. This was part of a series of initiatives

intended to address the problem of political violence (Bonner, 1993; Donohue, 2001; Gearty,

1991; Greer, 1995, 2001; Hogan and Walker, 1998). The adoption of the supergrass came

on the back of refinements in counter-terrorist intelligence-gathering. Beginning in 1979,

there was a serious attempt by the authorities in Northern Ireland to rationalise intelligence

acquisition using human sources and the idea of 'turning' criminal or terrorist informants and

persuading them to give Queen's evidence was a natural extension of these developments

(Greer, 2001, p. 129). In the middle years of the 1980s some six hundred suspects were

placed under arrest as a consequence of evidence supplied by about twenty-five

supergrasses, who had been offered immunity from prosecution.

One key issue was that of

corroboration, since some individuals were being convicted in Diplock trials on the strength of

criminal informant evidence alone. In his analysis of the arrival and departure of the

'supergrass process,' Greer observed that it:

turned out to be merely a fresh means for identifying defendants, obtaining

confessions, and ensuring that certain key suspects were remanded in

custody, which in a handful of cases, lasted for up to four years as those

concerned were shunted from one supergrass to another before eventually

being acquitted or having their convictions quashed. (ibid. p. 130)

Issues surrounding the lack of corroboration of supergrass testimony cannot be viewed in

isolation. In Diplock trials, where there is no jury and only the judge to decide on the facts,

the traditional rule that the arbiter of facts must be warned about the dangers of drawing

inferences from uncorroborated testimony "in effect merely requires the trial judge to warn

himself to take utmost care in assessing the evidence and drawing inferences from it"

(Bonner, 1993, p. 184). To close observers, the use of the Diplock Court system and the use

At least in the early stages. Public criticism was substantial and full immunity was later replaced by the offer of

reduced sentences and the promise of new identities after serving them.

of uncorroborated testimony from people who were essentially involved in the criminal-

terrorist nexus brought the British system of justice into disrepute. Concerns were raised in

the House of Commons about this to the extent that "When we sit in and observe trials in

[other] countries, our position and the respect in which British law is held in those countries

will be severely diminished and hampered" (quoted in Donohue, 2001, p. 348).

It may seem odd that, in the face of the danger that members of the public faced from

paramilitary violence and terrorist activity, that these legal innovations were thought to bring

British law into disrepute. However, as Gearty (1991) has argued, in the grey area that exists

somewhere between the use of political terror and the response of a strong state to suppress

it, there lies the issue of legitimacy. Legitimacy is undermined when the balance is wrong.

Legitimacy is weak when carefully calculated and restrained subversion meets draconian and

secretive state action, but it is emboldened when hardened terrorism meets a 'culture of

lawfulness' (Godson, 2003), wherein state action is transparent, rule-governed, upright and

dispassionate. The question of legitimacy is not unconnected to the more practical issue of

effectiveness. Greer argues that the effectiveness of the supergrass system in Northern

Ireland was "at best marginal and short-lived" (Greer, 2001, p. 130). He suggested that the

locking up of large numbers of active members of paramilitaries on supergrass evidence may

have had some effect on the terrorist murder rate (and, by extension, on other aspects of

paramilitary action), but he went on to observe that the diminution of paramilitary violence was

equally explicable as the consequence of the political rise of Sinn Fein (ibid. p. 130) and by

the changing tactics of paramilitary groups. Commenting on the problems the 'supergrass

years' created for law enforcement in Northern Ireland, Ellison and Smyth observe that:

The legal system was called into further disrepute and the integrity of the

judiciary questioned as the sight of members of the legal profession earning

vast fees by engaging in supergrass trials was not a pleasant one . . . the

RUC had once again embarked upon a high-risk strategy to defeat the IRA

with no appreciable success apart from damaging their own battered

reputation (Ellison and Smyth. (2000, pp. 113-114)

Bonner (1993) is less condemnatory, arguing that the question of effectiveness is 'highly

complex and controversial' and 'while anti-terrorist measures may play some role', the

Republican pursuit of an electoral strategy may also have had some effect. His conclusion is

"that anti-terrorist measures [he examines a raft of them including economic, legal and

political ones], are likely to be with us for the foreseeable future and must continue to be

strictly scrutinised" (p. 200).

The Irish experience: The Republic of Ireland

The overspill in violence from the civil conflict in Northern Ireland to the Irish Republic has had

important implications for how the Irish Government tackles organised crime. In particular,

there is evidence of the increasing use of the Special Criminal Court for trying organised

crime cases. Established under section 38.3 of the Irish Constitution 'for the trial of offences

in cases where it may be determined in accordance with such law that the ordinary courts are

inadequate to secure the effective administration of justice and the preservation of public

peace and order', the Special Criminal Court comprises three judges and no jury. The main

reason for its current phase of operation, which began in 1972, was a concern that juries in

cases connected with paramilitary activities were likely to be the subject of intimidation. Its

continuation reflects the Irish Government's belief that organised crime poses a threat to

juries equivalent to that posed by paramilitaries. This has, however, proved controversial. In

a recent report of the Committee set up to review the Offences Against the State Acts and

other matters (Department of Justice, Equality and Law Reform, 2002), a majority of the

Committee supported the use of the Special Criminal Court in organised crime cases given

the potential threat to the administration of justice posed by organised criminals. The

Committee concurred with the view put forward by Charleton and McDermott that given 'The

extent to which [organised crime] can grow and dominate society, the arrogance of those

involved with their gangs and their determination not to abide by any rules of decency and

standards makes ... a reasonable case for the measured use of multi-judge, non-jury courts

on an emergency basis' (Charleton and McDermott, 2000, p.141-142). More generally given

that the Special Criminal Court has the same rules of evidence and rights of appeal as other

criminal courts in Ireland and is comparable to other standard, non-jury court systems in other

countries, its use in organised crime cases is viewed as an acceptable measure (Davis,

2003).

A minority of the review Committee, dissented from the view that using Special Criminal

Courts was acceptable in organised crime cases. They drew attention to other common law

jurisdictions, like the United States, England and Australia, where the risks of jury intimidation

in organised crime cases have not provided grounds for non-jury trials in a Special Criminal

Court. Further, there are other measures which could be employed to address the risk of jury

intimidation, including the use of anonymous juries, providing jury protection during a trial, and

allowing a jury to observe the trial remotely via a video link. The minority report of the

Committee concludes that the case in favour of the Special Criminal Court has not been

made and that concerns about jury intimidation can be addressed in other ways: 'if the jury

are anonymous and at a secure and secret location, the risk of effective jury intimidation

would not be very great' (Department of Justice, Equality and Law Reform, 2002, para.9.95).

Given, too, that it is at the discretion of the Director of Public Prosecutions (DPP) as to

whether a person is tried before the Special Criminal Court and that the DPP's decision

cannot be effectively reviewed, the United Nations Human Rights Committee have been

critical of the mechanisms for referring cases to the Court. Specifically, the UN have been

concerned that, 'The law establishing the Special Criminal Court does not specify clearly the

cases which are to be assigned to that Court but leaves it to the broadly defined discretion of

the Director of Public Prosecutions' (Department of Justice, Equality and Law Reform, 2002,

para.9.61). In response to this criticism, however, the Committee reviewing the Offences

against the state Acts recommends that any decisions by the DPP to send an accused to trial

to the Court should be subject to 'a positive review mechanism' (ibid., para. 9.64).

Witness compellability

Although the extensive literature search yielded no materials on witness compellability in

organised crime investigations, there is some discussion in the academic literature on the

significance of compellability in the context of cases of domestic violence in the UK and US

(Cretney and Davis, 1997; Morley and Mullender, 1992). While there are, of course,

fundamental differences between cases involving domestic violence and those involving

organised crime, it is possible to distil from the literature a set of general issues about the use

of compellability that are relevant to informing debate about the relative merits of witness

compellability in organised crime investigations. For example, both rely on the testimony of

someone involved with the offender, that is an accomplice in one case and a spouse or

partner in the other

. These issues are set out in Table 3.2.

It should be clear from Table 3.2 that the arguments for and against the compellability of

witnesses are finely balanced. As Cretney and Davis conclude in the context of their

discussion of domestic violence but which might equally apply to organised crime cases,

Hobbs (1997) has noted that familial and inter-personal relationships are often a key aspect of organised and

serious crime groups. However, it seems unlikely that the witness-defendant relationship will commonly have the

same degree of intensity and complexity (often involving children, shared finances and property) evident in domestic

violence relationships. This may make it relatively easier to manage a witness in organised crime cases.

Conversely, a witness in an organised crime case may have more to fear than retaliation from the defendant, as

there may be a wider network of criminal associates who will seek to intimidate or punish the witness for testifying.

'While compellability may be acceptable in some cases it will never be acceptable in all; while

it may be successful in some cases it will never be successful in all' (p.82). One further point

to emphasise is that while witness compellability might be considered a necessary measure in

organised crime investigations, compulsion must be accompanied by adequate protection and

support. As Victim Support stated in their evidence to a UK Home Affairs committee, it is

'against natural justice to compel a witness who is thereby endangered without offering

protection and support' (quoted in Cretney and Davis, 1997, p.79).

Table 3.2: Arguments for and against the compellability of witnesses

4. Discussion

Emerging from this international review of the literature on witness protection programmes

and crown witnesses in tackling organised crime are a set of overlapping issues which

suggest the need for caution in the development and use of these measures. This concluding

section focuses on the 'costs' and 'effectiveness', and the legitimacy and accountability of

different measures for facilitating witness co-operation.

"Costs' and 'effectiveness'

Despite the widespread use of witness protection programmes and informants to tackle

organised crime, there is remarkably little research evaluating either the 'costs' or the

'effectiveness' of these measures. Terms like 'costs' and 'effectiveness' are, of course, open

to a wide variety of interpretations and this section considers a range of substantive and

conceptual evidence to illustrate some of the issues raised in trying to evaluate the use of

witness protection and informants.

There is some evidence relating to the financial costs of witness protection programmes in

different parts of the world:

• In the US, WITSEC expenditure on witnesses in the program has gone from just over

$17million (£10.5m) in 1997 to $27.5 million (£17m) in 2001. In 2000, the expenditure

was $24million (£15m) and the program contained 7137 primary witnesses and 9091

family members making a total of 16,228 participants (United States Marshals, 2001).

• In one European country, the costs (as of 1999) varied between US$80,000 (c.£50,000)

and US$160,000 (c.£100,000) per participant; in another European country the average

cost for a witness and family of three people was US$80,000 (£50,000) a year and the

average duration of protection was 5 years making total cost US$400,000 (c.£250,000)

(Heijden, 2001).

• In the UK the average costs to protect a witness and their family has been estimated at

between £10,000 and £50,000 but one study found the average cost to be nearer £4,000

(Fyfe, 2001).

• In Australia, the total annual expenditure of the National Witness Protection Program has

ranged from AUS$1,729,500 (c.£670,000) during its first full year of operation (in 1995/6)

to a low of AUS$668, 289 (c.£260,000) in 1998/99. The most recent figures for 2001/02,

during which there were 12 witness protection operations providing assistance to 26

people, show a total expenditure of AUS$773,948 (c.£300,000). For this year, this gives

an average cost, including the witness protection service, of £11,500 per participant

(Australian Federal Police, 2002).

• In Italy the Central Protection Service, which has 700 employees and protects just over

5000 people (1100 are witnesses and the remainder members of their families) has a

budget of approximately 200 billion Lire a year (c. £72 million), which is equivalent to

£14,500 per participant (Italian Ministry of the Interior, 2001).

In term of the 'effectiveness' of witness protection programmes, the available evidence is a

mix of anecdotal observations (largely from self-interested practitioners) and conviction data.

In the US, the federal government has declared that 'The Federal Witness Protection

Program has become so essential that it is difficult to imagine federal law enforcement without

it' (quoted in Earley and Schur, 2002, p.228) while Greer, drawing largely on newspaper

reportage, claims that WITSEC 'had hit the Italian Mafia hard' (2001, p. 126). Similar claims

have been made in Europe. According to Heijden, witness protection programmes in

Germany, Italy and the Netherlands 'are highly effective in the sense that not a single

participant nor a relative of a protected witness has become the victim of an attack by the

source of the threat' (p.6). Echoing the views of Earley and Schur in the US, Heijden (2001)

goes on to note, 'The major effect of witness protection programmes is the conviction of

numerous leaders and other important members of highly organised criminal groups on the

basis of testimonies that otherwise would not have been given'. Interestingly, however, other

researchers have argued that the success in tackling organised crime in places like New

York, has been "accomplished by a shift of reliance away from criminal enforcement (usually

followed by the replacement of one jailed Mafioso with another) and toward a regulatory

strategy" which focused on denying business licences to mob controlled firms (Braithwaite,

2000, p. 229).

In addition to this anecdotal evidence, there are a handful of studies that attempt to

demonstrate the efficacy of witness protection schemes in terms of convictions:

• In the US, the Government Accounting Office reviewed cases that were initiated via

protected witness testimony between 1979-1980. On this period, 1,283 defendants,

mainly involved in organised crime, were charged with crimes from the testimony of 220

protected witnesses resulting in 965 (75%) of the defendants being found guilty

(Montanino, 1990).

• In South Africa between January and June 2002, there were 45 cases involving 49

protected witnesses, resulting in 37 convictions (of which 19 were life terms), there were

six acquittals, and three cases were still awaiting trial/judgement (Adam, 2002).

An important issue in interpreting any conviction data in comparative perspective relates to

differences in the legal systems between countries. In Italy, for example, the minimum

requirements of evidence are less in cases in which the defendant is accused of organised

crime. Furthermore, in such cases the defendant can be convicted on the basis of the

testimonies of two co-defendants who are collaborating with justice, as long as there is some

material evidence as well. However, recent concerns about the validity of the statements of

collaborators has led to a new law which requires collaborators to be isolated from other

witnesses and to declare everything they know about a criminal groups and its activities in a

limited time period (Heijden, 2001, p.2).

The effectiveness of witness protection measures also needs to be viewed from three other

perspectives. First, relocation programmes represent an essentially situational approach to

dealing with the problem of witness intimidation and are focused on the management and

manipulation of the environment to reduce the risk and opportunities for intimidation. Like

other situational measures, witness protection programmes are therefore vulnerable to the

criticism that they displace rather than tackle the problems they seek to address by shifting

the target of intimidation to the families and friends of witnesses or even to jurors. Moreover,

witness protection programmes clearly deal with the symptoms rather than the causes of

intimidation. They are concerned with 'risk minimisation' not the reasons behind witness

intimidation. If progress is to be made in terms of tackling the causes of intimidation then it is

vital to explore the possibility of so-called 'social', rather than situational, approaches to the

problem of crime by addressing some of the underlying reasons that lead to witness

intimidation.

Second, this situational perspective is important because it means that the success or

effectiveness of witness protection programmes is viewed largely in terms of the physical

security of the witness and their participation in the legal process. Indeed, as indicated

above, one measure of the success of these programmes is whether any witness has been

physically harmed or killed while under protection. Yet for relocated witnesses, and other

programme participants, their physical security is a necessary but insufficient condition for

their sense of social and psychological well-being. As studies focusing on the experiences of

those relocated via these programmes graphically illustrate, participants experience chronic

feelings of anxiety and insecurity, causing extreme mental distress. Indeed, there is a tragic

irony that the above average suicide rate for protected witnesses revealed in the US means

that it is often their own anxieties that lead them to endanger their physical safety (Fyfe and

McKay, 2000a and 2000b; Koedam, 1993).

Third, while some have argued that witness protection and immunity for criminal informants

"represents an exit strategy for career criminals" and that programmes such as WITSEC

"provides the criminal with a place in legitimate society" (Kelly, Scharzberg and Chin, 1994, p.

501), the empirical evidence is not so unequivocal. Citing data from the US General

Accounting Office (GAO), Albanese (1996) observed that approximately 21 percent of

protected witnesses were arrested within two years of being put on the programme (p. 195).

In one notorious instance, the US federal government was sued by a woman after her brother

was killed by a 'protected witness' relocated to her city. It is not surprising to find that career

criminals are recidivists and the 20 percent figure suggests that witness protection and

immunity from criminal informants is a risky strategy. Moreover, given the reputation that

many professional criminals have for violence (Hobbs, 1995), it should be considered a high-

risk strategy.

The difficulties of evaluating the 'costs' and 'effectiveness' of witness protection programmes

also apply to crown witnesses. Writing about the UK, Maguire notes that an Audit

Commission report (Audit Commission, 1993) 'confidently concluded that the systematic use

of informants was one of the most cost-effective methods of clearing up crime' but that it 'had

little good evidence for this claim' (Maguire, 2000, p. 330). Another UK perspective on the

effectiveness of informants is provided by Innes (2002) who notes that working police officers

are somewhat dubious about the efficacy of informant-sourced information characterising it as

'too unreliable'. He also notes that 'particular reservations were expressed about a number of

informants who it was suspected were participating in serious criminal activity' (ibid. p. 378).

He characterised the relationship between criminal informants as a 'Faustian pact' (ibid. p.

376) and argued that 'once one progresses beyond the somewhat naive and simplistic cost-

benefit analysis upon which the Audit Commission conclusions are based' the utility of such

tactics become questionable (see also: Morgan and Newburn, 1997). In particular, 'a host of

other economic and non-economic costs relating to the recruitment, cultivation, and running of

an informant become apparent' (ibid. p. 378). In his view, the heavy costs accrued against

the legitimacy of the police and the criminal process must be carefully weighed (see also,

Innes, 2003). Similar conclusions are reached by Dunningham and Norris (1996a, 1996b).

They argue that the claims made about the cost-effectiveness of informers are 'highly

dubious' and that while the police use of informers has been articulated in government policy-

making circles as having few negative aspects, in fact 'negative consequences abound' (ibid,

p. 407). According to them:

...crime is facilitated as well as repressed; criminals are licensed to commit

crime rather than apprehended for their violations; police rule bending is

often organisationally condoned rather than condemned; police morale is

sapped as well as boosted; relationships with colleagues are based on

distrust and secrecy rather than honesty and openness; the courts are

deceived, defendants misled, and in the end justice is as likely to be

undermined rather than promoted. (ibid. p. 407)

Legitimacy and accountability

General concerns

A second set of issues centres on concerns about the legitimacy and accountability of the use

of witness protection programmes, plea bargaining and witness immunity statutes, and the

use of crown witnesses. In the US several commentators have expressed their concerns

about the 'intrinsic morality and fairness to the public' of WITSEC:

If witness testimony is the fundamental weapon in the fight against organised

crime, is protection and relocation of those witnesses through concerted

government-supported subterfuge the most effective method to employ that

weapon? Are there 'reasonable alternatives to carry out the government's duty

of protection that do not involve purposeful misrepresentation, excessive

expenditures, and interference in the lives of innocent third parties?' (Lawson,

1992, p. 1455).

Indeed, Lawson goes on to argue that while the US government should continue to improve

the program, there must also be serious consideration of viable alternatives to WITSEC in

order 'to ensure the proper balance between the public interest in fighting organised crime

and the interest of the public in protection from deceit and violence perpetrated by participants

in the government's Witness Protection Program' (ibid., p. 1459). The possible alternatives to

WITSEC Lawson identifies are:

• Independent relocation and protection arrangements: this would involve a witness

providing testimony in exchange for a lump-sum payment after which a witness would be

required to make his or her own protection arrangements. One advantage of this

approach, Lawson argues, is that 'it disentangles the government from the web of formal

misrepresentation and manipulation that characterises the current Witness Protection

Program' (ibid., p.1457). Although there might be concerns about the safety of witnesses,

Lawson claims there is evidence that fears about the risks of retributive physical harm

have been exaggerated.

• Special Protective Incarceration: the suggestion here is to make use of a network of safe

houses in order 'to reduce the cost both in terms of financial expenditures as well as in

terms of the potential human cost occasioned by infiltrating former criminals back into

society'. Another element to this would be to require witnesses who have committed

crime to serve reduced sentences prior to entering the protection programme.

• Establishment of an independent review board to oversee and/or make recommendations

concerning the acceptability of a witness for the protection programme. The intention

here is to eliminate 'any of the lingering government bias in favour of offering protection to

"important" witnesses without regard to their potential for success' in the protection

programme.

In relation to witness immunity, several arguments have been advanced in the US against

witness immunity, with Menza (1999) describing it as 'unconstitutional, unfair and

unconscionable'. The objections Menza raises to witness immunity laws include:

• Witness immunity undermines the principle of equal protection under the law: Menza

argues that witness immunity violates that Equal Protection Clause of the Fourteenth

Amendment (which provides that no state shall "deny to any person within its

jurisdiction the equal protection of the laws") by allowing unequal enforcement of the

laws. For example, a person who has committed a murder might be granted

immunity if s/he agreed to testify against a person who the prosecutor believes is a

'Mafia boss', and therefore, in the prosecutor's opinion, is more important to convict.

The concern is that a prosecutor has wide discretion in deciding whom to prosecute

and to whom to grant immunity but is not required to explain his/her immunity

decisions to a court and that no standard exists for selection in immunity cases other

than a general assertion that it is 'in the interests of justice'. With no standards to

guide a prosecutor in the exercise of his/her discretion to grant immunity and no

judicial oversight to restrain it, there is, Menza contends, great potential for abuse.

• Witness immunity might be seen as a 'bribe': in 1998, in a judgement that 'sent shock

waves through the national criminal justice community' (Bowman, 1999), a panel of

judges declared in the case of the United States against Singleton that the granting of

leniency to a co-operating witness constituted a bribe and drew attention to the US

Criminal Code which prohibits the giving, offering or promising anything of value to a

witness for-or because of-his/her testimony (Hollis, 2000; Johnston, 1999). In its

judgement the court stated, 'If justice is perverted when a criminal defendant seeks to

buy testimony from a witness, it is no less perverted when the government does so'.

Although this judgement was quickly over-ruled on the basis that the federal bribery

statute does not apply to the government and that discontinuing the pervasive

practice of buying testimony for leniency would jeopardise law enforcement, Menza

argues that the use of offers of leniency or immunity in exchange for testimony may

still undermine public confidence in the legal system.

• Witness immunity may encourage corruption: Menza also contends that immunity

laws may corrupt by encouraging people to give false testimony. 'Does one really

believe', Menza asks, 'that a person whom the prosecutor has accused of, for

example, nineteen murders, would not say anything, true or untrue, on the witness

stand if that "anything" would relieve him or her of appropriate penalties... for those

crimes? ". In support of this position, Menza quotes the Supreme Court which stated

that 'To think that criminals will lie to save their fellows but not to obtain favours from

the prosecution for themselves is indeed to clothe the criminal class with more nobility

than one might expect to find in the public at large'.

In terms of 'crown witnesses', studies in the United States by academic lawyers tend to

emphasise the moral hazards of police use of informants giving state's evidence (e.g.

Schreiber, 2001; Zimmerman, 1994). These studies catalogue the many instances where

federal law enforcement agencies and prosecutors offices in the United States have engaged

in questionable, unethical and illegal conduct involving criminal informants. They criticise

curative suggestions from law enforcement such as the issuance of new guidelines, training,

lists of do's and don'ts, and anti-corruption investigations as being unproductive of meaningful

results. They also offer considerable criticism of attempts to curtail misuse of informants in

the trial process (which happen, very often, with the collusion of prosecuting attorneys). This

tack on reform suggests that expanded defence cross-examinations of informants, prior

judicial review of informant use, and expansion of habeas corpus relief for those convicted

with perjured testimony would provide sufficient safeguards. According to its critics, this

rights-based approach maintains the assumption that the basic informant-handler relationship

is not problematic. While acknowledging that the rights-based approach might contribute to

identifying violations of specific rights resulting from the mishandling or misconduct of

informants in specific cases, they argue that, absent is a thorough understanding of the moral

hazards that infuse the informant-handler relationship and this approach fails at a systemic

level. They conclude that, without an effective deterrent, informant mishandling and

misconduct are destined to continue.

Similarly, Schreiber (2001) notes that the anonymity that confidential informants enjoy

presents some of the most vexing issues in law enforcement. She draws attention to the

criticisms that can be made with regard to the role that criminal informants may play in the

instigation of criminal activity, their questionable motivations for providing information, and

their credibility as witnesses. Zimmerman, (1994) cites cases where a multitude of benefits

(including monetary payments, immunity from prosecution, and the anonymity that placement

on a witness protection programme provides) produces ample motives for giving detailed

information 'regardless of the truth'. In concluding his exhaustive analysis of the moral

hazards that pertain, Zimmerman argues that the key to stopping the abuses of the informant

relationship is to 'apply a rebuttable presumption that informant conduct is state action and

action under the color of law'. According to his analysis, this will 'instil responsibility in law

enforcement agencies for their decision to use informants' and will 'have a significant impact

on how courts view informants' since it creates a linkage between government and informant

that will force law enforcement to take responsibility for the use of informants'.

The European Convention on Human Rights

In assessing the legitimacy of measures to facilitate witness co-operation it is also important

to acknowledge the importance of the European Convention on Human Rights (ECHR) and

specifically Article 6 concerning the guarantee of a fair trial. According to Tak (1997), in a

series of cases that have come before the European Commission on Human Rights relating

to the use of crown witnesses in the UK, Italy and the Netherlands, the Commission has

assumed that using evidence from accomplices in exchange for a promise of impunity could

be in breach of Article 6. However, in the specific cases considered this has not been an

issue for a combination of reasons:

• the Prosecution Service has been open with regard to a deal with witnesses from the

beginning;

• the statements of witnesses were corroborated by other evidence;

• the defence had been able to interrogate the crown witness in order to establish their

credibility; and,

• the courts had thoroughly examined the evidence.

Tak concludes that the ECHR 'has given the green light to the institution of the crown witness

if only it is used openly and carefully' (1997, p.26).

In addition, concerns have been raised in the UK about the way in which protective measures

designed to assist vulnerable and intimidated witnesses give evidence (such as the use of

video-links or screens during a trial) might breach the Convention's guarantee of a fair trial

and, in particular, the right of the defendant to examine witnesses against him under Article

6(3)d. According to Hoyano (2001) the crux of the problem is this:

The common law adversarial trial model, unlike the inquisitorial trial

systems familiar to most judges on the European Court of Human Rights

places primacy on the rights of the defendant, as the only person in

jeopardy of punishment in the trial. Therefore it is intrinsically

incompatible with that model to balance those rights against the interests

of other participants in the criminal justice system, and specifically

witnesses. It is feared that special protective measures for witnesses will

inevitably collide with the distinctive defendant-centred conception of

justice.

In assessing these concerns, however, Hoyano makes two important observations. First, in

relation to the admissibility of evidence and the mode of taking and assessing evidence, the

stance of the European Court is that these are matters for regulation under national law and

its remit is confined to considering whether the trial as a whole was fair. Second, although

Article 6 does not explicitly require the interests of witnesses called upon to testify to be taken

into consideration, the interests of witnesses and victims are in principle covered by other

substantive provisions of the convention, given that their life, liberty or security of person may

be at stake. States are therefore expected to organise their criminal proceedings in such a

way that those interests are not 'unjustifiably imperilled'. Thus the principles of a fair trial

require that in appropriate cases the interests of the defence are balanced against those of

witnesses called upon to testify. Hoyano therefore concludes that measures designed to

assist vulnerable and intimidated witnesses being introduced into UK courts, like the physical

separation of the accused from the witness, the use of screens and live video-links for cross-

examination are ECHR-compliant. 'Affording witnesses some protection by mitigating the

rigours of the orthodox adversarial trial does not', she observes, 'necessarily mean that one is

hollowing out the defendant's rights'.

Accountability

The use of witness protection programmes and informants also raises important issues about

accountability. In terms of witness protection there are clearly major differences between

countries in the ways in which the operation of witness protection programmes can be held to

account. As section three of this report has shown, in some countries (such as the US and

those included in Tables 1.1 and 1.2) witness protection arrangements are formally

accountable in law, whereas in other countries (such as the UK), there are no specific

legislative provisions covering witness protection. Indeed, in the UK the government

specifically rejected the need for a legal framework for witness protection on the basis that

informal arrangements are working satisfactorily (Home Office, 1998). Other countries,

however, have moved to formalise existing informal arrangements. In Canada, a Witness

Protection Program Act was passed in 1996, twelve years after the establishment of a

Witness Protection Program. According to the Solicitor General, the intention behind the Act

is to ensure the program can:

operate more openly and effectively by establishing a solid legislative

and regulatory foundation for the program. ... The proposed changes will

help ensure that both applicants who enter the program and the RCMP,

who operate the program, have a clear understanding of their rights and

obligations, and the extent and scope of the benefits and protection to be

provided. This should go a long way to eliminating any potential

misunderstandings between the RCMP and protected individuals.

(Solicitor General, 1996)

Several countries also require reports to be published on the operation of their witness

protection programmes. In Canada and Australia, these are annual reports which must be

submitted by the chief of the federal police forces to their respective parliaments; in the US,

the Office of Enforcement Operations is required to submit a quarterly report to the Deputy

Attorney General detailing the results of the testimony provided by relocated witnesses.

There is also a question of how participants on witness protection programmes hold those

operating the programmes to account given several high profile examples in the US and

elsewhere of witnesses claiming they were let down by those protecting them (see Earley and

Schur, 2002).

Accountability has also emerged as an important issue in the police use of informants. For

South (2001), 'one of the key unresolved issues ... in relation to accountability and justice is

the tension between the need for transparency versus dispute over the disclosure of evidence

in a court trial' (p. 76). He argues further that:

This generally hinges on the arguments that, on the one hand, withholding of

prosecution evidence from the defence can evidently lead to miscarriages of

justice, versus the proposition that such disclosure would lead to the

identification of the informer and place them in danger. This is a fundamental

aspect of the overall problem of accountability. On the one hand, justice and

the courts might reasonably expect that evidence submitted be open to

further questioning; on the other hand, courts are also likely to accept the

arguments of prosecution and police that sources must be protected 'in the

public interest'. In various jurisdictions, including the UK, there have been

recent initiatives to try to resolve these tensions. Nonetheless, the problem

remains. (ibid. p. 77)

Conclusion

Without witnesses, the rudiments of prosecution, such as identifying the

accused and establishing the requisite nexus between the accused and the

crime, would become insurmountable obstacles to conviction, and the criminal

justice system would cease to function.

(Harris, 1991, p. 1285)

Given that the evidence provided by witnesses is vital to the effective investigation of crime

and in building a case against the accused, facilitating witness co-operation is a key objective

of all criminal justice systems. In the context of organised crime, however, the serious

problems posed by witness intimidation and the importance of accomplice testimony provided

by 'crown witnesses' make the issue of witness co-operation particularly acute. In recent

years, however, an increasing number of countries have introduced measures allowing a

witness-centred approach to develop to the investigation and prosecution of organised crime.

As this report has illustrated, there are significant variations between countries in terms of

when and what types of measures they have introduced, partly reflecting differences in the

scale and nature of the organised crime problem they face and partly linked to differences in

the legal traditions and environments of different countries. In the United States, for example,

measures like witness protection and plea bargaining are well established; France, by

contrast, is only now considering introducing plea bargaining but in the face of strong

opposition from magistrates and lawyers who have denounced it as a 'decline in the rights of

the defence and the presumption of innocence, and the ... marginalization of the function of

judges in favour of ever more powerful prosecutors' (Tagliabue, 2003, p.A14). The measures

discussed in this report also vary considerably in terms of their spatial and temporal scope

and in terms of the range of individuals and agencies involved. Plea bargaining and

discussions of witness immunity are, for example, typically focused around the trial phase of

the criminal justice process and involve a narrow range of actors, such as judges, prosecutors

and defence teams. Witness protection arrangements, by contrast, are normally introduced

early on in an investigation, typically provide long-term support in a variety of different

locations, and involve a wide range of agencies both within and outside the criminal justice

system. Despite these differences between the various measures considered here, however,

there clearly needs to be a degree of integration between them if they are to be effective.

Offers of witness immunity or compelling individuals to testify may only be successful if there

are arrangements to provide witnesses with long-term protection.

The increasing development and use of different measures to facilitate witness co-operation

in organised crime investigations also raises key questions about how to measure the

'success' and 'effectiveness' of these approaches. The evidence-base on which to judge their

efficacy, however, is very weak and there remains a real need for empirical research

assessing the application of these approaches. Indeed, although there are clearly difficulties

of researching in this field it is reasonable to concur with Bean (2001) that no policing strategy

should remain so secretive that it remains outside the boundaries of critical evaluation. As

the discussion in this chapter has indicated, any research agenda focused on issues of 'costs'

and 'effectiveness' also needs to consider questions of legitimacy and accountability. Given

that the overwhelming majority of witnesses on protection programmes and those involved in

investigations and prosecutions as 'crown witnesses' are themselves criminals, the potential

for scandal is ever-present and needs to be guarded against. If these techniques are to be

utilised, therefore, they should ideally be employed in the context of enhanced evidence as to

their effectiveness, and under the umbrella of transparent legislative guidelines as to their

appropriate use.

References

Adam, D. (2002) Keynote address at National Prosecuting Authority, Witness Protection Unit

Strategy Workshop, South Africa.

Albanese, J.S. (1996) Organized Crime in America (3

edition). Cincinnati, OH: Anderson

Publishing.

Audit Commission, (1993) Helping with enquires; tackling crime effectively, London: HMSO.

Australian Federal Police (2002) Memorandum of arrangement concerning the provision of

services under the National Witness Protection Program. Canberra: Australian Federal Police.

Bean, P. (2001) 'Informers and Witness Protection Schemes', in R. Billingsley, T. Nemit and

P. Bean (eds.) Informers; policing, policy and practice. Devon: Willan Publishing.

Boer, M den (1997) Undercover policing and accountability from an international perspective.

Maastricht: European Institute of Public Administration.

Bonner, D. (1993) 'The United Kingdom's Response to Terrorism' in A.P. Schmidt and R.D.

Crelinsten (eds.) Western Responses to Terrorism. London: Frank Cass.

Bowman, F.O. (1999) Departing is such sweet sorrow: a year of judicial revolt on 'substantial

assistance' departures follows a decade of prosecutorial indiscipline', Stetson Law Review,

Summer.

Braithwaite, J. (2000) 'The New Regulatory State and the Transformation of Criminology',

The British Journal of Criminology, 40(2): 222-238.

Carroll, R. (2001) 'How Italy betrays innocent witnesses against the Mafia', in The Guardian,

April 9, 2001.

Charleton, P. and McDermott, P.A. (2000) 'Constitutional Aspects of Non-Jury Courts', Bar

Review, 106 (Part I).

Council of the European Union (1995) 'Resolution of the Council of 23 November 1995 on

the protection of witnesses in the fight against international organised crime', Official Journal

of the European Communities, C 327.

Council of the European Union (1997) 'Council resolution of 20 December 1996 on

individuals who co-operate with the judicial process in the fight against international organised

crime', Official Journal of the European Communities, 97/C 010.

Cretney, A. and Davis, G. (1997) 'The significance of compellability in the prosecution of

domestic assault', British Journal of Criminology, 37(1): 75-89.

Davis, F. (2003) 'The strengths and weaknesses of the Special Criminal Court' (personal

communication, September, 2003).

Department of Justice, Equality and Law Reform (2002) Report of the Committee to

review the Offences against the State Acts 1939-1998 and related matters (Dublin).

Donohue, L.K. (2001) Counter-Terrorist Law and Emergency Powers in the United Kingdom

1922-2000, Dublin: Irish Academic Press.

Dunningham, C. and Norris, C. (1996a) 'The Detective, the Snout and the Audit

Commission; the real costs of using informants', The Howard Journal, 38(1): 67-86.

Dunningham, C. and Norris, C. (1996b) 'The Nark's Game', (Parts 1 & 2), New Law Journal,

146, 6736: 402-404; and 146, 6737: 456-457.

Earley, P. and Schur G. (2002) WITSEC: inside the Federal Witness Protection Program,

New York: Bantam Books.

Ellison, G. and Smyth, J. (2000) The Crowned Harp; Policing Northern Ireland, London:

Pluto Press.

Europol (1999) Europol's contribution and future role within the framework of witness security

programmes of member states of the European Union, Europol, 2510-21.

Europol (2003) Witness protection: common criteria for taking a witness into a protection

programme, Europol, 2510-82 rev 3.

Ewing, K.K. (2000) 'Establishing an equal playing field for criminal defendants in the

aftermath of United States versus Singleton", Duke Law Journal, March.

Fyfe, N.R. and McKay, H. (2000a) 'Witness intimidation, forced migration and resettlement: a

British case study', Transactions of the Institute of British Geographers, NS25: 77-90.

Fyfe, N.R.and McKay, H. (2000b) 'Desperately seeking safety: witnesses' experiences of

intimidation, protection and relocation', British Journal of Criminology, 40(4): 671-687.

Fyfe, N.R. (2001) Protecting intimidated witnesses, Ashgate: Aldershot.

Gabor, T., Baril, M., Cusson, M., Elie, D., LeBlanc, M. and Normandeau, A. (1987) Armed

Robbery: Cops, Robbers, and Victims, Springfield, Illinois: Charles C Thomas.

Gearty, C. (1991) Terror, London: Faber and Faber.

Godson, R. (2003) Menace to Society: political-criminal collaboration around the world, New

Brunswick, NJ.: Transaction Publishers.

Greer, S. (1995) Supergrasses: a study of anti-terrorist law enforcement in Northern Ireland,

Oxford: Clarendon Press.

Greer, S. (2001) 'Where the grass is greener?; Supergrasses in comparative perspective', in

R. Billingsley, T. Nemit and P. Bean (eds.) Informers; policing, policy and practice. Devon:

Willan Publishing.

Hampson, F.J. (1998) 'The international Criminal Tribunal for the former Yugoslavia and the

reluctant witness,' International and Comparative Law Quarterly, 47(1): 50-74.

Harris, R.J. (1991) 'Whither the witness? The Federal Government's Special Duty of

Protection in Criminal Proceedings after Piechowicz v United States', Cornell Law Review,

76(6): 1285-1316.

Heijden, T. van der (2001) Witness Protection Programmes Compared, a paper given at the

Second World Conference on the Investigation of Crime; Modern Criminal Investigation,

Organized Crime and Human Rights, Durban, Dec. 3-7, 2001 (available at

www.legal.coe.int/economiccrime/organisedcrime/BestPractice1E.pdf.).

Hilger, J.P.W. (2001) 'Organized Crime and Witness Protection in Germany", in Effective

Methods to Combat Transnational Organized Crime in Criminal Justice Processes, Tokyo:

Asia and Far East Institute for the Prevention of Crime and the Treatment of Offenders

(UNAFEI): 99-105.

Hobbs, D. (1995) Bad business: Professional Crime in Modern Britain, Oxford: Clarendon.

Hobbs, D. (1997) 'Professional Crime: Change, Continuity and the Enduring Myth of the

Underworld', Sociology, 31(1): 57-72.

Hogan, G. and Walker, C. (1998) Political Violence and the law in Northern Ireland,

Manchester: Manchester University Press.

Hollis, T. (2000) 'An offer you can't refuse? United States V. Singleton and the effects of

witness/prosecutorial agreements', Boston University Public Interest Law Journal, 433.

Home Office (1998) Speaking Up for Justice: report of the interdepartmental working group

on the treatment of vulnerable and intimidated witnesses in the criminal justice system,

London: Home Office.

Hoyano, L.C.H. (2001) 'Striking a balance between the rights of defendants and vulnerable

witnesses: will special measures directions contravene guarantees of a fair trial?', Criminal

Law Review.

Illinois Criminal Justice Information Authority, (1997) Evaluation of the Illinois State Police

Gang Crime Witness Protection Program, Chicago: Illinois Criminal Justice Information

Authority.

Innes, M. (2002) 'Professionalising the role of the police informant', Policing and Society,

9(4): 357-384.

Innes, M. (2003) Investigating Murder; detective work and the police response to criminal

homicide, Oxford: Clarendon Press.

Italian Ministry of the Interior (2001) The witness protection scheme and the Servizio

Centrale di protezione, Dipartimento della Pubblica Sicurezza: Rome.

Jeffries, J.C. and Gleeson, J. (1995) 'The federalization of organized crime: advantages of

federal prosecution', Hastings Law Journal, April.

Johnston, J.R. (1999) 'Plea bargaining in exchange for testimony: has Singleton really

resolved the issues?', Criminal Justice, 24.

Klipp, A. (1996) 'Witnesses before the international tribunal for the former Yugoslavia', Revue

internationale de droit penal, 67,1(2): 267-295.

Kelly, R.J., Schatzberg, R. and Chin, K-L. (1994) 'Without Fear of Retribution: The Witness

Security Program', in R.J. Kelly, Ko-Lin Chin and Rufus Schatzberg (eds.), The Handbook of

Organized Crime in the United States, Westport Conn.: Greenwood Press.

Koedam, W.S. (1993) 'Clinical considerations in treating participants in the Federal witness

protection program', The American Journal of Family Therapy, 21(4): Winter 1993.

Kolinsky, E. (1988) 'Terrorism in West Germany' in J. Lodge (ed.), The Threat of Terrorism,

Brighton: Harvester Wheatsheaf.

Levi, M. and Smith, A. (2002) A comparative analysis of organised crime conspiracy

legislation and practice and their relevance to England and Wales, London: Home Office.

Lawson, R.J. (1992) 'Lying, cheating and stealing at Government expense: striking a balance

between the public interest and the interests of the public in the witness protection program',

Arizona State Law Journal, 24: 1429-1459.

Maguire, M. (2000) 'Policing by Risks and Targets: some Dimensions and Implications of

Intelligence-Led Crime Control', Policing and Society, 9(4): 315-337.

McCusker, E. (2002) 'Protecting the truth': comparative analysis of witness protection

services, Hague: United Nations International Criminal Tribunal for the Former Yugoslavia.

Menza, A.J. (1999) 'Witness immunity: unconstitutional, unfair, unconscionable', Setron Hall

Constitutional Law Journal, Spring.

Mittermaier, C. (1851) Das englische, schottische und nordamerikanische Strafverfahren: im

Zusammenhange mit den politischen, sittlichen und socialen Zustanden und in den

Einzelheiten der Rechtsubung.

Montanino, F. (1990) 'Protecting organised crime witnesses in the United States',

International Journal of Comparative and Applied Criminal Justice, 14(1): 123-131.

Morgan, R. and Newburn, T. (1997) The Future of Policing, Oxford: Oxford University Press.

Morley R. and Mullender A. (1992) 'Hype or Hope? The importation of pro-arrest policies

and batterers' programmes from North America to Britain as key measures for preventing

violence against women in the home', International Journal of Law and the Family, 6: 265-88.

Norris, C. and Dunningham, C. (2000) 'Subterranean Blues; conflict as unintended

consequence of the police use of informers', Policing and Society, 9(4): 357-385.

Radzinowicz, L. (1956) A History of English Criminal Law and Its Administration from 1750

Vol. 2, The clash between private initiative and public interest in the enforcement of the law,

London: Stevens.

Roberts-Smith, L. (2000J Review of the Western Australia Police Witness Protection

Program, Volume 1 Perth: Government of Western Australia.

Schreiber, A.J. (2001) 'Dealing with the Devil: an examination of the FBI's troubled

relationship with its confidential informants', Columbia Journal of Law and Social Problems,

34.

Schur, G. (1988)'A father's dream come true', The Pentacle, 8(1): 13-14.

Silke, A. (2001) 'The Impact of Witnesses on Violent Offenders: Social Facilitation and

Deindividuation Effects in Violent Interpersonal Assaults', paper presented at the 11th

European Conference of Psychology and Law, June 2001, Lisbon, Portugal.

Solicitor General (1996) Witness Protection Program Act proclaimed into law, Press release,

Solicitor General of Canada.

South, N. (2001) 'Informers, Agents and Accountability', in Informers; policing, policy,

practice^ R. Billingsley, T. Nemitz and P. Bean (eds.), Devon: Willan.

Tagliabue, J. (2003) 'France proposes adoption of plea bargaining in legal overhaul', New

York Times, April 10, 2003, pA14.

Tak, P.J.P. (1997) 'Deals with Criminals: Supergrasses, Crown Witnesses and Pentiti',

European Journal of Crime, Criminal Law and Criminal Justice, 5(1): 2-26

United States Marshals (2001) 'The Witness Security Programme', Conference

presentation, Scottish Police College.

Vercher, A. (1992) Terrorism in Europe: an international comparative legal analysis, Oxford:

Oxford University Press.

Zimmerman, C.S. (1994) 'Toward a New Vision of informants: A History of Abuses and

Suggestions for Reform', Hastings Constitutional Law Quarterly, 22(1): 81-178.

Appendix

Databases consulted to identify literature for inclusion in the review

• Bath Information Data Services (BIDS) International Bibliography of the Social

Sciences (IBSS): covers the core social science disciplines and draws data from

over 2400 international social science journals and 7000 books a year;

• LegalTrac: provides international coverage of all major law reviews, law journals,

speciality law and bar association journals and legal newspapers;

• LexisNexis: provides international coverage of legislation, case law and legal

journals with particular emphasis on EU, US and Commonwealth materials;

• ProQuest: provides access to a range of data bases including Academic Research

Library (covering 2300 journals), Academic Research Newspapers (including New

York Times) and Social Science Plus (covering social science journals);

• Sociological Abstracts: provides international coverage of abstracts from

sociological, including criminological, journals;

• SOSIG (Social Science Information Gateway): provides internet information for

researchers and practitioners in the social sciences and law;

• Westlaw: provides mainly EU and US coverage of legislation, case law and legal

journals.

Produced by the Research Development and Statistics Directorate, Home Office

This document is available only in Adobe Portable Document Format (PDF) through the RDS

website

Home Office

Research, Development and Statistics Directorate

Communication Development Unit

Email: [email protected]k

ISBN 1 84473 650 4