. 12
( 34)


• Twenty-three cars parked wholly or partly on the footway.
• One lorry with its lowered tailboard hiding the rear number plate while in
• A furniture van with its rear number chalked on the back.
• One ¬‚ag-day girl shaking a collecting box in people™s faces.
• One boy throwing a half-eaten egg sandwich into the roadway.
• Three shop awnings that you had to duck under.
• A cycling window cleaner carrying a ladder on his shoulders.
• And a painter on a window sill wearing no means of preventing a fall.

I would say it was a typical lot and among the things too numerous to count were
cars bearing advertising ˜stickers™, vehicles waiting on double yellow lines, and
disembarking bus passengers throwing their tickets away.
I™ve known policemen who would have hated to let any of these escape. They
would all have been seen as personal a¬ronts, but even a policeman like that
couldn™t have coped with more than one of them. If he chose the cycling man with
the ladder, who was actually the most dangerous, all the rest would have got away.
So perhaps he would have chosen the three shop blinds, on the ground that they
might have knocked his helmet o¬ (they ought to be 8ft 6in from the ground).
But it has to be faced that the great majority of policemen would have chosen
none of them. Which in itself would have been a choice and the chief constable™s
˜discretion™ is merely the same choice writ large, with the di¬erence that chief
constables, who have no time to go around looking for car licences, number
plates, ¬‚ag-day o¬ences, litter bugs and men on window sills, don™t exercise their
choice until an o¬ence is actually reported and the papers come before them.

A more systematic exploration of the discretionary element in policing was part
of a study of policing conducted in the late 1960s by John Lambert. What he
wrote then is as relevant today:

J. Lambert, ˜The Police Can Choose™, New Society, 18 September 1969
The policeman is not, and never has been, simply a ˜law enforcement o¬cer™. He
has discretion, in almost all circumstances except catching a murderer actually
on the job, about whom he will arrest, investigate or harass, and whom he won™t.
In this sense, the problem is that of ˜normal™ policing, because in this exercise of
discretion, which is central to all his work, the policeman™s own private view of
the world comes into play: his opinion, as a citizen, of other citizens; his reac-
tion, as a member of one class or race, towards other classes or races.
The part that discretion plays, necessarily, in British police work is seldom
acknowledged publicly. The legal philosophy of a democratic society sees police
activities as potentially threatening to individual liberty. So the police, in enforc-
ing the law, are themselves bound by numerous regulations. The theory is that
243 The prosecution process

laws apply to all men and the police must enforce the law always, everywhere
Yet full enforcement is not possible. Law-breaking is so common that to inves-
tigate every infringement, to prosecute every known o¬ender, would require
police forces of a size, and involve expenditures on a scale, that would be imprac-
ticable and intolerable. So small police forces with small budgets have to enforce
laws selectively. Both as an organisation and as individuals, the police have con-
siderable choice about how to organise, which crimes and criminals to prose-
cute, how to allocate what number of men to di¬erent law enforcement tasks,
and so on. It™s almost a question of artistry, and certainly it™s craftsmanship . . .
Crime occurs unevenly in di¬erent neighbourhoods of towns and cities. This
puts more policemen in some areas than others, with di¬erent opportunities to
discover o¬ences and to ¬nd o¬enders to process as clients. Opportunities
abound for legalistic policemen to cram police cells with the drunk and disor-
derly and police o¬ces with papers relating to motoring o¬ences. In practice,
many are seen but few are processed. How the drunk or the motorist reacts to
the policeman™s intervention determines the outcome.
What matters is whether the client shows deference or respect to the police-
man and which of these is shown depends very much on a two-way perception
of status, rank, position and power between policeman and client. Thus law
enforcement depends quite precisely on relations between police and public.
The legal role of the police is de¬ned by the perceptions of policemen and their
ability to manage relationships.

Class bias in prosecutions
Research conducted by Professor Andrew Sanders, then of Birmingham
University, explored the question of how far prosecution policy was in¬‚uenced
by class bias. He took 1,200 (non-motoring) cases from six police divisions (two
from a large metropolitan county force, two from a small rural force and two
from a force that policed both rural areas and a city). He compared police deci-
sions with those of non-police agencies and especially the Factory Inspectorate
(HMFI). (See A. Sanders, ˜Class Bias in Prosecutions™, The Hansard Journal, vol.
24, 1985, p. 176.)
He found:

• That the police cautioned very little “ in about 4 per cent of cases, whereas the
HMFI used cautions as the norm “ in the period 1978“1982 cautions were
used in 65“73 per cent of cases.
• The overwhelming majority of HMFI prosecutions are directed at the
middle class, that is companies and managers, whereas most police prose-
cutions are of working class or unemployed (previously working class)
persons. Despite considerable demographic di¬erences, the police divisions
produced very similar class patterns with around only 5 per cent from the
middle or upper class. (Also a statistically signi¬cant di¬erence emerged in
the ability of middle class persons suspected of crime to avoid prosecution
by the police.)
244 Pre-trial criminal proceedings

• In the police, the decision to prosecute is taken at a relatively junior level
(inspector), while the decision not to prosecute is made at a relatively high
level (superintendent or above). In the HMFI it is the other way round. The
junior (inspector) can caution but it requires a more senior o¬cer to institute
a prosecution. Prosecutions are only started by the HMFI for what are
regarded as the most serious of the serious cases, whereas the police often
prosecute trivial cases.
• The police do not take poverty into account when deciding whether to pros-
ecute. By contrast, if a factory owner or trader says he broke the law because
he could not a¬ord to comply, it would be taken into account and would be
regarded as a valid reason for not prosecuting.
• Whereas in the police there was an institutional bias in favour of prosecution
(re¬‚ected in the phrase ˜let the court decide™), the HMFI regarded prosecu-
tions as a last resort. The role of HMFI o¬cials was one more of advice and
persuasion, getting ¬rms to comply with the law. The police saw their role in
prosecutions as such.

In Sanders™ view the di¬erent decisions of the agencies were the result not of the
people concerned but rather of the perspective of the agencies. The police in
some circumstances behaved in a similar way “ e.g. in fraud cases where it was
widely agreed that prosecution was thought of as a last resort for the real rogues.
In tax evasion, too, very few are prosecuted. In 1980 there were 22,000 serious
cases of tax evasion. One in 122 was prosecuted. By contrast there were 107,000
social security frauds, of which one in four were prosecuted. The total value of
social security fraud was estimated at some £108 million in 1979 compared to
£3“3.5 billion in tax evasion in the same year. Tax evasion therefore resulted in
thirty times more loss to the public purse and yet there were far more prosecu-
tions of social security fraud. On when the Crown Prosecution Service can pros-
ecute, although the Inland Revenue has accepted a pecuniary settlement, see S.
Elwes and R. Clutterbuck, ˜Tax and Criminal Prosecutions™, Criminal Law
Review, 1999, pp. 139“43.
On class bias see also the research sample reported by McConville, Sanders
and Leng in The Case for the Prosecution (Routledge, 1991) p. 123.
For evidence of remarkable regional di¬erences in prosecution policies see
research by Dr G. Slapper referred to at p. 268 below.

Proposals for an independent prosecution process
In most countries the decision whether to initiate a criminal prosecution is
taken not by the police but by the prosecutor (in the USA, the district attorney,
in Scotland, the procurator ¬scal and in France, the procureur or parquet). Until
1986, England was one of the few countries in the world where the decision was
taken by the police. The power of the police went so far that in the magistrates™
courts the police themselves often actually conducted the prosecution. Where
245 The prosecution process

they did not conduct the prosecution they instructed either solicitors or solici-
tors and barristers. In the Crown Court where neither they nor solicitors had the
right to appear as advocates, they had to instruct both solicitors and barristers.
This was changed by the Prosecution of O¬ences Act 1985 which set up the
Crown Prosecution Service (CPS). The police still initiated the charge but the
question whether the case was continued and, if so, on what charges was trans-
ferred to the CPS and it was the CPS which carried the prosecution forward. The
change represented a revolution in both the principles and the actual practice of
prosecutions. The Criminal Justice Act 2003 took the matter further by introduc-
ing the new principle that the CPS rather than the police formulate the initial
charge. The police power to lay the charge now only applies in routine minor cases
and in serious cases where an early holding charge is required to justify detention.
The origin of the CPS was the 1970 JUSTICE report The Prosecution Process
in England and Wales, the main thrust of which was that even the honest, con-
scientious police o¬cer may become psychologically committed to successful
prosecution. ˜He wants to prosecute and he wants to win™. He is therefore more
likely to continue with a prosecution where the evidence may be weak. Also, the
police were not well suited to evaluate the public policy aspects of the discretion
not to prosecute. The police should not both collect the evidence and conduct
the proceedings. The Committee recommended the introduction of the
Scottish system where the decision to prosecute in all but very minor cases is
taken by the procurators ¬scal, under the Lord Advocate, wholly independent
of the police. The police report all cases to the procurators ¬scal, who decide
whether a prosecution is warranted.

The Philips Royal Commission™s report
The Philips Royal Commission agreed that a new system was needed but it did
not go so far as JUSTICE in recommending the Scottish model. It thought that
the initial decision to charge a suspect should continue to be taken by the police
but that thereafter all decisions, including any decision to alter or drop the
charges, would be taken by a prosecution agency. Each area would have a pros-
ecuting solicitors™ department presided over by a Crown Prosecutor of equal
status to the chief constable and answerable to the same authority. Each police
authority area should have a new committee to be known as the Police and
Prosecution Authority, to which the Crown Prosecutor and the chief constable
would both be accountable. The minister responsible for the prosecution
system should be either the Home Secretary or the Attorney General.

The Government™s response
The Government rejected the Royal Commission™s view that the new prosecu-
tion system would be based on local committees but otherwise it accepted the
thrust of the Royal Commission™s recommendation.316 Its White Paper stated

For the history of the proposals see the 6th edition of this work, p. 220.
246 Pre-trial criminal proceedings

that there should be a single national prosecution service, controlled and
directed by the Director of Public Prosecutions. The investigation of criminal
o¬ences would remain with the police and they would continue to lay the initial
charges but thereafter the responsibility for all prosecution decisions (including
the dropping or alteration of charges) would be that of the prosecution

The Crown Prosecution Service (CPS)
The Crown Prosecution Act 1985 (CPA 1985) established the Crown Prosecution
Service as a national prosecution service for the whole of England and Wales
under the general direction of the Director of Public Prosecutions (DPP).
The country was originally divided into areas “ twenty-nine in England and
two in Wales “ with each area headed by a chief Crown Prosecutor, responsible
to the DPP for the operation of his area. In 1993 this structure was altered. The
thirty-one areas were amalgamated into thirteen larger areas, including one for
the whole of London.
The function of the CPS is to conduct all criminal cases against both adults
and juveniles (apart from minor motoring o¬ences which have been excluded
from the system) that are instituted by or on behalf of the police.
It was also given a power in CPA 1985, s. 23 to discontinue proceedings which
is used very often (see further below). This is in addition to the power not to
start proceedings where charges have been laid. In summary cases before the
court has heard any evidence, and in proceedings in the Crown Court before it
or the magistrates™ court has heard any evidence, the proceedings can be
stopped by notice to the court with reasons. They can also be stopped where
someone has been arrested without a warrant before the court has been
informed of the charge by notice to the suspect.
The CPS does not, however, have its own investigation machinery or facili-
ties. It relies for that role on the police, but unlike the procurator ¬scal in
Scotland, the CPS cannot direct the police to carry out an investigation or
further investigations. It can only request. The Runciman Royal Commission
considered whether the CPS should be given such a power but by a majority of
ten to one decided against it. Any dispute between the CPS and the police as to
such a question should, it thought, be resolvable between the two agencies, if
necessary with the help of the Chief Inspector of Constabulary.318
A rocky start The CPS started operation in 1986. For many years the CPS had
a poor press. In the early years there were many media stories of muddle and
confusion, of lost ¬les, delays and cases bungled. It is generally agreed that the
operational e¬ciency of the service has greatly improved. Such stories are now
much rarer, but the CPS continues to be the butt of criticism, often coming

An Independent Prosecution Service for England and Wales, Cmnd. 9074, 1983.
Runciman, p. 74, para. 26.
247 The prosecution process

from the police because of decisions to discontinue cases. There was also a
problem of sta¬ morale “ due partly to issues of management.319
In April 1997, a month before the General Election, the Labour Party pub-
lished a document in which it promised that the CPS would be re-organised yet
again “ this time into forty-two areas each with the same boundaries as the
forty-two police areas. It also announced that there would be a review of the
working of the CPS by a three-man team headed by Sir Iain Glidewell, a former
Lord Justice of Appeal.

The Glidewell Report
The Glidewell Report was published in June 1998.320 It proved to be a hard-
hitting, wide-ranging document highly critical of the CPS. The 216-page report
made seventy-¬ve recommendations for changes. The basic themes were:
• A need for improved sta¬ morale.
• Greater devolution of decision-making from the centre “ approval for the
Government™s decision to move from thirteen areas to forty-two.
• Senior lawyers to spend less time on administration and more on casework
and prosecuting.
• Headquarters should be ˜slimmer, tougher and more directly in control of
matters with which it is properly concerned™ (p. 165).
• The chief executive should be a lay person.
• Better working arrangements for the interface with the police.
• The police to retain responsibility for investigation and charging but there
should be new Criminal Justice Units (CJUs) with a CPS lawyer in charge,
civilian employees of the police and senior police o¬cers attached as liaison
to direct further investigations. CJUs would deal with ˜fast track™ cases, mag-
istrates™ court cases generally and would instruct counsel for Crown Court
• Better support for prosecuting barristers in the Crown Court.

The Government™s response to Glidewell
The Government™s reaction to the report was basically positive.321 It said it saw
the future in collaboration and partnership between the police and the CPS.

A survey in November 1993 by the First Division Association, the union which represents
three-quarters of CPS lawyers, found that their morale was at an all-time low and that the
majority had no con¬dence in the senior management. A poll of the entire membership of the
union in November 1995 found the lowest morale and highest dissatisfaction was in the CPS
section. Members complained that management was poor and that Government-imposed
e¬ciency and economy drives had put intolerable pressures on them. Disa¬ection was
especially pronounced among senior sta¬. See D. Bindman, ˜Crown Jewels™, Law Society™s
Gazette, 7 February 1996, p. 22.
Review of the Crown Prosecution Service, 1998, Cm. 3960 (˜Glidewell™). For a summary and
editorial comment see Criminal Law Review, 1998, pp. 517“20.
See statement of the Attorney General on publication, House of Commons, Hansard, 1 June
1998, vol. 313, col. 42 and Written Answers, 30 November 1998, col. 67.
248 Pre-trial criminal proceedings

The Attorney General told the House of Commons: ˜co-location, common
administration and integrated working™ would ˜streamline casework and ¬le
handling processes, remove duplication and unnecessary burdens and reduce
delay™. The police would retain their responsibility for ¬le preparation and
witness warning. The Government had decided ˜that it would not be practical
or a proper re¬‚ection of the respective constitutional priorities of the CPS and
the police to require a transfer of responsibilities™.322

The Auld Review
Lord Justice Auld in his Review said that most of the Glidewell recommenda-
tions had been adopted after local pilots. CPS sta¬ were increasingly located in
or close to police stations working in liaison with the police in CJUs and were
receiving papers for review shortly after charge. Early signs were that the new
system was producing some improvements in e¬ciency and some savings
though not, in the main, in the accuracy of charging.323
Disagreeing with both Runciman and Glidewell, Auld recommended that
charging of suspects should to a large extent be taken over by the CPS. The chief
reason was overcharging by the police. (˜A signi¬cant contributor to delays in
the entering of pleas of guilty and in identifying issues for trial and, in conse-
quence, the prolonged and disjointed nature of many criminal proceedings, is
“overcharging” by the police and failure by the CPS to remedy it at an early
stage™.324) Overcharging, Auld said, led the defence to maintain tactical pleas of
not guilty until the last minute. It could also give rise to ˜hasty, ill-considered
and inappropriate™ acceptances by the prosecution of guilty pleas which bewil-
dered and distressed victims.325
Auld thought that ˜consideration should be given to a move towards
earlier and more in¬‚uential involvement of the CPS in the process to the point
where, in all but minor, routine cases, or where there is a need for a holding
charge, it should determine the charge and initiate the prosecution™. This
became the basis for the new system established by the Criminal Justice Act

The CPS takes on the task of charging
In response to the Auld proposal, the Home Secretary and the Attorney General
decided to start a pilot scheme within the then current legal framework. This
took place on nine sites in ¬ve areas from February to August 2002. Reports on
the pilot were commissioned from independent researchers, PA Consulting

Attorney General, House of Commons, Hansard, 19 April 1999, WA col. 398.
An Early Assessment of Co-located Criminal Justice Units “ available on the CPS Website
324 325
www.cps.gov.uk. Auld, Ch. 10, p. 408, para. 35. Ibid.
See Crown Prosecution Service Charging Suspects: Early Involvement by CPS. A Pilot Final
Evaluation, April 2003.
249 The prosecution process

The pilot involved 3,324 cases in which the CPS gave the police oral advice as
to charge and 2,875 cases in which the advice was written. The reported results
were very positive:
• Conviction rates improved in six of the nine pilot areas.
• Discontinuance rates were lower than the pre-pilot sample in all areas.
• The proportion of defendants who pleaded guilty increased and the pleas
came at an earlier stage.
• There was a consistent fall in all areas in the number of cases where charges
were changed or dropped. The early intervention of the CPS, it seemed, made
it more likely that the charge was right from the start.
• The proportion of last-minute changes of plea (˜cracked trials™) decreased.
• The time from arrest to charge increased by an average of twenty-four days
but the time from charge to completion reduced by ten days.
• There was some evidence that the quality of ¬les had improved “ and not just
the ¬les in the pilot.
The Executive Summary of the Report said that all areas were ˜fully supportive
of early charging advice™ and were continuing with the scheme even though the
formal pilot had ended. It added: ˜we must also highlight the qualitative bene-
¬ts which have been seen in all areas including enhanced joint working, better
quality, skills transfer, and improved con¬dence, trust and mutual respect™
(p. vii). However, not all reactions were equally positive.327
In the meantime, the Government™s White Paper Justice for All had indicated
that Auld™s recommendation would be adopted and provisions to give it e¬ect
were included in Part 4 and Sch. 2 of the Criminal Justice Act 2003.328
Under these provisions, s. 37 of PACE is amended to require that the custody
o¬cer has regard to guidance from the DPP329 when determining whether the
suspect should be released without charge on bail, or without charge and without
bail or charged. Where the case is referred to the CPS to determine whether pro-
ceedings should be instituted and, if so, on what charges, the defendant can be
released on police bail with or without conditions. In routine minor cases the

An earlier assessment of the proposed placement of prosecutors in police stations to o¬er pre-
charge advice to police investigators had been sceptical about its likely bene¬ts “ see J.
Baldwin and A. Hunt, ˜Prosecutors Advising in Police Stations™, Criminal Law Review, 1998,
pp. 521“36. After the scheme was implemented, the Law Society™s Gazette, 4 August 2004, p. 3,
reported: ˜concerns are growing among criminal practitioners that the “charging project” is
damaging their health, scuppering the legal aid fund, and seeing o¬enders roam the streets
owing to lack of staª. The article said that both prosecution and defence lawyers were
concerned about the ˜plight of exhausted lawyers who were doing back-to-back shifts in the
police station and courts™. The Director of the Criminal Law Solicitors Association was quoted
as saying that CPS sta¬ shortages had left defence solicitors sitting in the police station for
hours with their clients waiting for a charging decision “ which was likely to increase the
burden on the legal aid fund. The police also su¬ered from the temptation to bail defendants
rather than have them sit and wait. Cm. 5563, July 2002, para. 3.31.
The Director™s Guidance document runs to sixteen pages. (See www.cps.gov.uk “ Publications
“ Prosecution Policy and Guidance.)
250 Pre-trial criminal proceedings

police continue to charge. In cases where bail would be inappropriate and where
the CPS have not yet informed the police what charges to lay, the police are per-
mitted under the DPP™s guidance to lay holding charges. In deciding whether to
charge, the police and the CPS have to follow the same basic principles.
The new charging system was brought in gradually but by April 2006 it had
been activated in all forty-two areas.
For an excellent description of the system see I.D. Brownlee, ˜The Statutory
Charging Scheme in England and Wales: Towards a Uni¬ed Prosecution
System™, Criminal Law Review, 2004, pp. 896“907.
See also R.M. White, ˜Investigators and Prosecutors or Desperately Seeking
Scotland: Re-formulation of the “Philips Principle”™, 69 Modern Law Review,
2006, pp. 143“82. White considers the respective roles of investigation and
prosecution in the CPS model as compared with that in Scotland and Northern
Ireland and that used in the Serious Fraud O¬ce, for prosecutions by the
Revenue and Customs and by the new Serious and Organised Crime Agency.

The decision to prosecute
The Code for Crown Prosecutors
The CPS decision as to whether to prosecute is based on the Code for Crown
Prosecutors. The current (2004) version is on the CPS Website (www.cps.gov.
uk) and is also published at the back of its annual report.330
The following text gives key passages as regards the tests that have to be
applied in the new charging system. Under that system, unless the new
˜Threshold Test™ applies, a prosecution can only go ahead if the case has passed
both parts of the ˜Full Code Test™.
Paragraph 2.2 sets out the basic principle of prosecution:

Code for Prosecutors, 2004
2.2 Crown Prosecutors must be fair, independent and objective. They must not
let any personal views about the ethnic or national origin, disability, sex, reli-
gious beliefs, political views or the sexual orientation of the suspect, victim or
witness in¬‚uence their decisions. They must not be a¬ected by improper or
undue pressure from any source . . .
The Full Code Test is set out in section 5. It has two stages.

5 The Full Code Test
5.1 . . . The ¬rst stage is consideration of the evidence. If the case does not pass
the evidential test, it must not go ahead, no matter how important or serious it
may be. If the case does pass the evidential test, Crown Prosecutors must
proceed to the second stage and decide if a prosecution is needed in the public
interest . . .
For an explanatory article by the DPP see K. Macdonald QC, ˜The New Code for Crown
Prosecutors™, 155 New Law Journal, 7 January 2005, p. 12.
251 The prosecution process

The Evidential Stage
5.2 Crown Prosecutors must be satis¬ed that there is enough evidence to provide
a ˜realistic prospect of conviction™ against each defendant on each charge. They
must consider what the defence case may be, and how that is likely to a¬ect the
prosecution case.
5.3 A realistic prospect of conviction is an objective test. It means that a jury
or bench of magistrates or judge hearing a case alone, properly directed in accor-
dance with the law, is more likely than not to convict the defendant of the charge
alleged . . .
5.4 When deciding whether there is enough evidence to prosecute, Crown
Prosecutors must consider whether the evidence can be used and is reliable.
There will be many cases in which the evidence does not give cause for concern,
but there will also be cases in which the evidence may not be as strong as it ¬rst
appears . . .
The Code refers to the question whether the evidence might be excluded because
of the way it was gathered and whether the evidence is su¬ciently reliable.

The Public Interest Stage
5.6 In 1951, Lord Shawcross, who was Attorney General, made the classic state-
ment on public interest, which has been supported by Attorneys General ever
since: ˜It has never been the rule in this country, I hope it never will be, that sus-
pected criminal o¬ences must automatically be the subject of prosecution™.
(House of Commons Debates, vol. 483, col. 681, 29 January 1951.)
5.7 The public interest must be considered in each case where there is enough
evidence to provide a realistic prospect of conviction. Although there may be
public interest factors against prosecution in a particular case, often the prose-
cution should go ahead and those factors should be put to the court for consid-
eration when sentence is being passed. A prosecution will usually take place
unless there are public interest factors tending against prosecution which clearly
outweigh those tending in favour, or it appears more appropriate in all the cir-
cumstances of the case to divert the person from prosecution.
Factors stated to militate in favour of prosecution (para. 5.9) include: (1) the
likelihood of a signi¬cant sentence; (2) a weapon was used or violence was
threatened; (3) the o¬ence was committed against someone who serves the
public such as a police or prison o¬cer or a nurse; (4) the accused was in a posi-
tion of authority or trust; (5) the accused was the ringleader or an organiser; (6)
the o¬ence was premeditated; (7) that it was carried out by a group; (8) the
victim was vulnerable or was put in fear or su¬ered personal attack, damage or
disturbance; (9) the o¬ence involved discrimination on grounds of ethnic or
national origin, sex, religion, political belief or sexual orientation; (10) a
marked di¬erence between the actual or mental ages of the accused and the
victim, or if there is an element of corruption; (11) the relevance of the
accused™s record; (12) whether the accused was subject to a court order; (13) the
likelihood of repetition and (14) the o¬ence, though not serious in itself, is
widespread in that area.
252 Pre-trial criminal proceedings

Factors stated to militate against prosecution (para. 5.10) include: (1) the
likely penalty would be very small or nominal; (2) the o¬ence was committed
as a result of a mistake or misunderstanding; (3) the loss or harm is minor; (4)
long delay between the o¬ence and the trial, unless the o¬ence is serious or it
has only just come to light; (5) prosecution will have a very bad e¬ect on the
accused™s physical or mental health, always bearing in mind the seriousness of
the o¬ence; (6) the accused is old, su¬ering from signi¬cant mental or physical
ill-health, unless the o¬ence is serious or there is a real possibility of it being
repeated; (7) the accused has already made reparation or paid compensation
(˜but defendants must not avoid prosecution or diversion solely because they
pay compensation™) and (8) details may emerge at the trial which could harm
sources, international relations or national security.
The CPS prosecutes on behalf of the public at large and not in the interests
of any individual. ˜However, when considering the public interest, Crown
Prosecutors should always take into account the consequences for the victim
of the decisions whether or not to prosecute, and any views expressed by the
victim or the victim™s family™ (para. 5.12). It was important that a victim is told
about a decision which ˜makes a signi¬cant di¬erence to the case™ (para. 5.13).

The Threshold Test
The Threshhold Test was introduced in 2004. It is applied to cases in which it is
not appropriate for the suspect on bail but the evidence to apply the Full Code
Test is not yet available and the PACE detention time limit requires that he either
be charged or released from custody.
6.1 The Threshold Test requires Crown Prosecutors to decide whether there
is at least a reasonable suspicion that the suspect has committed an o¬ence, and
if there is, whether it is in the public interest to charge that suspect.
6.4 The evidential decision in each case will require consideration of a
number of factors including:
the evidence available at the time;
the likelihood and nature of further evidence being obtained;
the reasonableness for believing that evidence will become available;
the time it will take to gather that evidence and the steps being taken to do so;
the impact the expected evidence will have on the case;
the charges that the evidence will support.
6.5 The Public Interest Test means the same as under the Full Code Test, but
will be based on the information available at the time of charge which will often
be limited.
When the police charge a suspect they are supposed to apply the same Evidential
Test, Threshold Test and Public Interest Test as the CPS.331
For analysis of the ˜public interest element™ under the previous version of the Code see A.
Ashworth, ˜The “Public Interest” Element in Prosecutions™, Criminal Law Review, 1987,
p. 595. See also A. Ashworth and J. Fionda, ˜Prosecution, Accountability and the Public
Interest™, Criminal Law Review, 1994, p. 894 and R. Daw, ˜A Response™, ibid, p. 904; A. Hoyano
et al, ˜A Study of the Impact of the Revised Code for Crown Prosecutors™, Criminal Law
Review, 1997, pp. 556“64.
253 The prosecution process

For a strongly argued view that the evidential requirement in the Code is
misconceived, see G. Williams, ˜Letting o¬ the Guilty and Prosecuting the
Innocent™, Criminal Law Review, 1985, p. 115. Professor Williams™ contention
was that if the test for proceeding is whether a conviction was likely to succeed
(the ˜51 per cent or realistic prospect of conviction rule™) many prosecutions
that ought to be brought would not be. The test rather should be whether the
prosecutor is satis¬ed on the evidence that the suspect is guilty, subject to the
public interest questions as to whether a prosecution is desirable. Certainly
there had to be at the least a reasonable possibility of a conviction, but the
e¬ect of the rule that there must be a reasonable probability of a conviction
meant, for instance, that corrupt police o¬cers might not be prosecuted
because it was notoriously di¬cult to get a jury to convict a police o¬cer.
Where the prosecutor did not believe that the accused was guilty, he should
drop the case. An exception to this principle might be where failure to charge
someone may bring about a loss of public con¬dence in the integrity of the
prosecution service.
It is not only the Code for Crown Prosecutors that in¬‚uences the decision
whether or not to start and continue a prosecution. Another consideration is
the CPS sta¬ member™s concern to maintain his employer™s approval by not
having too many cases that go wrong “ i.e. end in acquittal. In an era when per-
formance targets dominate thinking, that will tend to militate in the direction
of dropping cases when there is a doubt as to the prospects of a conviction. In
former times the prosecution might have left the issue to be resolved by the jury.
The same tendency is promoted by the current concern to reduce costs which
powerfully a¬ects the members of all public agencies.
In addition to the published Code, CPS sta¬ also work under the in¬‚uence of
the unpublished Policy Manual which is for internal use only.

Guide to case disposal
A step in the direction of reducing both discretion and prosecutions came in 1995
with the issue to the police of a new Case Disposal Manual. This ranked every
o¬ence, motoring, criminal and alcohol-related, on a scale of points from one to
Five-point o¬ences, such as murder, would always be prosecuted.
Four-point o¬ences have what the Manual terms ˜a high probability of prose-
cution™. This category included GBH, forgery, arson, perjury, burglary and per-
verting the course of justice.
Three-point o¬ences included indecent assault, theft, handling stolen goods,
buggery, prostitution o¬ences, resisting arrest, criminal damage and ABH. In
three-point o¬ences the decision whether to charge was to be made by listing
the ˜aggravating™ and the ˜mitigating™ factors. The Manual listed general factors,
for instance, the impact on the victim, the accused™s prior criminal record, the
likelihood of penalty and whether the crime is a prevalent o¬ence causing local
concern. The Manual also indicated factors speci¬c to particular o¬ences. So
254 Pre-trial criminal proceedings

possession of drugs (even Class A substances such as cocaine, crack and heroin)
would usually be cautioned if only ˜small amounts, for personal use™ are
involved. The possessor of an o¬ensive weapon would not be prosecuted under
the Manual™s guidelines if there was ˜no risk, weapon not on display, mistaken
belief that there would be no o¬ence if carried for protection only™. Deception
o¬ences would not be prosecuted if they were ˜committed over a short period,
low value™ or ˜driven by poverty/personal need™. ABH would not be prosecuted
if it was a single blow causing only super¬cial injury.
Two-point o¬ences were those where there was a high probability of a caution
and the decision-maker needed to be able to justify the decision not to caution.
This category includes begging, kerb-crawling and being drunk and incapable
in a public place. A charge for this o¬ence would only be laid if the o¬ender was
arrested four times for the same o¬ence within a four-week period.
One-point o¬ences were minor o¬ences for which a formal warning was
appropriate or where there was a decision not to proceed with a prosecution.
They included such things as throwing litter in the street or sending someone
to buy liquor for an underage person.
D. Rose, Home A¬airs Correspondent of The Observer, commented on this
remarkable development in his book In the Name of the Law: The Collapse of
Criminal Justice (Jonathan Cape, 1996) pp. 163“4:
The Case Disposal Manual, introduced at ¬rst in London and several counties,
with others rapidly following suit, enlarges police discretion on an unprece-
dented scale. It requires that o¬cers of junior rank take fundamental deci-
sions with massive implications for the lives of those they arrest, without
reference to any court or outside authority. The judgments it demands are
even more subjective than some of those required by the Code for Crown
Prosecutors. How, for example, do you measure whether a sexual assault is
˜trivial™? It is inescapable that many of those judgments will be shaped by
factors which have nothing to do with the true merits of the case: the o¬cer™s
workload; his opinion of the suspect; and the possibility that he may, in return
for non-prosecution, become a useful informant in future . . . The Manual,
drawn up after consultation between the Metropolitan Police, the Association
of Chief Police O¬cers and the CPS, alters institutionalised police practice
signi¬cantly, but its introduction took place without any trace of public or
parliamentary debate.

Reducing discretion by new ˜charging standards™
In August 1994 the police and the CPS introduced a new piece of machinery, ˜charg-
ing standards™. The ¬rst such standards were for assault.332 They were later extended
to public order o¬ences, driving o¬ences, dishonesty o¬ences and drugs.

Justice of the Peace, 20 August 1994, pp. 554“5. The standards de¬ned the degree of injury that
would justify charges respectively of common assault, assault occasioning actual bodily harm
(˜ABH™), unlawful wounding/in¬‚icting grievous bodily harm (˜GBH™) and wounding/causing
GBH with intent.
255 The prosecution process

The motive for this innovation was stated to be to increase consistency of
decision-making throughout the country. It may be that there were also other
reasons behind the decision. One such could have been to minimise the occa-
sions when the CPS incurred the annoyance of the police by reducing charges.
Another was to reduce costs. The concern expressed in some quarters was that
the overall e¬ect would be to downgrade o¬ences.333
The Glidewell Report (1998, p. 247 above) said: ˜We are very much in favour
of charging standards as a useful guide to both members of the police and the
CPS™ (p. 83). On the question of downgrading of o¬ences, Glidewell said that
although it suspected that charges were sometimes downgraded when they
should not be, it had no evidence on the question. It noted that charges were
only rarely upgraded. Downgrading was usually where the defendant was
pleading guilty (pp. 84“5).

Cautioning as an alternative to prosecution
One way to avoid prosecuting the suspect is to administer a caution. This has
been a part of the system for decades and has been used especially for young
o¬enders. From 1978 cautioning of young o¬enders was guided by Home O¬ce
circulars which established three conditions for a caution: there had to be
su¬cient evidence to justify a prosecution, the o¬ender had to admit the o¬ence
and the parent or guardian had to consent to the giving of a caution. A caution
was not a conviction but on any subsequent court appearance it could be cited.334
The Home O¬ce guidelines on cautioning were changed in 1990, largely
because of the research showing wide variations in cautioning rates between
forces. They were also partly the result of a drop in cautions of persons over sev-
enteen. The presumption in favour of not prosecuting juveniles and the elderly
was also to be extended ˜to other groups “ young adults and adults alike “ where
the criteria for caution are met™.
Mr Michael Howard, who became Conservative Home Secretary in 1993,
took a signi¬cantly less positive view of cautioning at least for more serious
o¬ences. His tougher approach was re¬‚ected in a new Circular (18/1994) issued
in March 1994. Its purpose was to provide guidance on cautioning and in par-
ticular ˜to discourage the use of cautions in inappropriate cases,335 for example
for o¬ences which are triable on indictment only; to seek greater consistency
between police forces; and to promote the better recording of cautions™ (para. 1).

See for instance F.G. Davies, ˜CPS Charging Standards: A Cynic™s View™, Justice of the Peace, 1
April 1995, p. 203.
See Judge Richard May, ˜The Legal E¬ect of a Police Caution™, Criminal Law Review, 1997,
pp. 491“3 and R. Evans, ˜Challenging a Police Caution using Judicial Review™, Criminal Law
Review, 1996, pp. 104“8. For a warning about the unreliability of cautions and therefore the
danger of treating them like previous convictions by citing them in subsequent proceedings
see Criminal Law Review, 1996, p. 453.
The circular said that, despite earlier discouragement of cautions for the most serious
o¬ences, cautions had been administered for o¬ences as serious as attempted murder and
rape which ˜undermines the credibility of the disposal™ (para. 5).
256 Pre-trial criminal proceedings

Cautions for juveniles replaced by reprimands and warnings The Crime and
Disorder Act 1998 (CDA 1998), ss. 65 and 66336 provided for the replacement of
cautions for juveniles by ˜reprimands and warnings™. (The term ˜caution™ is
retained for adult o¬enders.) A reprimand is a ¬rst caution. Normally there is
only one reprimand because s. 65(3) provides that if the o¬ender has previously
been reprimanded he cannot be reprimanded a second time unless the o¬ence
was committed two years after the last warning “ and no one can be warned
more than twice.
The ˜warning™ (˜¬nal warning™) is similar to what has come to be called
˜caution plus™ schemes which provide some kind of counselling, mentoring or
other community support for the young person. The o¬ender is referred to a
Youth O¬ending Team (YOT)337 which determines whether a rehabilitation
programme is appropriate. This may involve some form of mediation. The
Home O¬ce has indicated support for the restorative-cautioning initiative pio-
neered in the Thames Valley area.
Like the old caution, a reprimand or warning is not a conviction338 and does
not constitute a criminal record but the fact of the reprimand or warning and
any report on failure to participate in a rehabilitation programme can be cited
in court in the same way as a conviction.339 A person reprimanded or warned
for certain sex o¬ences is required to register with the police under the Sex
O¬ences Act 1997.340 Fingerprints are taken.341 Reprimands and warnings are
entered on the Police National Computer (PNC).342
However, unlike the old caution, a reprimand or warning does not require
the consent of the young o¬ender or of his parent or guardian. In R (R) v.
Durham Constabulary343 the Divisional Court held that despite the decision not
to prosecute the ¬nal warning process was incompatible with the person™s
ECHR Article 6 right to a fair and public hearing in the determination of a crim-
inal charge. (Entry on the PNC and the sex o¬ender™s register amounted to a
public pronouncement of guilt.) The House of Lords344 unanimously reversed
the decision on the ground that neither a reprimand nor a warning were the
determination of a criminal charge. Nor was their recording on the PNC and
the sex o¬ender™s register as access to them was controlled and limited to a small
number of authorised persons.
A ¬rst o¬ence by a young o¬ender can therefore be met with a reprimand,
a ¬nal warning or criminal charges depending on its seriousness. After a

As amended by the Criminal Justice and Court Services Act 2000, s. 56 and the Powers of
Criminal Courts (Sentencing) Act 2000, Sch. 9, para. 198.
Under CDA 1998, s. 39.
However, the assurance given to the person that the caution is instead of prosecution
precludes the bringing of a private prosecution by the victim “ Jones v. Whalley [2006] UKHL
41, [2007] Crim LR 74.
339 340
Home O¬ce, Final Warning Scheme “ Guidance for Police, April 2000. Ibid, para. 77.
341 342
Ibid, para. 80. Ibid, paras. 82“3.
[2002] EWHC 2486 Admin, [2003] 3 All ER 419.
[2005] UKHL 21, [2005] 2 All ER 369. For comment see Criminal Law Review, 2006, p. 88.
257 The prosecution process

reprimand, a further o¬ence leads either to a warning or charge. Normally it
would be prosecution.
The CPS Code states:
Reprimands and ¬nal warnings are intended to prevent re-o¬ending and the
fact that a further o¬ence has occurred indicates that attempts to divert the
youth from the court system have not been e¬ective. So the public interest will
usually require a prosecution in such cases, unless there are clear public interest
factors against prosecution (para. 8.9).
For an early and highly critical assessment see R. Evans and K. Puech, ˜Repri-
mands and Warnings: Populist Punitiveness or Restorative Justice?™ Criminal
Law Review, 2001, pp. 794“805. Their conclusion was that rather than provid-
ing an opportunity for new style restorative justice, ˜the legislation is punitive
and controlling in principle and in practice™ (p. 804). Many of the young people
and YOT workers saw the warning scheme as ˜arbitrary, unfair, and dispropor-
tionate especially as it may involve compulsory participation in a rehabilitation
(change) programme™ (ibid). It suggested that that there was a considerable gap
between the rhetoric of the Home O¬ce and the Youth Justice Board and what
was actually happening on the ground.

Simple cautions for adults The Home O¬ce updated its guidance on ˜simple™
cautions in Circular 30/2005 (Cautioning of adult o¬enders). The police can
issue a simple caution in all cases except indictable-only o¬ences which must be
referred to the CPS.
The precondition for a simple caution is that there is enough evidence to
satisfy the Threshold Test (above),345 that the suspect has admitted the o¬ence
and that, given the seriousness of the o¬ence, it is in the public interest to use a
simple caution to dispose of the case.
As for juveniles, where previously a caution required the consent of the
person, this requirement has been dropped.
In April 2006 it was reported that the Home O¬ce had sent the police a doc-
ument, the Gravity Factor Matrix, as part of a strategy to widen use of cau-
tions. Providing there were no aggravating factors, ¬rst-time o¬enders could
receive a caution for nearly one-third of the 180 crimes listed in the document.
The crimes were in four categories from the least serious (Level 1) to the most
serious (Level 4). The sanction in Level 2 covering some sixty o¬ences was
˜normally a simple caution for a ¬rst o¬ence™. Level 2 included criminal
damage up to £500, theft up to £200 and sex with thirteen- to ¬fteen-year-
In July 2006 the Lord Chancellor, the Home Secretary and the Attorney
General jointly issued a forty-two page policy document entitled Delivering

The evidentiary test previously required the higher standard of a realistic prospect of
conviction. Sunday Times, 2 April 2006.
258 Pre-trial criminal proceedings

Simple, Speedy, Summary Justice. It stated that a key component of a simple,
speedy, summary criminal justice system was the ability to deal rapidly and
e¬ectively with cases where formal court proceedings were disproportionate
and remedies such as ¬xed penalty notices, warnings or cautions were more

Conditional cautions The Criminal Justice Act 2003 introduced a new concept
of conditional cautions for adults “ a caution with conditions attached “ which
in e¬ect is a form of diversion from the criminal justice system. The conditions
must have one or more of the objectives of reparation for the o¬ence or reha-
bilitation of the o¬ender, or his punishment (s. 22 as amended by the Police and
Justice Act 2006). The 2006 Act (s. 17) added as possible conditions the payment
of a penalty of up to £250 or up to twenty hours of community work. The
requirements are that the CPS consider that there is su¬cient evidence to
charge the individual and that the individual signs a document in which it is
stated that he admits that he committed the o¬ence, that he consents to the cau-
tions and the conditions imposed (s. 23). Failure without reasonable excuse to
comply with the conditions makes the person liable to prosecution for the
o¬ence (s. 24). The 2006 Act, s. 18 gives the police a right of arrest for breach of
the conditions.348
The Delivering Simple, Speedy, Summary Justice policy paper of July 2006
(above) said that the aspiration was to implement conditional cautioning across
the whole of England and Wales by April 2008.349
The Government™s Fraud Review thought that in exceptional circumstances
conditional cautions might be appropriate even in fraud cases “ for instance if
the condition was to compensate victims of the fraud.350

Discontinuance by the CPS
One of the stated objectives for setting up the CPS was better and earlier iden-
ti¬cation of cases that for any reason should not go forward to prosecution.
The 1983 White Paper An Independent Prosecution Service for England and
Wales said the objectives of the CPS included the promotion of greater consis-
tency of policy and uniformly high standards of case preparation and decision-
making across the country. The e¬ect, it was said, ˜should be that cases which
are unlikely to succeed should be weeded out at an early stage™ (Cmnd. 9074,
p. 14), but the CPS™ power to drop a prosecution has given rise to much

Paragraph 7.6, p. 40.
See I. Brownlee, ˜Conditional Cautions and Fair Trial Rights in England and Wales: Form
versus Substance in the Diversionary Agenda?™, Criminal Law Review, 2007, pp.129“40.
Paragraph 7.12, p. 41.
Fraud Review, pp. 272“3, para. 11.72.
259 The prosecution process

The Code for Prosecutors (paras. 4.1“4.3) states:
4.1 Each case the Crown Prosecution Service receives from the police is
reviewed to make sure that it is right to proceed with a prosecution . . .
4.2 Review is a continuing process and Crown Prosecutors must take account
of any change in circumstances. Wherever possible, they should talk to the
police ¬rst if they are thinking about changing the charges or stopping the case.
Crown Prosecutors should also tell the police if they believe that some addi-
tional evidence may strengthen the case. This gives the police the chance to
provide more information that may a¬ect the decision.
4.3 The Crown Prosecution Service and the police work closely together, but
the ¬nal responsibility for the decision whether or not a charge or a case should
go ahead rests with the Crown Prosecution Service.
The CPS discontinue around 12“13 per cent of the cases it starts. It is di¬cult
to form a view as to whether this is too high, too low or about right. The police
often complain that the CPS drop too many cases, but there is also evidence that
it drops too few. (The Crown Court Study done for the Runciman Royal
Commission showed that in the view of both prosecuting and defence barris-
ters and of judges, the prosecution was weak in about one-¬fth of contested
cases and that over 80 per cent of these cases ended in acquittal.351)
Another possible indication comes from acquittal statistics. There are three
kinds of acquittals “ ordered acquittals, directed acquittals and jury acquittals.
˜Ordered acquittals™ are where the prosecution o¬ers no evidence at all. The case
is dropped at court because the prosecution decide at the last moment not to
pursue it, perhaps because a crucial witness fails to turn up or refuses to give evi-
dence. (The rules do not permit the CPS to discontinue a case between com-
mittal for trial by the magistrates and trial at the Crown Court. They must
therefore go through the process of formally o¬ering no evidence even in cases
which their process of review has identi¬ed as too weak to continue.) ˜Directed
acquittals™ are where the judge stops the case, usually half way, after a submission
by the defence that the prosecution™s case is not strong enough even to require a
response. Jury acquittals are where the jury has deliberated and found the defen-
dant not guilty. At ¬rst blush, ordered and directed acquittals seem to be in the
category of weak cases that arguably could and should have been aborted earlier.
In the six years leading to the establishment of the CPS the proportion of
acquittals that were either ordered or directed acquittals ranged from a low of
43 per cent in 1982 to a high of 48 per cent in 1985. It might have been expected
that with the establishment of the CPS and its presumably better screening
methods, the proportion of ordered and directed acquittals would go down. In
fact, however, it went up. It is now some two-thirds of all acquittals. (In the ¬ve
years from 2001“5 it ¬‚uctuated between 62 per cent and 69 per cent.) (See the
annual Judicial Statistics, Table 6.10).)

M. Zander and P. Henderson, Crown Court Study (Royal Commission on Criminal Justice,
Research Study No. 19, 1993) Tables 6.20 and 6.21.
260 Pre-trial criminal proceedings

There have been a number of studies as to the reasons why cases are termi-
nated by the CPS or by the court352 and various initiatives have been tried, but
it is not clear that much impact has yet been made on the issue.

Judicial control of police discretion in prosecution policy
The problem of controlling police discretion with regard to prosecuting has
only rarely come before the courts. The ¬rst modern examples of importance
were the cases brought by a private citizen, former Member of Parliament Mr
Raymond Blackburn, to compel the police to enforce the gambling and then the
obscenity laws.

R v. Metropolitan Police Commissioner, ex p Blackburn [1968] 2 WLR
893, Court of Appeal
[In April 1966, a con¬dential instruction was issued to senior o¬cers of the
Metropolitan Police. Underlying this instruction was a policy decision not to
take proceedings against clubs for breach of the gaming laws unless there were
complaints of cheating or they had become the haunts of criminals. The appli-
cant, being concerned at gaming in London clubs, brought proceedings for
mandamus to get the Commissioner to withdraw the con¬dential instruction.]
Lord Denning MR: the result of the police decision of 22 April 1966, was that
thenceforward, in this great metropolis, the big gaming clubs were allowed to
carry on without any interference by the police . . .

The duty of the Commissioner of Police of the Metropolis
I hold it to be the duty of the Commissioner of Police of the Metropolis, as it is
of every chief constable, to enforce the law of the land. He must take steps so to
post his men that crimes may be detected; and that honest citizens may go about
their a¬airs in peace. He must decide whether or not suspected persons are to
be prosecuted; and, if need be, bring the prosecution or see that it is brought,
but in all these things he is not the servant of anyone, save the law itself. No min-
ister of the Crown can tell him that he must, or must not, prosecute this man or
that one. Nor can any police authority tell him so. The responsibility for law
enforcement lies on him. He is answerable to the law and to the law alone. That
appears su¬ciently from Fisher v. Oldham Corpn,353 and A-G for New South
Wales v. Perpetual Trustee Co Ltd.354
Although the chief o¬cers of police are answerable to the law, there are many
¬elds in which they have a discretion with which the law will not interfere. For
instance, it is for the Commissioner of Police of the Metropolis, or the chief con-
stable, as the case may be, to decide in any particular case whether inquiries

See D. Crisp and D. Moxon, Case Screening by the Crown Prosecution Service: How and Why
Cases are Terminated (Home O¬ce Research Study No. 137, 1994); B. Block, C. Corbett and
J. Peay, Ordered and Directed Acquittals in the Crown Court (Royal Commission on Criminal
Justice Research Study No. 15, 1993); the study reported in the CPS Annual Report, 1994“5,
pp. 7“8 and ¬gures cited in the Glidewell Report, 1999 at p. 90.
353 354
[1930] 2 KB 364, 46 TLR 390. [1955] AC 457.
261 The prosecution process

should be pursued, or whether an arrest should be made or a prosecution
brought. It must be for him to decide on the disposition of his force and the con-
centration of his resources on any particular crime or area. No court can or
should give him directions on such a matter. He can also make policy decisions
and give e¬ect to them, as, for instance, was often done when prosecutions were
not brought for attempted suicide, but there are some policy decisions with
which, I think, the courts in a case can, if necessary, interfere. Suppose a chief
constable were to issue a directive to his men that no person should be prose-
cuted for stealing any goods less than £100 in value. I should have thought that
the court could countermand it. He would be failing in his duty to enforce the
. . . On 30 December 1967, the Commissioner issued a statement in which he
said: ˜it is the intention of the Metropolitan Police to enforce the law as it has
been interpreted™. That implicitly revoked the policy decision of 22 April 1966;
and the Commissioner by his counsel gave an undertaking to the court that the
policy decision would be o¬cially revoked. We were also told that immediate
steps are being taken to consider the ˜goings-on™ in the big London clubs with a
view to prosecution if there is anything unlawful. That is all that Mr Blackburn
or anyone else can reasonably expect.
See also R v Metropolitan Police Commissioner, ex p Blackburn (No 3)355 and R v
Chief Constable of Devon and Cornwall, ex p Central Electricity Generating
But there are limits to police and prosecutorial discretion. In R (Mondelly) v
Metroplitan Police Commissioner 357 the Divisional Court considered the impli-
cations of the downgrading by legislation of cannabis from Class B to Class C
and consequential national guidance from the police and the Home O¬ce that
a person found in simple possession of cannabis should not be arrested unless
there were aggravating factors. The applicant sought judicial review to quash the
decision to caution him for simple possession after the police had, by mistake,
gone to his premises and were invited in, at which point they noticed a smell of
cannabis. He was arrested for allowing his premises to be used for the smoking
of cannabis but this charge was not pursued. Instead he was cautioned for simple
possession. He argued that in view of the national guidance this was unlawful.
The court, in a 2“1 decision (Moses LJ and Ouseley J) refused the application.
M™s argument was based on the proposition that the national guidance issued to

[1973] 1 All ER 324. Mandamus to enforce the pornography laws refused on the court being
persuaded that the police were doing their best. (Subsequent prosecutions of members of the
Obscene Publications Squad for corruption suggested that this was not quite the case.)
[1981] 3 All ER 826. Mandamus refused to require the police to help the Electricity Board to
remove squatters blocking the building of a nuclear power station. It was for the police on the
spot and not the Court to decide when and how to exercise their power.
See further Coxhead [1986] Crim LR 251, CA; R v General Council of the Bar, ex p Percival
[1990] 3 All ER 137, 149“52, Div Ct: In R v Chief Constable of Kent County Constabulary,
ex p L and R v DPP, ex p B (a minor) [1993] 1 All ER 756 Div Ct; R v IRC, ex p Mead [1993]
1 All ER 772; R v Chief Constable of Sussex, ex p International Trader™s Ferry [1999] 1 All ER
129, HL. [2006] EWHC 2370 (Admin).
262 Pre-trial criminal proceedings

the police had changed the law. That could not be so. If there was a police or CPS
policy that an arrest or prosecution for the o¬ence required aggravating factors
it would be unlawful. Executive discretion could not change the law. The police
retained their discretion as to whether to arrest and to caution.358

Remedies for the prosecution™s failure to prosecute
In R v. DPP, ex p C359 the Divisional Court, most unusually, allowed an applica-
tion for judicial review of the decision of the CPS not to prosecute for buggery
of a wife by a husband on the ground that the prosecutor had not had in mind
certain relevant considerations. Several of the cases have involved ethnic minor-
ity complaints about the failure of the CPS to prosecute police o¬cers involved
in deaths in police custody of family members.360
For the argument that the courts should be prepared to review prosecutorial
decisions see generally Y. Dotan, ˜Should Prosecutorial Discretion Enjoy Special
Treatment in Judicial Review? A Comparative Analysis of the Law in England
and Israel™, Public Law, 1997, pp. 513“31.
In 1988, the House of Lords ruled that the police could not be held liable in
negligence for failing to prevent crimes. There was no duty of care to individ-
ual members of the public to identify and apprehend an unknown criminal,
even though it was reasonably foreseeable that harm was likely to be caused to
a member of the public if the criminal was not detected and apprehended. Even
if such a duty did exist, it would be against the public interest to hold the police
liable.361 In 2005, the House of Lords applied much the same reasoning in
rejecting the negligence claim brought against the police by Duwayne Brooks
who survived an attack by a gang of white thugs on him and Stephen Lawrence.
Brooks claimed that the psychiatric injury su¬ered in the attack had been exac-
erbated by the failings of the police investigation as identi¬ed in the report of
the inquiry into the Stephen Lawrence case chaired by Sir William Macpherson.
The House of Lords held that to impose such a duty of care on the police would
cause a diversion of resources from crime investigation to defending claims,
would inhibit o¬cers™ fearless discharge of their duties as they would tend to act
in a detrimentally defensive frame of mind and would involve the courts in
making policy judgments better suited to the police.362
Where the police are on notice that an attack on a named individual is fore-
seeable there may be a duty on the police to provide appropriate protection “ a
duty that could be the basis of an action for damages. This follows from the

The dissenting judge, Walker J, considered that there was no su¬cient reason for the police to
have departed from their policy as explained in the guidance and that it was right for the
court to intervene to stop an abuse of process. [1995] 1 Cr App Rep 136.
See R v. DPP, ex p Manning [2001] QB 330 in which the Divisional Court quashed the
decision not to prosecute after a death in custody and M. Burton, ˜Reviewing Crown
Prosecution Service decisions not to prosecute™, Criminal Law Review, 2001, pp. 374“84.
Hill v. Chief Constable of West Yorkshire [1989] 1 AC 53, [1988] 2 All ER 238, a case brought by
relatives of one of the victims of Peter Sutcli¬e, the ˜Yorkshire Ripper™.
Brooks v. Metropolitan Police Commissioner [2005] UKHL 24, [2005] 2 All ER 489.
263 The prosecution process

decision of the European Court of Human Rights in Osman v. United
Kingdom.363 A teacher had formed an attachment to a ¬fteen-year-old pupil. He
changed his name to that of the boy, broke windows at the family home and
slashed car tyres. The school met the police to discuss the matter. Eventually the
teacher shot and killed the boy™s father and seriously injured the boy. The Court
of Appeal rejected an action for negligence against the police in light of the deci-
sion in Hill. The ECHR held unanimously that there had been a breach of
Article 6 of the Convention which guarantees a right to have one™s civil rights
and obligations determined by a court or tribunal. The ECHR accepted that in
Hill there were su¬cient public policy reasons for excluding liability, but in
Osman the proximity test seemed to be satis¬ed as the police appeared to have
assumed some responsibility for the Osmans™ safety. A blanket immunity for the
police established by the House of Lords decision in Hill was therefore a dis-
proportionate restriction on the applicant™s right of access to a court.

Normally no duty to give reasons for not prosecuting
Normally the police cannot be required to give reasons for not prosecuting,364
but very occasionally the circumstances might be such as to require reasons to
be given. In R v. DPP, ex p Manning and Melbourne 365 Lord Chief Justice
Bingham held that where there had been a death in custody, the inquest resulted
in a verdict of unlawful killing and the identity of the person responsible and
his whereabouts were known, it was to be expected that a prosecution would
follow. If none did follow, it was appropriate to require the DPP to give his

Is the CPS independent of the police?
A central part of the case for the establishment of the CPS was that it should be
more independent of the police. In one sense the CPS certainly is independent
of the police in that the CPS has always taken the decision whether to proceed
with the case and under the new statutory charging arrangements established
by the Criminal Justice Act 2003 the CPS generally now determines the charge
from the very outset, but the police still play a crucial role which to an extent
constrains the CPS.
The entire investigation of the o¬ence is in the hands of the police. The CPS
gets the ¬le prepared by the police and nothing much else.366

363 364
(2000) 29 EHRR 245. R v. DPP, ex p Treadaway (1997) Times, 31 October.
[2000] 3 WLR 463.
In R (on the application of Joseph) v. DPP [2001] Crim LR 489 the Divisional Court held that
the CPS had been entitled to rely on summaries of video evidence given by the police which
led it to discontinue the case despite the fact that in interview one of the alleged culprits
admitted carrying a weapon which undermined the self-defence argument on which the CPS
decision was based. The commentator in the Criminal Law Review wrote: ˜this is simply
another example of the dependence of the CPS on the accuracy and bona ¬des of the police in
preparing and disclosing material™ (p. 490).
264 Pre-trial criminal proceedings

The Philips Royal Commission envisaged that the CPS would supervise and
check the work of the police but in practice this does not happen. The
Runciman Royal Commission did not recommend any change in that regard.
One reason is simply the lack of manpower resources. Another is that it would
generate tensions between the two agencies. Thirdly, since the CPS has no inves-
tigatory powers under the Act, it lacks the standing to do so. The CPS basically
work with what is provided to them by the police “ and in cases where the police
decide to take no further action the CPS will not be involved at all.
In Scotland where the prosecutors do have investigatory powers, they tend
not to use them. Research in the early 1980s showed that normally they acted
on the basis of information supplied by the police. They used the power to ask
for further information in only 6 per cent of cases. In most cases the decision-
making was largely routine. (In 63 per cent of cases the decision was taken on
the same day that the procurator ¬scal received the papers.) The real discretion
of the ¬scal at that date came in the ˜trial avoidance arrangements™ “ bargaining
over a guilty plea to lesser charges in return for other more serious charges being
dropped.367 A decade later the same researchers found a dramatic increase in the
rate of cases not prosecuted from 8“47 per cent resulting from a wider range of
formal alternatives to prosecution, notably ¬scal penalties and ¬scal ¬nes.368
At the time when the CPS was established it was regarded as vital that the new
organisation be moved physically into its own buildings away from the police,
but, as has been seen, the trend is now in the opposite direction toward co-
location. Glidewell welcomed closer co-operation and did not think it threat-
ened the independence of the prosecutor. (˜When they are working in close
proximity to a police station and in association with one or more police o¬cers,
they will, we are con¬dent, continue to maintain that professional indepen-
dence™ (p. 132).)

The three stages of the CPS™s history
In January 2003, Lord Goldsmith QC, the Attorney General, summed up what
he described as the ¬rst two stages of the CPS™s history:369
1986“1999 In this ¬rst stage the professional culture of the Crown Prosecution
Service was established. It is marked, rightly, by a great emphasis on establish-
ing the independence of the CPS from the police. It was essential to move from
what had essentially been a solicitor/client relationship with the police to estab-
lish instead a culture of independence, and to bring home that a new organisa-
tion had come into being.
These were di¬cult times. The idea was good in concept, but the execution
was poor. The CPS was undoubtedly under-resourced. The organisation itself

See S. Moody and J. Tombs, Prosecution in the Public Interest (Scottish Academic Press, 1982).
J. Tombs and S. Moody, ˜Alternatives to Prosecution: The Public Interest Rede¬ned™, Criminal
Law Review, 1993, pp. 357“67.
Speech to the Crown Prosecution Service Senior Management Conference.
265 The prosecution process

struggled to ¬nd the right balance between local autonomy and central direc-
tion. It probably became over-bureaucratic. There was a long period of inade-
quate work, loss of public con¬dence, and a lack of self-worth, but despite all
these pressures, a strong professional culture did develop . . .
1999“2003 The second phase began in early 1999 after the present
Government in its ¬rst term set about reform. It commissioned the Glidewell
Report, and put its recommendations into e¬ect. The CPS was restructured into
forty-two areas which matched police force boundaries . . .
The process was started of creating Glidewell co-located units where much
police and CPS work could be done under one roof, increasing e¬ciency and
reducing bureaucracy.
In this phase too, this Government addressed the historic under-funding of
the CPS. This chronic under-funding, which meant far too few Crown
Prosecution sta¬ having to do far too much work, has now been remedied by
this Government. In 2001, the CPS received a net increase in funds of 23 per
cent. This year it received a further 6 per cent increase in real terms and the
¬gure for the forthcoming ¬nancial year is 9 per cent in real terms. These are
substantial ¬gures. A nearly 40 per cent increase in resources is a substantial
uplift. It takes time for extra resources to translate into additional sta¬
recruited, trained and in post, especially in organisations where a very high
degree of professional expertise is required. So we are only now starting to see
the bene¬ts of these extra resources in terms of additional, quali¬ed sta¬ on
the ground, delivering results.

The next period heralded the third stage for the CPS:
I see the third stage of the CPS especially as one in which the CPS is more
outward facing and outcome focused. It is one in which the CPS has an
enhanced role at every stage of the criminal process working in co-operation
with our criminal justice partners while retaining that independence of prose-
cution decision-making which is the hallmark of the CPS, but it also means
increasing the e¬ciency and the accountability which were also part of its reason
for being.

Some of it would involve a change of culture. Lord Goldsmith listed the features
of the new era:
Getting the cases right from the start: This is the most important part of what
happens later. The CPS will have an increased role through a new relationship
of co-operation between the police and the CPS. The Criminal Justice Act will
give the CPS the responsibility for determining the charge in all but routine
Building on the Glidewell co-location and so improving the administration of
cases; reducing bureaucracy; getting operational o¬cers in touch with opera-
tional lawyers.
Improving the review of cases: The focus of review will be not only to keep
careful track of changes which are taking place to see whether cases should be
continued but seeing what should be done to strengthen weak cases. This is a
very important element and part of the culture change needed.
266 Pre-trial criminal proceedings

Taking charge of witnesses: Far too many cases fail because witnesses do not
turn up to give evidence. Changing this is a key part of improving performance.
The West Midlands pilot will show the potentials for one agency to be in the lead
in handling witnesses and that this agency should be the CPS . . .
The new powers will give a role to the prosecutors even at the outset: The new
power for prosecutors™ cautions in the Criminal Justice Act [see p. 258 above
(ed.)] and greater emphasis on getting cases in the right court as changes to
magistrates™ sentencing powers take place mean prosecutors will have from the
start to be concerned with what the eventual disposal ought to be.
The role of the CPS is gradually expanding. A speech by the new DPP, Ken
Macdonald, in May 2005 saw the CPS taking a central role in the criminal justice
system “ controlling the progress of cases from start to ¬nish, dealing directly
with victims and prosecution witnesses and appearing as advocates in many
more cases.370
For an exploration of the ethical dimensions of these developments see J.
Jackson, ˜The Ethical Implications of the Enhanced Role of the Public Prosecu-
tor™, 9 Legal Ethics, 2006, pp. 35“55.
For a critical assessment of the CPS™s lack of e¬ciency see the report of
the House of Commons Public Accounts Committee in October 2006:
Crown Prosecution Service: E¬ective Use of Magistrates™ Courts Hearings. The
report said the CPS, as e¬ectively the largest law ¬rm in the country, needed to
learn how to run its business from the most successful private practices.

Other prosecutors
The Law Officers
The Attorney General has the power to enter a nolle prosequi in cases tried on
indictment, which has the e¬ect of stopping the proceedings. Equally, he can give
or refuse his permission (known as ¬at) in the considerable number of cases where
by statute his consent is required for a prosecution. In 1998 the Law Commission
recommended that the consent requirement should be abolished save where it was
required in the public interest.371 No action followed. In 2007 however reform
seemed increasingly likely as a result of two high-pro¬le cases. One was Lord
Goldsmith™s involvement in the controversial dropping of a major corruption
inquiry by the Serious Fraud Of¬ce into defence sales by BAE Systems to Saudi
Arabia. The other was his insistence that as Attorney General he would be the
person who ultimately would decide whether the Prime Minister and others close
to him should be prosecuted in connection with the ˜cash for honours™ affair.372
Adam Cowell memorial lecture, Law Society, 17 May 2005. He called, inter alia, for the
scrapping of the rule banning the prosecution from conducting pre-trial interviews with
witnesses. Allowing the prosecution to interview witnesses to test for themselves the
strength of their evidence would allow them to reach better informed prosecution
decisions (29 Independent Lawyer, June 2005, p. 1).
Consents to Prosecution, Law Com. No. 255, 1998.
For the argument that the AG™s power should be transferred to the DPP see J. Jackson, ˜Let the
Director Direct™, 157 New Law Journal, 5 January 2007, p. 23.
267 The prosecution process

Serious Fraud Office
The Serious Fraud O¬ce (SFO) was set up by the Criminal Justice Act 1987 fol-
lowing the recommendation of the Roskill Committee on Fraud Trials (1986).
There is a Director, separate from the DPP, but also under the supervision of the
Attorney General. The Director may, in conjunction with the police or any other
person, investigate any o¬ence which appears to him to involve ˜serious or
complex fraud™ and may institute and have conduct of any criminal proceedings
relating to such fraud investigated. So in the case of the SFO the functions of
investigation and prosecution are joined in one organisation.

Customs and Excise
HM Customs and Excise (HMCE) was a large agency enforcing revenue and
regulatory law with a wide range of criminal, civil and administrative enforce-
ment options including prosecution, compounding, seizure, forfeiture and civil
penalties. It conducted large and complex investigations involving surveillance
and undercover investigations. Investigations were carried out by the National
Investigations Service. Prosecutions were the task of the Solicitor™s O¬ce con-
taining a Prosecutions Group. The Solicitor™s O¬ce was solicitor to HMCE gen-
erally. The position was therefore like pre-CPS police prosecuting departments
with the relationship being that of solicitor and client, but while the investiga-
tors decided what to investigate and the solicitors decided if there was su¬cient
evidence, the decision whether to prosecute was taken by administrators. This
led to a series of spectacular failures resulting in a number of severely critical
reports.373 The London City Bond warehouse case, for instance, involved thir-
teen prosecutions involving 109 accused persons. There were no convictions.
The legal aid bill was estimated at £20 million. The Butter¬eld Report on this
case374 recommended that all prosecution functions should be transferred to a
separate prosecuting authority. This was accepted. HMCE and the Inland
Revenue were merged into HM Revenue and Customs. As from April 2005,
prosecutions are handled by the Revenue and Customs Prosecution O¬ce
(RCPO) headed by a Director appointed by and under the supervision of the
Attorney General.375 RCPO will also handle prosecutions resulting from the
work of the Serious Organised Crime Agency established in 2006 under the pro-
visions of the Serious Organised Crime and Police Act 2005.

Prosecution by other public bodies
Many prosecutions are conducted by Government departments, nationalised
industries, local authorities and other statutory bodies. A study done for the

See R.M. White, ˜Investigators and Prosecutors or Desperately Seeking Scotland: Re-
formulation of the “Philips Principle”™ at p. 161, n. 120.
Review of the Criminal Investigations and Prosecutions Conducted by HM Customs and Excise
by Mr Justice Butter¬eld, 2003.
Commissioners for Revenue and Customs Act 2005. See ˜Introducing RCPO™, 170 Justice of the
Peace, 11 February 2006, pp. 91“4.
268 Pre-trial criminal proceedings

Philips Royal Commission on Criminal Procedure showed that they amounted
to something like one-quarter of all prosecutions. Prosecutions were con-
ducted, in order of frequency, by the Post O¬ce (mainly for television licence
o¬ences), the British Transport Police (e.g. for non-payment of fares), the
Department of the Environment (in relation to vehicle excise licences), the
Department of Social Security (for social security frauds), HM Customs and
Excise and Regional Tra¬c Commissioners (for o¬ences connected with the use
of heavy lorries).
Other public bodies with some prosecution functions included the Health
and Safety at Work Inspectorate, Water Authorities, the Inland Revenue,
Department of Trade and Ministry of Agriculture.376
Research has shown that there are extraordinary regional variations in such
prosecutions. Analysis of all 9,689 prosecutions brought by organisations other
than the CPS at three magistrates™ courts in London, Milton Keynes and
Newcastle showed that of the 2,320 cases prosecuted by the Department of
Transport and DVLA, 80 per cent were in Milton Keynes and only 1 per cent
was in the London court. Of the ninety-seven cases brought by the
Environment Agency, 89 per cent were in the Newcastle court.377
For a study of prosecutions by the Health and Safety Executive see K.
Hawkins, Law as Last Resort: Prosecution Decision-Making in a Regulatory
Agency (OUP, 2002).378

Private prosecutions
A private person can bring a prosecution even though he has no direct inter-
est in the matter. (So private bodies such as the NSPCC or NSPCA can bring
prosecutions.) The private prosecutor must persuade a magistrate to issue a
summons which will be refused if it appears to be a vexatious or improper
proceeding. He would also normally have to bear his own costs and if the
prosecution fails he might in addition have to pay something in respect of
the costs of the defence. The DPP has the right to take over a private prose-
cution and the Attorney General has the power to stop one by entering a nolle
A study done for the Philips Royal Commission showed that at that time
private prosecutions were 2 per cent of all prosecutions.379 The great major-
ity were for common assault. In some areas the police had a policy of not
prosecuting in shoplifting cases and all prosecutions for this o¬ence were
brought by retail stores.

See K.W. Lidstone, R. Hogg and F. Sutcli¬e, Prosecutions by Private Individuals and Non-Police
Agencies (Royal Commission on Criminal Procedure, Research Study No. 10, 1980) Table 2.3,
p. 15. G. Slapper, Organisational Prosecutions (Ashgate, 2001).
A critical review in Modern Law Review, July 2004, pp. 704“9 ended: ˜. . . notwithstanding its
¬‚aws it is an important source of material for students of health and safety in particular, and
more generally for those concerned to know and explore the nature and limits of regulatory
enforcement in a capitalist democracy™. Note 376 above, Table 2.13, p. 23.
269 The prosecution process

When the CPS refuses to prosecute or refuses to bring appropriate charges,
sometimes the victim (or in cases of death, a relative) wants to bring a
private prosecution. In R v. Tower Bridge Metropolitan Stipendiary
Magistrate, ex p Chaudhry380 the mother of the deceased victim of a driving
accident tried to get a summons for causing death by reckless driving. The
driver had been charged with summary-only tra¬c o¬ences. The Divisional
Court held that the decision whether to allow a private prosecution to go
forward should be based on consideration of various matters. One was
whether the case had already been investigated by a responsible prosecuting
authority which was pursuing what it considered to be appropriate charges.
Regard should be had to the provision in the Code for Crown Prosecutors “
whether the charges re¬‚ect the seriousness of the o¬ence and give the court
su¬cient sentencing powers and whether the charges can be presented in a
clear and simple manner. A second consideration would be whether the issue
of a summons for a more serious o¬ence would override the discretion of
the CPS in a way that would be oppressive to the defendant. Thirdly, the
court should bear in mind that the DPP could always intervene to discon-
tinue the proceedings under s. 6(2) of the Prosecution of O¬ences Act 1985
or to reduce the charges under s. 23. Lord Justice Kennedy suggested that
there would have to be special circumstances ˜such as apparent bad faith on
the part of the public prosecutor™ (at p. 51). The court refused the applica-
tion for judicial review.
In 1995, the family of eighteen-year-old Stephen Lawrence succeeded in
launching a private prosecution for murder after the CPS dropped charges
against two teenagers who had been charged with the killing. It was believed to
be the ¬rst time that a private prosecution had been brought in this country in
a murder case. Stephen Lawrence, who was black, was killed by a white gang in
a racially motivated murder. The case aroused much public attention and a fund
was established to pay for the private prosecution. However, in the event, the
trial of the three defendants collapsed after the judge ruled that crucial eye
witness evidence for the prosecution was too unreliable to be put before the
See also Elguzouli-Daf v. Metropolitan Police Commissoner382 holding that the
CPS owes no duty of care toward defendants such that they could sue for fail-
ure to dismiss charges earlier after the forensic evidence against them had been

[1994] 1 All ER 44. See also R v. Bow Street Stipendiary Magistrate, ex p South Coast Shipping
Co Ltd [1993] 1 All ER 219.


. 12
( 34)