Lord Diplock
my lords,
The appellant was indicted at the Old Bailey before Judge Buzzard and a
jury for conspiracy to utter counterfeit American bank notes. On his arraign-
ment he pleaded not guilty to the charge and, in the absence of the jury, alleged,
through his counsel, that he had been induced to commit the offence by an
informer acting on the instructions of the police, and that, but for such per-
suasion, he would not have committed any crime of the kind with which he was
charged. Faced, as he was, by recent decisions of the Criminal Division of the
Court of Appeal that "entrapment" is no defence in English law (Reg. v.
McEvilly and Lee [1974] Crim.L.R. 239; Reg. v. Mealey & Sheridan [1974]
60 Cr.App.R. 59), counsel for the appellant sought to achieve by a different
means the same effect as if it were. He submitted that if the judge were satisfied
at a "trial within a trial" that the offence was instigated by an agent provocateur
acting on the instructions of the police and, but for this, would not have been
committed by the accused, the judge had a discretion to refuse to allow the
prosecution to prove its case by evidence.
In support of this submission counsel was able to cite a number of dicta from
impressive sources which, on the face of them, suggest that judges have a
very wide discretion in criminal cases to exclude evidence tendered by the
prosecution on the ground that it has been unfairly obtained. In addition
there is one actual decision of the Court of Criminal Appeal in Reg. v. Payne
[1963] 1 All.E.R. 848 where a conviction was quashed upon the ground that
the judge ought to have exercised his discretion to exclude admissible evidence
upon that ground—though this was not a case of entrapment. Moreover
there had also been a recent decision at the Central Criminal Court (Reg. v.
Ameer and Lucas [1977] Crim.L.R. 104) in a case which did involve an agent
provocateur where Judge Gilles, after a lengthy trial within a trial, had exercised
his discretion by refusing to allow the prosecution to call any evidence to prove
the commission of the offence by the accused.
In order to avoid what promised to be a lengthy "trial within a trial", which
would be fruitless if Judge Buzzard were to rule as a matter of law that he had
no discretion to exclude relevant evidence tendered by the prosecution to
prove the commission of the offence, even though it had been instigated by an
agent provocateur and was one which the accused would never have committed
but for such inducement, the judge first heard legal submissions on this question.
He ruled that even upon that assumption he had no discretion to exclude the
prosecution's evidence. In consequence of this ruling the appellant withdrew his
plea of not guilty and pleaded guilty.
It is only fair to the police to point out that there never was a trial within a
trial. The judge's ruling made it unnecessary to go into the facts relating to the
appellant's claim that he was induced by a police informer to commit a crime
of a kind which but for such persuasion he would never have committed;
so no evidence was ever called to prove that there had been any improper
conduct on the part of the police or of the prosecution.
The appeal to the Criminal Division of the Court of Appeal (Roskill and
Ormrod L.JJ. and Park J.) was dismissed. Their judgment which was delivered
2
by Roskill L.J. includes a helpful and wide-ranging review of the previous cases,
embracing not only those in which agents provocateurs had been involved
but also those in which the existence of a wide discretion in the judge to exclude
any evidence tendered by the prosecution which he considered had been
unfairly obtained, had been acknowledged in obiter dicta by courts of high
authority. As a result of their examination of these authorities they certified
as the point of law of general importance involved in their decision, a much
wider question than is involved in the use of agents provocateurs. It is:
" Does a trial judge have a discretion to refuse to allow evidence—being
evidence other than evidence of admission—to be given in any circum-
stances in which such evidence is relevant and of more than minimal
probative value?"
I understand this question as inquiring what are the circumstances, if
there be any, in which such a discretion arises; and as not being confined to
trials by jury. That the discretion, whatever be its limits, extended to whoever
presides in a judicial capacity over a criminal trial, whether it be held in the
Crown Court or in a magistrates' court was expressly stated by Lord Widgery
C.J. in Jeffrey v. Black [1977] 3 W.L.R. 895, an appeal by the prosecution to a
Divisional Court by way of case stated from magistrates who had exercised
their discretion to exclude evidence of possession of drugs that had been
obtained by an illegal search of the accused's room by the police. The Divisional
Court held that the magistrates had exercised their discretion wrongly in the
particular case; but Lord Widgery C.J., while stressing that the occasions on
which the discretion ought to be exercised in favour of excluding admissible
evidence would be exceptional, nevertheless referred to it as applying to "all
the evidence tendered by the prosecution" and described its ambit in the
widest terms: "If the case is such that not only have the police officers entered
without authority but they have been guilty of trickery, or they have misled
someone, or they have been oppressive, or they have been unfair, or in other
respects they have behaved in a manner which is morally reprehensible,
then it is open to the justices to apply their discretion and decline to allow the
particular evidence to be let in as part of the trial".
One or other of the various dyslogistic terms which Lord Widgery uses to
describe the kind of conduct on the part of the police that gives rise to a judicial
discretion to exclude particular pieces of evidence tendered by the prosecution
can be found in earlier pronouncements by his predecessor Lord Parker of
Waddington, notably in Callis v. Gunn [1964] 1 Q.B. 495 at 502, where he adds
to them false representations, threats and bribes; while unfairness and trickery
are referred to in dicta to be found in a judgment of the Privy Council in
Kuruma v. The Queen [1955] AC 197 at 204, the case which is generally
regarded as having first suggested the existence of a wide judicial discretion
of this kind. What is unfair, what is trickery in the context of the detection
and prevention of crime, are questions which are liable to attract highly
subjective answers. It will not have come as any great surprise to your Lordships
to learn that those who preside over or appear as advocates in criminal trials
are anxious for guidance as to whether the discretion really is so wide as these
imprecise expressions would seem to suggest and, if not, what are its limits.
So, although it may not be strictly necessary to answer the certified question
in its full breadth in order to dispose of the instant appeal I think that your
Lordships should endeavour to do so.
Before turning to that wider question however, I will deal with the narrower
point of law upon which this appeal actually turns. I can do so briefly. The
decisions in McEvilly and Lee and Mealey and Sheridan that there is no defence
of "entrapment" known to English law are clearly right. Many crimes are
committed by one person at the instigation of others. From earliest times at
common law those who counsel and procure the commission of the offence
by the person by whom the actus reus itself is done have been guilty themselves
of an offence, and since the abolition by the Criminal Law Act 1967 of the
distinction between felonies and misdemeanours, can be tried, indicted and
punished as principal offenders. The fact that the counsellor and procurer is a
3
policeman or a police informer, although it may be of relevance in mitigation
of penalty for the offence, cannot affect the guilt of the principal offender;
both the physical element (actus reus) and the mental element (mens rea) of
the offence with which he is charged are present in his case.
My Lords, this being the substantive law upon the matter, the suggestion
that it can be evaded by the procedural device of preventing the prosecution
from adducing evidence of the commission of the offence, does not bear
examination. Let me take first the summary offence prosecuted before
magistrates where there is no practical distinction between a trial and a "trial
"within a trial". There are three examples of these in the books, Brannan v.
Peek [1948] 1 K.B. 68; Browning v. Watson [1953] 2 All.E.R. 775; Reg. v.
Sneddon [1967] 1 W.L.R. 1051. Here the magistrates in order to decide whether
the crime had in fact been instigated by an agent provocateur acting upon police
instructions would first have to hear evidence which ex hypothesi would involve
proving that the crime had been committed by the accused. If they decided that
it had been so instigated, then, despite the fact that they had already heard
evidence which satisfied them that it had been committed, they would have a
discretion to prevent the prosecution from relying on that evidence as proof of
its commission. How does this differ from recognising entrapment as a
defence—but a defence available only at the discretion of the magistrates ?
Where the accused is charged upon indictment and there is a practical
distinction between the trial and a "trial within a trial", the position, as it
seems to me, would be even more anomalous if the judge were to have a
discretion to prevent the prosecution from adducing evidence before the jury
to prove the commission of the offence by the accused. If he exercised the
discretion in favour of the accused he would then have to direct the jury to
acquit. How does this differ from recognising entrapment as a defence—but a
defence for which the necessary factual foundation is to be found not by the
jury but by the judge and even where the factual foundation is so found, the
defence is available only at the judge's discretion.
My Lords, this submission goes far beyond a claim to a judicial discretion
to exclude evidence that has been obtained unfairly or by trickery; nor in any
of the English cases on agents provocateurs that have come before appellate
courts has it been suggested that it exists. What it really involves is a claim to a
judicial discretion to acquit an accused of any offences in connection with
which the conduct of the police incurs the disapproval of the judge. The conduct
of the police where it has involved the use of an agent provocateur may well be
a matter to be taken into consideration in mitigation of sentence; but under
the English system of criminal justice, it does not give rise to any discretion on
the part of the judge himself to acquit the accused or to direct the jury to do so,
notwithstanding that he is guilty of the offence. Nevertheless the existence of
such a discretion to exclude the evidence of an agent provocateur does appear
to have been acknowledged by the Courts-Martial Appeal Court of Northern
Ireland in Reg. v. Murphy [1965] N.I. 138. That was before the rejection of
"entrapment" as a defence by the Court of Appeal in England; and Lord
McDermott C.J. in delivering the judgment of the court relied upon the dicta
as to the existence of a wide discretion which appeared in cases that did not
involve an agent provocateur. In the result he held that the court-martial had
been right in exercising its discretion in such a way as to admit the evidence.
I understand your Lordships to be agreed that whatever be the ambit of the
judicial discretion to exclude admissible evidence it does not extend to excluding
evidence of a crime because the crime was instigated by an agent provocateur.
In so far as Reg. v. Murphy suggests the contrary it should no longer be regarded
as good law.
I turn now to the wider question that has been certified. It does not purport
to be concerned with self incriminatory admissions made by the accused
himself after commission of the crime though in dealing with the question
I will find it necessary to say something about these. What the question is
concerned with is the discretion of the trial judge to exclude all other kinds of
evidence that are of more than minimal probative value.
4
Recognition that there may be circumstances in which in a jury trial the
judge has a discretion to prevent particular kinds of evidence that is admissible
from being adduced before the jury, has grown up piecemeal. It appears first
in cases arising under section 1 proviso (f) of the Criminal Evidence Act 1898,
which sets out the circumstances in which an accused may be cross-examined
as to his previous convictions or bad character. The relevant cases starting in
1913 with R. v. Watson 8 Cr.App.R. 249 are conveniently cited in the speech
of Lord Hodson in Selvey v. D.P.P. [1970] A.C. 304, a case in which this House
accepted that in such cases the trial judge had a discretion to prevent such
cross-examination, notwithstanding that it was strictly admissible under the
statute, if he was of opinion that its prejudicial effect upon the jury was likely
to outweigh its probative value.
Next the existence of a judicial discretion to exclude evidence of similar
facts", even where it was technically admissible, was recognised by Lord du Parq
delivering the opinion of the Privy Council in Noor Mohamed v. The King [1949]
A.C. 182 at p. 192. He put the grounds which justified its exercise rather more
narrowly than they had been put in the "previous conviction" cases to which I
have been referring; but in Harris v. D.P.P. [1952] A.C. 694 Viscount Simon
with whose speech the other members of this House agreed, said that the dis-
cretion to exclude "similar facts" evidence should be exercised where "the
probable effect" [sc. prejudicial to the accused] would be out of proportion
to the true evidential value".
That phrase was borrowed from the speech of Lord Moulton in Rex v.
Christie [1914] A.C. 545 at p.559. That was neither a "previous conviction"
nor a "similar facts" case, but was one involving evidence of an accusation
made in the presence of the accused by the child victim of an alleged indecent
assault and the accused's failure to answer it, from which the prosecution
sought to infer an admission by the accused that it was true. Lord Moulton's
statement was not confined to evidence of inferential confessions but was
general in its scope and has frequently been cited as applicable in cases of cross-
examination as to bad character or previous convictions under the Criminal
Evidence Act 1898 and in "similar facts" cases. So I would hold that there has
now developed a general rule of practice whereby in a trial by jury the judge
has a discretion to exclude evidence which, though technically admissible,
would probably have a prejudicial influence on the minds of the jury, which
would be out of proportion to its true evidential value.
Ought your Lordships to go further and to hold that the discretion extends
more widely than this, as the comparatively recent dicta to which I have already
referred suggest? What has been regarded as the fountain head of all subsequent
dicta on this topic is the statement by Lord Goddard delivering advice of the
Privy Council in Kuruma v. The Queen (ubi sup. at p.204). That was a case in
which the evidence of unlawful possession of ammunition by the accused was
obtained as a result of an illegal search of his person. The Board held that this
evidence was admissible and had rightly been admitted; but Lord Goddard
although he had earlier said that if evidence is admissible "the court is not
concerned with how the evidence was obtained", nevertheless went on to say:
No doubt in a criminal case the judge always has a discretion to disallow
evidence if the strict rules of admissibility would operate unfairly against
the accused. This was emphasised in the case before this Board of Noor
Mohamed v. The King, and in the recent case in the House of Lords of
Harris v. D.P.P. If, for instance, some admission of some piece of evidence,
e.g., a document, had been obtained from a defendant by a trick, no doubt
the judge might properly rule it out".
Up to the sentence that I have italicised there is nothing in this passage to
suggest that when Lord Goddard spoke of admissible evidence operating
unfairly against the accused he intended to refer to any wider aspect of
unfairness than the probable prejudicial effect of the evidence upon the minds
of the jury outweighing its true evidential value; though he no doubt also had
in mind the discretion that had long been exercised in England under the
5
Judges' Rules to refuse to admit confessions by the accused made after the
crime even though strictly they may be admissible. The instance given in
the passage I have italicised appears to me to deal with a case which falls
within the latter category since the document obtained from a defendant
by a trick is clearly analogous to a confession which the defendant has been
unfairly induced to make, and had, indeed, been so treated in R. v. Barker
[1941] 2 K.B. 381 where an incriminating document obtained from the defendant
by a promise of favours was held to be inadmissible.
It is interesting in this connection to observe that the only case that has been
brought to your Lordships' attention in which an appellate court has actually
excluded evidence on the ground that it had been unfairly obtained (Reg. v.
Payne ubi sup.) would appear to fall into this category. The defendant, charged
with drunken driving, had been induced to submit himself to examination by a
doctor to see if he was suffering from any illness or disability, upon the under-
standing that the doctor would not examine him for the purpose of seeing
whether he were fit to drive. The doctor in fact gave evidence of the defendant's
unfitness to drive based upon his symptoms and behaviour in the course of
that examination. The Court of Criminal Appeal quashed the conviction on
the ground that the trial judge ought to have exercised his discretion to exclude
the doctor's evidence. This again, as it seems to me, is analogous to unfairly
inducing a defendant to confess to an offence, and the short judgment of the
Court of Criminal Appeal is clearly based upon the maxim nemo debet prodere
se ipsum.
In no other case to which your Lordships' attention has been drawn has
either the Court of Criminal Appeal or the Court of Appeal allowed an appeal
upon the ground that either magistrates in summary proceedings or the judge
in a trial upon indictment ought to have exercised a discretion to exclude
admissible evidence upon the ground that it had been obtained unfairly or by
trickery or in some other way that is morally reprehensible; though they cover a
wide gamut of apparent improprieties from illegal searches, as in Kuruma v.
Reg. itself and in Jeffrey v. Black [1977] 3 W.L.R. 895 (which must be the high
water mark of this kind of illegality) to the clearest cases of evidence obtained
by the use of agents provocateurs. Of the latter an outstanding example is to
be found in Browning v. J. H. Watson (ubi sup.) where Lord Goddard remitted
the case to the magistrates with a direction that the offence had been proved,
but pointedly reminded them that it was open to them to give the defendant
an absolute discharge and to award no costs to the prosecution.
Nevertheless it has to be recognised that there is an unbroken series of dicta
in judgments of appellate courts to the effect that there is a judicial discretion
to exclude admissible evidence which has been obtained unfairly or by trickery
or oppressively, although except in Reg. v. Payne, there never has been a case
in which those courts have come across conduct so unfair, so tricky or so
oppressive as to justify them in holding that the discretion ought to have been
exercised in favour of exclusion. In every one of the cases to which your
Lordships have been referred where such dicta appear, the source from which
the evidence sought to be excluded had been obtained has been the defendant
himself or (in some of the search cases) premises occupied by him; and the dicta
can be traced to a common ancestor in Lord Goddard's statement in Kuruma v.
Regina which I have already cited. That statement was not, in my view, ever
intended to acknowledge the existence of any wider discretion than to exclude
(1) admissible evidence which would probably have a prejudicial influence
upon the minds of the jury that would be out of proportion to its true evidential
value; and (2) evidence tantamount to a self-incriminatory admission which was
obtained from the defendant, after the offence had been committed, by means
which would justify a judge in excluding an actual confession which had the
like self-incriminating effect. As a matter of language, although not as a matter
of application, the subsequent dicta go much further than this; but in so far
as they do so they have never yet been considered by this House.
My Lords, I propose to exclude, as the certified question does, detailed
consideration of the role of the trial judge in relation to confessions and
6
evidence obtained from the defendant after commission of the offence that is
tantamount to a confession. It has a long history dating back to the days
before the existence of a disciplined police force, when a prisoner on a charge of
felony could not be represented by counsel and was not entitled to give evidence
in his own defence either to deny that he had made the confession, which was
generally oral, or to deny that its contents were true. The underlying rationale
of this branch of the criminal law, though it may originally have been based
upon ensuring the reliability of confessions is, in my view, now to be found in
the maxim nemo debet prodere se ipsum, no one can be required to be his own
betrayer or in its popular English mistranslation "the right to silence". That is
why there is no discretion to exclude evidence discovered as the result of an
illegal search but there is discretion to exclude evidence which the accused
has been induced to produce voluntarily if the method of inducement was
unfair.
Outside this limited field in which for historical reasons the function of the
trial judge extended to imposing sanctions for improper conduct on the part
of the prosecution before the commencement of the proceedings in inducing
the accused by threats, favour or trickery to provide evidence against himself
your Lordships should, I think, make it clear that the function of the judge at a
criminal trial as respects the admission of evidence is to ensure that the accused
has a fair trial according to law. It is no part of a judge's function to exercise
disciplinary powers over the police or prosecution as respects the way in which
evidence to be used at the trial is obtained by them. If it was obtained illegally
there will be a remedy in civil law; if it was obtained legally but in breach of
the rules of conduct for the police, this is a matter for the appropriate
disciplinary authority to deal with. What the judge at the trial is concerned
with is not how the evidence sought to be adduced by the prosecution has
been obtained, but with how it is used by the prosecution at the trial.
A fair trial according to law involves, in the case of a trial upon indictment,
that it should take place before a judge and a jury; that the case against the
accused should be proved to the satisfaction of the jury beyond all reasonable
doubt upon evidence that is admissible in law; and, as a corrollary to this, that
there should be excluded from the jury information about the accused which is
likely to have an influence on their minds prejudicial to the accused which is
out of proportion to the true probative value of admissible evidence conveying
that information. If these conditions are fulfilled and the jury receive correct
instructions from the judge as to the law applicable to the case, the requirement
that the accused should have a fair trial according to law is, in my view,
satisfied; for the fairness of a trial according to law is not all one-sided; it
requires that those who are undoubtedly guilty should be convicted as well as
that those about whose guilt there is any reasonable doubt should be acquitted.
However much the judge may dislike the way in which a particular piece of
evidence was obtained before proceedings were commenced, if it is admissible
evidence probative of the accused's guilt it is no part of his judicial function to
exclude it for this reason. If your Lordships so hold you will be reverting to the
law as it was laid down by Lord Moulton in Christie, Lord du Parq in Noor
Mohamed and Viscount Simon in Harris before the growth of what I believe
to have been a misunderstanding of Lord Goddard's dictum in Kuruma.
I would accordingly answer the question certified in terms which have been
suggested by my noble and learned friend, Viscount Dilhorne, in the course of
our deliberations on this case.
A trial judge in a criminal trial has always a discretion to refuse to
admit evidence if in his opinion its prejudicial effect outweighs its
probative value.
Save with regard to admissions and confessions and generally with
regard to evidence obtained from the accused after commission of the
offence, he has no discretion to refuse to admit relevant admissible
evidence on the ground that it was obtained by improper or unfair
means. The court is not concerned with how it was obtained. It is
7
no ground for the exercise of discretion to exclude that the evidence
was obtained as the result of the activities of an agent provocateur.
I would dismiss this appeal.
Viscount Dilhorne
my lords,
The Court of Appeal (Criminal Division) (Roskill & Ormrod L.JJ. and
Park J.) when dismissing the appellant's appeal from the ruling of H. H. Judge
Buzzard, certified that the following point of law of general public importance
was involved:—
"Does a trial judge have a discretion to refuse to allow evidence—being
evidence other than evidence of admission—to be given in any circum-
stances" (my emphasis) in which such evidence is relevant and of more
than minimal probative value?"
Judge Buzzard had ruled that, if he had discretion to refuse to admit evidence
for the prosecution to prove the offences charged which, he was prepared to
assume for the purpose of the submissions made to him, the appellant had been
induced to commit by an agent provocateur and which would not otherwise
have been committed, it would be inappropriate to exercise it in this case.
He doubted whether he had discretion to refuse to admit relevant admissible
evidence on the ground that it had been unfairly obtained.
So if the question certified is answered in the affirmative, it would have made
no difference to Judge Buzzard's decision. On his rejection of the submissions
the appellant pleaded guilty to the first count of the indictment charging him
with conspiracy to utter counterfeit bank notes and was sentenced to 18 months
imprisonment, a sentence which, we were told, he has now served.
In Reg. v. Ameer and Lucas [1977] Crim.L.R. 104 Judge Gillis held that he
had a discretion to exclude evidence which had been obtained as a result of the
activities of an agent provocateur; and in the exercise of that discretion, he
ruled that the evidence for the prosecution was inadmissible with the result that
the accused were acquitted of serious charges. A similar course was taken in
Reg. v. Foulder, Foulkes and Johns [1973] Crim.L.R. 45 and in Reg. v. Burnett and
Lee [1973] Crim.L.R. 748.
One of the questions to be decided in this appeal is whether these cases were
rightly decided. If they were, it means, to quote from Judge Buzzard's ruling that
"facts which afford no defence to the charge should nevertheless require
the judge to secure the defendant's acquittal before any evidence is
heard by the jury".
If the answer to the question certified is that a judge has not a general and
unfettered discretion to exclude relevant admissible evidence, has he discretion
to do so in some circumstances; and, if so, what are those circumstances and
what are the criteria on which the exercise of that discretion should be based ?
These appear to me to be the questions to be resolved in this appeal.
That a judge has such a discretion in some circumstances is now established
beyond all doubt. He can refuse to allow the cross-examination of an accused
as to character when the provisions of the Criminal Evidence Act 1898 would
permit it and he can refuse to allow the prosecution to call evidence tending
to prove the commission of offences other than those charged. In my opinion
these are not the only cases in which he has that discretion. He can in my
opinion disallow the use in any trial of admissible relevant evidence if in his
opinion its use would be accompanied by effects prejudicial to the accused
which would outweigh its probative value.
8
In Reg. v. Selvey [1970] A.C. 304, a decision of this House which somehow
escaped the attention of the Court of Appeal and of counsel, the Crown
contended that a judge had no discretion to refuse to permit cross-examination
as to character when the Criminal Evidence Act 1898 sanctioned it. Dealing
with this contention, I said after reviewing a number of cases (at p.341):—
"In the light of what was said in all these cases by judges of great eminence,
one is tempted to say, as Lord Hewart said in Dunkley [1927] 1 K.B. 323
that it is far too late in the day even to consider the argument that a judge
has no such discretion. Let it suffice for me to say that in my opinion the
existence of such a discretion is now clearly established".
Lord Hodson in the same case at p.346 said that there were two answers to
the argument that a judge had no such discretion:
"First, there is a long line of authority to support the opinion that there is
such a discretion to be exercised under this subsection. In the second
place, what is, I think, more significant, there is abundant authority
that in criminal cases, there is discretion to exclude evidence, admissible
in law, of which the prejudicial effect against the accused outweighs its
probative value in the opinion of the trial judge".
and Lord Pearce at p.360 said the discretion came:
"from the inherent power of the courts to secure a fair trial for the accused,
or, to use the words of Viscount Simon (in Harris v. Director of Public
Prosecutions) [1952] A.C. 694, 707 the duty of a judge when trying a
charge of crime to set the essentials of justice above the technical rule if
the strict application of the latter would operate unfairly against the
accused".
In the "similar fact" cases of which Noor Mohamed v. The King [1949]
A.C. 182 and Harris v. Director of Public Prosecutions are examples, a similar
conclusion was reached. That the use of evidence of which the probative value
is outweighed by its prejudicial effect should not occur appears first to have been
clearly staled in R. v. Christie [1914] A.C. 545 in the speeches of Lord Moulton
and Lord Reading. That was a case in which the admissibility of a statement
made in the presence of the accused had to be considered and the fact that their
statements were made in that case is a strong indication that the exercise of
this power by a judge is not limited to "character" and "similar fact" cases.
I referred in Selvey to the overriding duty of the judge to ensure that a trial is
fair. His discretion to control the use of relevant admissible evidence is exercised
in the discharge of this duty. It is the use of the evidence, not, save in relation
to confessions and admissions by the accused, the manner in which it has been
obtained with which he is concerned.
Support for this conclusion is to be found in the judgment of Lord Goddard
in Kuruma v. The Queen [1955] A.C.197 where it was contended that evidence
illegally obtained was inadmissible. He rejected this contention, saying:—
"the test to be applied in considering whether evidence is admissible is
whether it is relevant to the matters in issue. If it is, it is admissible and the
court is not concerned with how the evidence was obtained. (My emphasis)
He went on to say:
"No doubt in a criminal case the judge always has a discretion to disallow
evidence if the strict rules of admissibility would operate unfairly against
an accused".
and then referred to Noor Mohamed and Harris. Pausing here, in view of his
reference to those cases one might conclude that when he said "operate unfairly
against the accused", he meant unfairly in the sense that the prejudicial effect
of the evidence outweighed its probative value; but he went on to say:—
"If, for instance, some admission of some piece of evidence, e.g., a document, had been obtained from a defendant by a trick, no doubt the judge might properly rule it out".
9
This instance is not an instance of evidence which a judge can exclude on
account of its prejudicial effect as compared with its probative value and is not
easily reconcilable with his statement that the court is not concerned with how
evidence was obtained. Perhaps when he said this, Lord Goddard was thinking
of admissions and confessions when the court is concerned with the manner in
which they were obtained, and of the decision in Rex v. Barker [1941] 2 K.B.381
where it was held that fraudulently prepared documents produced to a tax
inspector stood
"on precisely the same footing as an oral or a written confession . . .
brought into existence as a result of... a promise, inducement or threat".
In Brannan v. Peek [1948] 1 K.B.68 Lord Goddard strongly criticised the
conduct of a police officer in tricking the accused into the commission of an
offence and in Browning v. Watson (Rochester) Ltd. [1953] 2 All.E.R. 775
after saying
"No court in England has ever liked action by what are called agents
provocateurs resulting in imposing criminal liability."
and sending the case back, holding that the offence was proved, he reminded
the magistrates that they could grant an absolute discharge without making
any order as to costs. If he had thought that the magistrates in their discretion
could have refused to admit evidence on the ground that it was unfairly obtained
and that the accused had been tricked into its commission, there can be little
doubt that he would have said so in this case and also have referred to it in
Brannan v. Peek.
In Reg. v. Payne [1963] 1 W.L.R. 637 the accused was persuaded to allow
himself to be examined by a doctor on being told that it was no part of the
doctor's duty to examine him in order to give an opinion as to his fitness to
drive. The doctor gave evidence that the accused was unfit to drive. Lord
Parker C.J. held that in the exercise of discretion his evidence ought not to
have been admitted
"on the basis that if the [defendant] realised that the doctor was likely to
give evidence on that matter, he might refuse to subject himself to
examination".
The accused was tricked into allowing the examination and the ratio of this
decision appears to be that evidence of the result of the examination should
not have been admitted without the accused's consent to an examination for
that purpose. It is not, I think necessary to decide whether that case was rightly
decided. If Lord Parker based his conclusion on the reasons he gave in Callis v.
Gunn [1964] 1 Q.B. 495, then I think it was wrongly decided. In Callis v. Gunn
after citing Lord Goddard's statement in Kuruma that in every case a judge
has a discretion to disallow evidence if its admission would operate unfairly
against the accused, Lord Parker said :—
"in considering whether admissibility would operate unfairly against a
defendant, one would certainly consider whether it had been obtained
in an oppressive manner, by force or against the wishes of an accused
person. That is the general principle".
and that the overriding discretion would certainly be exercised by excluding the evidence if there was any suggestion of it having been obtained oppressively, by false representations, by a trick, by threats, by bribes, anything of that sort".
In Jeffrey v. Black [1977] 3 W.L.R. 895 Lord Widgery C.J. expressed the
same view, saying that it was open to justices to apply their discretion and to
decline to allow evidence to be given if it had been obtained by police officers
by trickery, oppressive conduct, unfairly or as a result of behaviour which
was morally reprehensible. With great respect I do not think that these observations were correct. I have not been able to find any authority for the general
10
principle enunciated by Lord Parker or for these statements by him and by
Lord Widgery. If there is any authority for it, it conflicts with Lord Goddard's
statement in Kuruma that the court is not concerned with how evidence is
obtained. If obtained in one of the ways referred to, its credibility may be
impaired. That will be a matter for the jury to consider. It cannot be said that in
addition to the probative value of evidence so obtained, it has a prejudicial
effect such as to render the trial unfair to the accused if it is admitted.
It has been held, rightly in my opinion, that entrapment does not constitute a
defence to a charge (Reg. v. McEvilly [1974] 60 Cr.App.R. 150: Reg. v.
Mealey and Sheridan [1974] 60 Cr.App.R. 59). It would indeed be odd if, although
proof that he was incited to commit an offence which he would not otherwise
have done is no defence to a charge, he could not be convicted of the offence
as a result of the exclusion of admissible evidence in the exercise of judicial
discretion. In Kuruma evidence was not held to be inadmissible because it was
illegally obtained. Evidence so obtained must surely be regarded as unfairly
obtained. Evidence may be obtained unfairly though not illegally but it is not
the manner in which it has been obtained but its use at the trial if accompanied
by prejudicial effects outweighing its probative value and so rendering the trial
unfair to the accused which will justify the exercise of judicial discretion to
exclude it.
Where the trial is with a jury, the judge can hear argument and decide whether
or not to exercise his discretion in the absence of the jury. In a trial in a
magistrates' court, that is not possible. When considering the admissibility of
any evidence, the magistrates must know what evidence it is proposed to
tender. If they decide that it is inadmissible, they will ignore it in reaching their
conclusion. In the same way, it falls upon them to decide whether on account
of its prejudicial effect outweighing its probative value, certain evidence should
not be given. Again they will be informed of the nature of the evidence and if
they rule that it should not be admitted, they no doubt will ignore it in reaching
their conclusion.
I do not think that it is possible just to give an affirmative or negative answer
to the question certified. My answer to it has with my consent been incorporated
by my noble and learned friend Lord Diplock in his speech and so I need not
repeat it.
In my opinion Reg. v. Ameer and Lucas, Reg. v. Foulder, Foulkes and Johns and
Reg. v. Burnett and Lee were wrongly decided and this appeal should be dismissed.
Lord Salmon
My lords,
This is a strange appeal which plainly has no hope of succeeding.
The appellant was convicted at the Central Criminal Court of conspiring with
others to utter counterfeit U.S.A. banknotes knowing them to be forged and
with intent to defraud.
Before the case for the Crown was opened, counsel for the accused adopted
the rather strange course of applying to the trial judge to have a "trial within a
"trial" before the trial itself began. He asserted that if he succeeded on "the
"trial within the trial", the judge would be obliged to rule that the Crown could
adduce no evidence against the accused and the jury would then be directed to
bring in a verdict of not guilty. Counsel then explained to the judge the facts
upon which he proposed to rely. They were as follows:— Whilst the accused
had been a prisoner in Brixton Prison, he met a fellow prisoner called Scippo
who, unbeknown to the accused, was alleged to be a police informer and an
agent provocateur. Shortly before the accused was about to be released, Scippo
who seemed to think (rightly) that the accused's business, or part of it, was to
11
deal in forged banknotes, told the accused that he knew of a safe buyer of
forged banknotes and that he would arrange for this buyer to get in touch
with the accused by telephone. Soon after the accused left prison he was
telephoned by a man who posed as a keen buyer of forged banknotes and
enquired whether the accused would sell him any. The accused said that he
would, and a rendezvous was arranged at which the deal was to be completed.
The accused had no idea that the man with whom he had been speaking may,
in fact, have been a sergeant in the police force.
The accused and some of his associates went to the rendezvous carrying with
them a large number of forged $U.S.A. banknotes and walked straight into a
police trap. The forged notes were confiscated and the accused and his comrades
were arrested.
Counsel for the accused hoped to prove the facts which he had opened by the
evidence of the police sergeant and Scippo during "the trial within the trial"
for which he was asking. Counsel submitted that if these facts were proved:
(1) they would establish that the accused had been induced by an agent
provocateur, i.e., the sergeant or Scippo or both, to commit the crime with
which he was charged and which, but for the inducement, he would never have
committed, and that accordingly the law required the judge to disallow any
evidence of the accused's guilt to be called by the Crown; alternatively (2) (a)
the trial judge had a discretion to reject any evidence of the offence because it
had been unfairly obtained and (b) he was bound by the authorities to exercise
that discretion in the accused's favour.
The judge held, rightly, in my view, that he had no such discretion and
rejected the submissions made on behalf of the accused. The accused then
withdrew his plea of not guilty, pleaded guilty and was sentenced to 18 months
imprisonment.
The Court of Appeal (Criminal Division) dismissed the appellant's appeal
from the trial judge's findings and the appellant now appeals to your Lord-
ship's House.
My Lords, it is now well settled that the defence called entrapment does not
exist in English law. Reg. v. McEvilly and Lee [1974] Crim.L.R. 239; Reg. v.
Mealey and Sheridan [1975] 60 C.A.R. 59. A man who intends to commit a
crime and actually commits it is guilty of the offence whether or not he has
been persuaded or induced to commit it, no matter by whom. This being the
law, it is inconceivable that, in such circumstances, the judge could have a
discretion to prevent the Crown from adducing evidence of the accused's
guilt—for this would amount to giving the judge the power of changing or
disregarding the law. It would moreover be seriously detrimental to public
safety and to law and order, if in such circumstances, the law immunised an
accused from conviction. There are, however, circumstances in which an
accused's punishment in such a case might be mitigated, and sometimes greatly
mitigated.
It is only fair to observe that in the present case there was not a shred of
evidence that the police sergeant was an agent provocateur. Even if he had been
told by an informer that the accused was a hardened dealer in forged bank
notes, it would, I think, have been his duty to carry out a test to discover
whether this information was correct—which events show that it obviously
was. No doubt, the accused would not have committed the crime of trying to
sell forged bank notes to the police had he known it was the police. There can,
however, be little doubt that he would have tried to sell the forged notes to
anyone else whom he "considered safe".
I would now refer to what is, I believe, and hope the unusual case, in which a
dishonest policeman, anxious to improve his detection record, tries very hard
with the help of an agent provocateur to induce a young man with no criminal
tendencies to commit a serious crime; and ultimately the young man reluctantly
succumbs to the inducement. In such a case, the judge has no discretion to
12
exclude the evidence which proves that the young man has committed the
offence. He may, however, according to the circumstances of the case, impose
a mild punishment upon him or even give him an absolute or conditional
discharge and refuse to make any order for costs against him. The policeman
and the informer who had acted together in inciting him to commit the crime
should however both be prosecuted and suitably punished. This would be a far
safer and more effective way of preventing such inducements to commit crimes
from being made, than a rule that no evidence should be allowed to prove that
the crime in fact had been committed.
At common law the person who successfully persuades or induces (counsels
or procures) another to commit an offence has always himself been guilty of a
criminal offence and, since the Criminal Law Act 1967, he can be indicted and
punished as a principal offender. He is regarded as being as guilty as the man
he has incited to commit the crime—and often far more culpable.
It is perhaps worth observing that the law relating to crimes caused by
duress is quite different from the law relating to crimes caused by incitement. As
the law now stands, a man who commits any offence under duress except
murder in the 1st degree is entitled to a clear acquittal. I think that serious
consideration should be given to reforming this branch of the law particularly
in view of the mounting wave of terrorism; but this could only be done by
statute. I respectfully agree with that great criminal lawyer Sir James Fitzjames
Stephen when he wrote: ". . . compulsion by threats ought in no case whatever
to be admitted as an excuse for crime, though it may and ought to operate in
mitigation of punishment in most though not in all cases". The punishment
would certainly vary according to the circumstances of the case; sometimes it
might be minimal. See the majority judgment in Abbott v. The Queen [1977]
A.C. 755.
It follows that Reg. v. Ameer and Lucas [1977] Crim.L.R. 104 which laid
down that a trial judge has a discretion to exclude evidence of the accused's
guilt called by the Crown because it had been improperly obtained by the
activities of an agent provocateur was wrongly decided and should be overruled.
There remains the point of law which has been certified to be of general
public importance thus:—
Does a trial judge have a discretion to refuse to allow evidence—being
evidence other than evidence of admission—to be given in any circumstances in which such evidence is relevant and of more than minimal probative value?"
This question relates no doubt to an interesting and important branch of the
criminal law about which a learned treatise might well be written. But, so far
as this appeal is concerned, the answer to this question can only be obiter. I
very much doubt the value of obiter dicta, but out of respect for your Lordships
and for the Court of Appeal which has posed the question, I will deal with it
shortly, without reciting in detail the authorities which your Lordships have
already so thoroughly analysed.
I do not propose to comment upon the obiter dicta in Callis v. Gunn [1964]
1 Q.B. at p.501 or in Jeffrey v. Black [1978] 1 Q.B. at p.498 or in any of the
many other cases which attempt to define the nature of the prosecution's
evidence and the circumstances in which it may be excluded in order to preserve
the fairness of the trial. In my opinion, the decision as to whether evidence
may be excluded depends entirely upon the particular facts of each case and
the circumstances surrounding it—which are infinitely variable.
I consider that it is a clear principle of the law that a trial judge has the power
and the duty to ensure that the accused has a fair trial. Accordingly, amongst
other things, he has a discretion to exclude legally admissible evidence if justice
so requires. See Lord Reid's speech at p.1024 in Myers v. D.P.P. [1965]
A.C.1001.
13
It follows that—
An accused cannot be convicted unless the prosecution proves his guilt
beyond a reasonable doubt. To allow an accused to be convicted when there
is no evidence before the court capable of proving his guilt beyond a reasonable
doubt would obviously be unfair.
A confession by an accused which has been obtained by threats or promises
is inadmissible as evidence against him, because to admit it would be unfair.
The judge has a discretion to exclude evidence procured, after the com-
mission of the alleged offence, which although technically admissible appears to
the judge to be unfair. The classical example of such a case is where the pre-
judicial effect of such evidence would be out of proportion to its evidential
value. Harris v. D.P.P. [1952] A.C. 694, 707; Kuruma v. Regina [1955] AC 197
Regina v. Selvey [1970] A.C. 304.
Very recently, at "a trial within a trial" an accused gave evidence (accepted
as true by the judge) that a confession upon which the Crown wished to rely
was forced out of him; but nevertheless the accused admitted in cross-examina-
tion that the confession was true. The Privy Council ruled that when the trial
was resumed the Crown could not offer evidence or cross-examine the accused
about anything he had said at the "trial within a trial". To allow the Crown to
do so would have been unfair. See Wong Kam-ming v. The Queen [1979] 2
W.L.R. 81.
I recognise that there may have been no categories of cases, other than those
to which I have referred, in which technically admissible evidence proffered by
the Crown has been rejected by the court on the ground that it would make the
trial unfair. I cannot, however, accept that a judge's undoubted duty to ensure
that the accused has a fair trial is confined to such cases. In my opinion the
category of such cases is not and never can be closed except by statute. I
understand that the answer given by my noble and learned friend Lord Diplock
to the certified question accepts the proposition which I have just stated. On
that basis, I respectfully agree with that answer.
My Lords, I would dismiss the appeal.
Lord Fraser of Tullybelton
MY LORDS,
The appellant was charged with conspiring to utter forged U.S. dollar notes.
He alleged that he had been induced by an agent provocateur to commit the
offence and that he would not have committed it if he had not been so induced.
The appellant accepts that the doctrine commonly known as entrapment has no
place in English law and therefore that his allegations, if proved, would not
constitute a defence to the charge against him. The law to that effect is well
established by decisions—see Regina v. McEvilly and Lee [1974] Crim.L.R. 239
and Regina v. Mealey and Sheridan [1974] 60 Cr.App.R. 59—the correctness of
which is not challenged in this appeal. These decisions appear to me to be right
in principle. The assertion by an accused person that he has been induced by some
other person to commit a crime necessarily involves admitting that he has in
fact committed the crime. Ex hypothesi he must have done the necessary act
and have done it intentionally in response to the inducement. All the elements,
factual and mental, of guilt are thus present and no finding other than guilty
would logically be possible. The degree of guilt may be modified by the in-
ducement and that can appropriately be reflected in the sentence—see Reg. v.
Birtles [1969] 53 Cr.App.R. 469 and Browning v. Watson [1953] 1 W.L.R. 1172
where Lord Goddard C.J. pointed out that the court could even grant an
absolute discharge in such circumstances. That is, I believe, in accordance with
common understanding; so when Eve, taxed with having eaten the forbidden
fruit, replied "the serpent beguiled me", her excuse was, at most, a plea in
mitigation and not a complete defence.
14
Nevertheless, although entrapment is not a defence and therefore not a
matter for the jury to consider on the facts, it is argued for the appellant that
the same result can be achieved by the judge, in the exercise of a discretion,
excluding all evidence of an offence which has been procured by an agent
provocateur. If that were the law, it would be very remarkable, but I am
satisfied that it is not so. In my opinion there is no doubt that, whatever discre-
tion the judge may have in his conduct of a criminal trial, it could not extend
so as to allow him to exclude the evidence in a case such as the present. I reach
that opinion on two grounds. First, there are several reported cases in which
the courts have expressed strong disapproval of the activities of agents provo-
cateurs without suggesting that their evidence should be excluded—see for
example Brannan v. Peak [1948] 1 K.B. 68 and Browning v. Watson supra.
Secondly, the relevant discretion of the judge is a discretion to exclude evidence
because the evidence itself is objectionable on certain grounds. But the present
case does not truly raise a question of evidence at all. On the assumed facts
here, the evidence against the accused would not have been obtained improperly
and would not be open to any objection as evidence. The objection to admitting
it would be that the accused had been unfairly induced to commit the offence
which the evidence tended to prove, and that would be in effect letting in the
defence of entrapment. Accordingly I am of opinion that the evidence was
rightly admitted by the learned judge in this case.
The certified question raises a much more general question as to what
discretion to exclude legally admissible evidence is enjoyed by the judge at a
criminal trial in England. The starting point is, in my opinion, that by the law
of England all evidence which is relevant is also admissible—see Kuruma v.
Regina [1955] AC 197 Lord Goddard at p.203. Nevertheless evidence that is
admissible in law may, in certain cases, be excluded by the judge in the exercise
of a discretion which he undoubtedly possesses. One such case is where evidence
of "similar facts" would be admissible, for example to prove guilty intent or to
exclude a defence of accident, but where the judge considers that its effect in
prejudicing the jury against the accused would exceed its value in tending to
prove his guilt. The judge in these circumstances has a discretion to exclude the
evidence not only if its probative weight is "trifling"—see Noor Mohamed v.
The King [1949] A.C. 182-192—but whenever its prejudicial effect would be
"out of proportion to [its] true evidential value"—see Harris v. D.P.P. [1952]
A.C. 694, 707, Viscount Simon quoting Lord Moulton in D.P.P. v. Christie
[1914] A.C. 545. I read the latter expression as meaning that the discretion can
be exercised where the prejudicial value of the evidence would greatly exceed
its probative value. Another such case is that a judge has a discretion to admit
or exclude evidence of the previous record or character of the accused and to
allow him to be cross-examined as to his character notwithstanding that such
evidence or cross-examination may be legally admissible under the Criminal
Evidence Act 1898 section l(f)(ii). In Regina v. Selvey [1970] A.C. 304, Viscount
Dilhorne at p.341 said that the existence of such a discretion was "now clearly
established".
These cases are in my opinion examples of the exercise of a single discretion
founded upon the duty of the judge to ensure that every accused person has a
fair trial. That is the basis upon which it was put by Lord Goddard in Kuruma
supra, where he said (at p.204):
"No doubt in a criminal case the judge always has a discretion to disallow
evidence if the strict rules of admissibility would operate unfairly against
an accused."
If there were not some underlying discretionary power it would be difficult to
explain how the judges were able, when the Criminal Evidence Act 1898 came
into force, to exclude legally admissible evidence of the type to which I have
referred. The statute does not in terms confer a discretion.
The important question is whether the discretion (a) is limited to excluding
evidence which is likely to have prejudicial value out of proportion to its
15
evidential value or (b) extends to excluding other evidence which might operate
unfairly against the accused and, if so, how far it extends. On the best con-
sideration that I can give to the authorities, I have reached the opinion that the
discretion is not limited to excluding evidence which is likely to have prejudicial
effects out of proportion to its evidential value. I take first the judgment of Lord
Goddard in Kuruma, from which I have already quoted. It is true that im-
mediately after saying the judge has discretion to disallow evidence if the strict
rules would operate unfairly against the accused, Lord Goddard referred to the
cases of Noor Mohamed and Harris, and might therefore seem to have had in
mind only cases that would fall within alternative (a) above. But he went on as
follows:
"If, for instance, some admission or some piece of evidence, e.g., a docu-
ment, had been obtained from a defendant by a trick, no doubt the judge
might properly rule it out. It was this discretion that lay at the root of
the ruling of Lord Guthrie in H.M. Advocate v. Turnbull 1951 J.C. 96. The
other cases from Scotland to which their Lordships' attention was drawn,
Rattray v. Rattray [1897] 25 R. 315, Lawrie v. Muir 1950 JC 19, and
Fairley v. Fishmongers of London 1951 J.C. 14, all support the view that if
the evidence is relevant it is admissible and the court is not concerned
with how it is obtained. No doubt their Lordships in the Court of Justiciary appear at least to some extent to consider the question from the point of view whether the alleged illegality in the obtaining of the evidence could properly be excused . . ."
I find this passage difficult to follow. The case of Lawrie v. Muir was one in
which the court held that evidence obtained by an illegal search of the accused's
business premises was not admissible, because the illegality could not be
excused. With the greatest respect, the case does not seem to me to support the
proposition, for which it was cited by Lord Goddard, that if the evidence is
relevant it is admissible. On the contrary, I think it is an application of the
principle, well established in Scots criminal law, that "An irregularity in the
"obtaining of evidence does not necessarily make that evidence inadmissible",
see Lord Justice-General Cooper at p.27 quoting from Lord Justice Clerk
Aitchison in M'Guigan 1936 J.C. 16, 18. A few lines lower down Lord Cooper
said "Irregularities require to be excused, and infringements of the formalities
of the law in relation to these matters are not lightly to be condoned". Nor
can I agree that what lay at the root of Lord Guthrie's decision in H.M.
Advocate v. Turnbull to exclude documentary evidence was that the evidence
had been obtained from the accused by a trick. The case is in my opinion
another example of evidence obtained from premises occupied by an accused
person, by an irregularity which could not be excused. Lord Guthrie's final
reason at p.103 was that it was unfair to the accused to admit the evidence. He
said this:
"If such important evidence upon a number of charges is tainted by the
method by which it was deliberately secured, I am of opinion that a fair
trial upon these charges is rendered impossible".
The decision in Turnbull may be contrasted with the decision in Jeffrey v.
Black [1978] 1 Q.B. 490, where it was held that the Justices would not have
been entitled to exclude evidence "simply because the evidence in question had
been obtained by police officers who had entered [the accused's residence]
without the appropriate authority". It is not particularly surprising that the
two decisions may not be easily reconcilable because the law on this matter is
not the same in Scotland as it is in England, as has been judicially recognised
on both sides of the border—see King v. The Queen [1969] AC 304, 315 per
Lord Hodson and H.M. Advocate v. Chalmers 1954 J.C. 66 per Lord Justice-
General Cooper at 77, 78. In Chalmers Lord Cooper referred to the English
courts being in use to admit certain evidence which would fall to be rejected
in Scotland. But the principle of fairness to the accused applied by Lord
Guthrie in Turnbull seems to be the same as that stated by Lord Widgery C.J.
in Jeffrey supra, where he said this at 497;
"... the justices sitting in this case, like any other tribunal dealing with
criminal matters in England and sitting under the English law, have a
16
general discretion to decline to allow any evidence to be called by the
prosecution if they think that it would be unfair or oppressive to allow
that to be done."
That was the principle that seems to have been recognised by Lord Goddard
in his reference to Turnbull and treated by him as applicable in England.
Lord Goddard's opinion in Kuruma supra was accepted by Lord Parker of
Waddington C.J. in Callis v. Gunn [1964] 1 Q.B. 495, 501, who said that "as
Lord Goddard C.J. points out, and indeed as is well known, in every criminal
case a judge has a discretion to disallow evidence, even if in law relevant and,
therefore, admissible, if admissibility would operate unfairly against a
defendant. I would add that in considering whether admissibility would
operate unfairly against a defendant, one would certainly consider whether
it had been obtained in an oppressive manner by force or against the wishes
of an accused person. That is the general principle." Later in his judgment
Lord Parker referred to the judge's overriding discretion".
Essentially the same principle was stated in King v. The Queen supra, where
Lord Hodson, giving the judgment of the Judicial Committee said this:
"Having considered the evidence and the submissions advanced, their
"Lordships hold that there is no ground for interfering with the way in
"which the discretion has been exercised in this case.
"This is not in their opinion a case in which evidence has been obtained
"by conduct of which the Crown ought not to take advantage. If they had
"thought otherwise they would have excluded the evidence even though
"tendered for the suppression of crime."
I recognise that there does not appear to be any decision by an appellate
court in England clearly based upon an exercise of the discretion except when
the excluded evidence either (1) is more prejudicial than probative or (2) relates
to an admission or confession. I do not regard the case of Regina v. Payne
[1963] 47 Cr.App.R. as an authority in favour of such a discretion. The
Court of Criminal Appeal held that evidence described by Lord Parker as
"clearly admissible" ought to have been excluded and the conviction was
quashed on that ground. The evidence in question was that of a doctor relating
to a medical examination of an accused person who was charged with driving
a motor car under the influence of drink. The accused had been induced by a
trick to permit (and, I would suppose, co-operate in) a medical examination of
himself and thus to provide material for incriminating evidence by the doctor
who examined him and I regard the decision as being based, at least in part, on
the maxim nemo tenetur se ipsum accusare. But notwithstanding the absence of
direct decision on the point, the dicta are so numerous and so authoritative that
I do not think it would be right to disregard them, or to treat them as applicable
only to cases where the prejudicial effect of the evidence would outweigh its
probative value. If they had been intended to have such a limited application,
the references to the Scottish cases would be inexplicable. In any event, I would
be against cutting down their application to that extent.
On the other hand, I doubt whether they were ever intended to apply to
evidence obtained from sources other than the accused himself or from premises
occupied by him. Indeed it is not easy to see how evidence obtained from other
sources, even if the means for obtaining it were improper, could lead to the
accused being denied a fair trial. I accordingly agree with my noble and learned
friends that the various statements with regard to the discretion to which I have
referred should be treated as applying only to evidence and documents obtained
from an accused person or from premises occupied by him. That is enough to
preserve the important principle that the judge has an overriding discretion to
exclude evidence, the admission of which would prevent the accused from
having a fair trial. That discretion will be preserved if the question in this
appeal is answered in the way proposed in paragraph (2) at the end of the
speech of my noble and learned friend, Lord Diplock, with which I agree.
17
The result will be to leave judges with a discretion to be exercised in accor-
dance with their individual views of what is unfair or oppressive or morally
reprehensible. These adjectives do undoubtedly describe standards which are
largely subjective and which are therefore liable to variation. But I do not
think there is any cause for anxiety in that. Judges of all courts are accustomed
to deciding what is reasonable and to applying other standards containing a
large subjective element. In exercising the discretion with which this appeal is
concerned, judges will have the benefit of the decision of this House fixing
certain limits beyond which they should not go and they will also have valuable
guidance of a more general nature in the opinion of Lord Widgery in Jeffrey v.
Black supra. I do not think it would be practicable to attempt to lay down any
more precise rules because the purpose of the discretion is that it should be
sufficiently wide and flexible to be capable of being exercised in a variety of
circumstances that may occur from time to time but which cannot be foreseen.
I have referred throughout to evidence being excluded by the judge from
consideration by the jury, but it follows of course that the same evidence ought
to be excluded by magistrates from their own consideration in cases where they
are the judges both of law and of fact.
I would dismiss the appeal, and answer the question in the way proposed by
my noble and learned friend Lord Diplock.
Lord Scarman
MY LORDS,
The certified question, though superficially concerned with the exercise of a
criminal judge's discretion as to the admission of evidence, raises profound
issues in the administration of criminal justice. What is the role of the judge?
How far does his control of the criminal process extend? It is his duty, as we all
know, to ensure that an accused has a fair trial: but what does "fair" mean in
this context? And does not the prosecution also have rights which the judge
may not by the exercise of his discretion override? These problems lie at the
root of the criminal justice of a free society.
The drama of the common law wears two faces. The first, and sterner, face
is that subject to exceptions, of which hearsay evidence is far the most import-
ant, evidence which a judge rules relevant is admissible, however obtained.
"It matters not how you get it; if you steal it even, it would be admissible"-
this was the stark assertion of principle by Crompton J. in R. v. Leatham [1861]
8 Cox 498 at 501. The modern formulation of the principle is to be found in the
opinion of the Judicial Committee of the Privy Council prepared by Lord
Goddard C.J. in Kuruma v. The Queen [1955] AC 197 at p.203:-
"In their Lordships' opinion the test to be applied in considering whether
evidence is admissible is whether it is relevant to the matters in issue. If
it is, it is admissible, and the court is not concerned with how the evidence
was obtained."
Consistently with this general rule of evidence the courts have resisted all
attempts to introduce into English substantive law a defence of entrapment:
Reg. v. McEvilly and Lee [1974] Crim. L. R. 239. In Reg. v. Mealey and Sheridan
[1974] 60 Cr.App.R. 59 at p.62, Lord Widgery C.J. said bluntly:-
"It is abundantly clear on the authorities, which are uncontradicted on
this point, that if a crime is brought about by the activities of someone
who can be described as an agent provocateur, although that may be an
important matter in regard to sentence, it does not affect the question of
guilty or not guilty."
18
Brannan v. Peek in 1948 [1948] 1 K.B. 68, Browning v. Watson in 1953, [1953]
1 W.L.R. 1172, Reg. v. Birtles in 1969 [1969] 1 WLR 1047, illustrate that this
is the practice of the courts, even when strongly critical of police methods in
the obtaining of evidence. The authorities are, my Lords, soundly based. It
would be wrong in principle to import into our law a defence of entrapment.
Incitement is no defence in law for the person incited to crime, even though the
inciter is himself guilty of crime and may be far the more culpable. It would
confuse the law and create unjust distinctions if incitement by a policeman or
an official exculpated him whom they incited to crime whereas incitement by
others—perhaps exercising much greater influence—did not. There are other
more direct, less anomalous, ways of controlling police and official activity
than by introducing so dubious a defence into the law. The true relevance of
official entrapment into the commission of crime is upon the question of
sentence when its mitigating value may be high: see Birtles supra.
The second, and merciful, face of the law is the criminal judge's discretion
to exclude admissible evidence if the strict application of the law would operate
unfairly against the accused. Viscount Simon so described the discretion in
Harris v. D.P.P. [1952] A.C. 694 at 707, but immediately proceeded to limit its
scope. He said:—
"If such a case arose, the judge may intimate to the prosecution that
evidence of 'similar facts' affecting the accused, though admissible,
should not be pressed because its probable effect 'would be out of propor-
tion to its true evidential value' (per Lord Moulton in Director of
Public Prosecutions v. Christie). Such an intimation rests entirely within
the discretion of the judge."
In this passage Lord Simon was certainly not envisaging a power in the judge to
stop the prosecution prosecuting, or presenting admissible evidence in support
of its case. He was speaking not of judicial power but of judicial influence; of
a judicial practice, not a rule of law. In so limiting the discretion he was agree-
ing with the views expressed by this House in D.P.P. v. Christie [1914] A.C. 545,
and in no way differing from the famous interjection of Lord Halsbury L.C.
(reported only in the Criminal Appeal Reports (1914) 10 Cr.App.R. 141 at
149) that he
"must protest against the suggestion that any judge has the right to exclude
evidence which is in law admissible, on the ground of prudence or
discretion, and so on."
I do not review the authorities as to the existence and scope of the discretion:
for the task has already been done by others of your Lordships and by the
Court of Appeal. There is also a valuable review of the law by the Law Com-
mission: Law Com. 83, paragraphs 5.7 to 5.20. The problem is however complex.
Is there one discretion or are there several? What is the scope of it (or each of
them)? Upon what principles should the discretion be exercised in modern
conditions?
In my judgment, certain broad conclusions emerge from a study of the case
law. They are:—
that there is one general discretion, not several specific or limited discre-
tions ;
that the discretion now extends further than was contemplated by Lord
Halsbury, and Lord Moulton in Christie's case, or even by Lord Simon in
Harris' case: it is now the law that "a judge has a discretion to exclude legally
admissible evidence if justice so requires" (Lord Reid in Myers v. D.P.P
[1965]A.C. 1001 at p. 1024);
that the formula of prejudical effect outweighing probative value which
has been developed in the "similar fact" cases, is not a complete statement of
the range or the principle of the discretion;
19
(4) that the discretion is, however, limited to what my noble and learned
friend Viscount Dilhorne, calls the "unfair use" of evidence at trial: it does not
confer any judicial power of veto upon the right of the prosecution to prosecute
or to present in support of the prosecution's case admissible evidence, however
obtained.
These broad conclusions leave unresolved the critical question as to the
limits of the discretion and the principle upon which it is founded. It may be,
as Lord MacDermott C.J. said in Reg. v. Murphy [1965] N.I. 138 at p.149 that
unfairness, which will be found to be its modern justification, cannot be closely
defined. One must, however, emerge from that last refuge of legal thought-
that each case depends on its facts—and attempt some analysis of principle.
It is tempting to accept that there are several discretions specific to certain
situations. Certainly the law has developed by reference to specific situations
in which admissible evidence has been either excluded or said to be liable, at
the judge's discretion, to be excluded.
A discretion has been recognised to exclude "similar fact" evidence where its
prejudicial effect would outweigh its probative value: Noor Mohamed v. The King
[1949] A.C. 182. A discretion to refuse to permit a cross-examination of the
accused to his record, though permissible under the Criminal Evidence Act
1898, was recognised by this House in Selvey v. D.P.P. [1970] A.C. 304. Other
relevant evidence may also be excluded. Examples are:— a voluntary confession
obtained in breach of the Judges' Rules: evidence obtained where the defendant
has been misled into providing it (Reg. v. Payne [1963] 1 W.L.R.637); evidence
obtained illegally after the commission of the offence (Kuruma's case, supra).
The instances of actual exclusion are rare: but too many distinguished judges
have said that the discretion exists for there to be any doubt that it does.
Notwithstanding its development case by case, I have no doubt that the
discretion is now a general one in the sense that it is to be exercised whenever
a judge considers it necessary in order to ensure the accused a fair trial. Selvey's
case can be seen to be of critical importance. Viscount Dilhorne, though he
was directing his attention to the specific situation in that case (cross-examina-
tion of the accused to his record) referred to cases concerned with other situa-
tions, e.g. Christie, Noor Mohamed, Harris and Kuruma, and concluded by
saying (pp.341-342):—
"It [i.e. its exercise] must depend on the circumstances of each case
and the overriding duty of the judge to ensure that a trial is fair" (my
emphasis).
Lord Hodson, Lord Guest and Lord Pearce, with whom Lord Wilberforce
agreed, were clearly of the opinion that the discretion was a general one.
Lord Hodson said (p.349):—
"Discretion ought not to be confined save by the limits of fairness."
Lord Guest (p.352) said the discretion "springs from the inherent power of the
judge to control the trial before him and to see that justice is done in fairness
to the accused": and Lord Pearce echoed his words (at p.360F).
The review of the authorities by this House in Selvey's case reveals how com-
paratively recent a judicial development this discretion is. Its history is associated
with the recognition of the admissibility of "similar fact" evidence. As this
rule of evidence became established, judges were alert to prevent its abuse
where probative value was slight and prejudicial effect upon a jury likely to be
great. But other more basic matters contributed to the development: in parti-
cular, the common law principle against self-incrimination, and the side-
effects of the Criminal Law Evidence Act 1898 which by conferring upon the
accused the right to give evidence on his own behalf exposed him to the perils
of cross-examination. Against this comparatively modern background the
judges have had to discharge their duty of ensuring the accused a fair trial.
20
Long before 1898, however, the courts were faced with the problem of re-
conciling fairness at trial with the admissibility of evidence obtained as a
consequence of an inadmissible confession. The problem was resolved in R. v.
Warickshall [1783] 1 Leach 263 by the court declaring, p.264:—
"Facts thus obtained, however, must be fully and satisfactorily proved,
without calling in the aid of any part of the confession from which they
may have been derived."
The discovery of the stolen goods in that case, or (as in Reg. v. Berriman
(1854) 6 Cox 388) the finding of the remains of the corpse, is the best possible
evidence of the truth of the confession (compare and contrast the Canadian
approach in the Supreme Court decision R. v. Wray, 11 D.L.R. 673): but in
English law the confession is inadmissible, not because it is unreliable (its
reliability is established by what has been found), but because to admit it
would be unfair. Similar considerations influenced the judges after 1898 in
protecting an accused from a permissible cross-examination to his record or in
excluding admissible "similar fact" evidence. At first, the judge would be
expected to use his influence (Christie's case, supra) to dissuade the prosecution
from doing what the statute or the common law allowed: but by the time
Reg. v. Kuruma supra was decided influence had become power. Lord Goddard
C.J. was able to say, at p.204, in that case:—
"No doubt in a criminal case the judge always has a discretion to disallow
evidence if the strict rules of admissibility would operate unfairly against
"an accused."
Christie's case is, therefore, only a staging-post in the development of the law.
The modern discretion is a general one to be exercised where fairness to the
accused requires its exercise.
Authority therefore, strongly suggests that the discretion is based upon,
and is co-extensive with, the judge's duty to ensure that the accused has a
fair trial according to law. The two faces of the law reveal the nature and limits
of this duty. The accused is to be tried according to law. The law, not the
judge's discretion, determines what is admissible evidence. The law, not the
judge, determines what defences are available to the accused. It is the law that,
subject to certain recognised exceptions, evidence which is relevant is admiss-
ible. It is the law that there is no defence of entrapment. The judge may not
use his discretion to prevent a prosecution being brought merely because he
disapproves of the way in which legally admissible evidence has been obtained.
The judge may not by the exercise of his discretion to exclude admissible
evidence secure to the accused the benefit of a defence unknown to the law.
Judges are not responsible for the bringing or abandonment of prosecutions:
nor have they the right to adjudicate in a way which indirectly usurps the
functions of the legislature or jury. For legislation would be needed to intro-
duce a defence of entrapment: and, if it were to be introduced, it would be for
the jury to decide whether in the particular case effect should be given to it.
I can now answer the questions posed at the beginning of this opinion. The
role of the judge is confined to the forensic process. He controls neither the
police nor the prosecuting authority. He neither initiates nor stifles a prosecu-
tion. Save in the very rare situation, which is not this case, of an abuse of the
process of the court (against which every court is in duty bound to protect
itself), the judge is concerned only with the conduct of the trial. The Judges'
Rules, for example, are not a judicial control of police interrogation, but notice
that, if certain steps are not taken, certain evidence, otherwise admissible,
may be excluded at the trial. The judge's control of the criminal process begins
and ends with trial, though his influence may extend beyond its beginning and
conclusion. It follows that the prosecution has rights, which the judge may not
override. The right to prosecute and the right to lead admissible evidence in
support of its case are not subject to judicial control. Of course when the
prosecutor reaches court, he becomes subject to the directions as to the conduct
of the trial by the judge, whose duty it then is to see that the accused has a
fair trial according to law.
21
What does "fair" mean in this context? It relates to the process of trial.
No man is to be compelled to incriminate himself; "nemo tenetur se ipsum
prodere". No man is to be convicted save upon the probative effect of legally
admissible evidence. No admission or confession is to be received in evidence
unless voluntary. If legally admissible evidence be tendered which endangers
these principles (as, for example, in Reg, v. Payne, supra), the judge may
exercise his discretion to exclude it, thus ensuring that the accused has the
benefit of principles which exist in the law to secure him a fair trial: but he
has no power to exclude admissible evidence of the commission of a crime,
unless in his judgment these principles are endangered.
In the light of these principles this appeal presents no difficulty. The learned
trial judge dealt with the case upon certain assumptions: he said:—
"For the purposes of these submissions I am prepared to assume that
Sang was induced by an agent provocateur to commit these crimes,
and that he would not have committed them if he had not been so
induced".
The crimes were conspiracy to utter counterfeit bank notes (U.S. dollars)
and unlawful possession of forged notes. After a full and illuminating review
of the case law the judge drew a distinction between evidence being unfairly
obtained and activity being unfairly induced. He held that the discretion
arose only in the case of evidence unfairly obtained. The distinction is a genuine
one; it does not, however, answer the question which the judge must ask
himself. That question is whether the use of the evidence at the trial would be
fair or unfair to the accused. And fairness has to be determined in the light of the
principles to which I have referred and in the context of the particular facts.
In the present case, the (assumed) evidence of crime was clear. The fact that the
criminal conduct was (upon the assumptions made) incited by an "agent
provocateur" did not, as a matter of law, diminish its criminality or weaken the
probative value of the evidence. There was, therefore, no justification for the
exercise of the discretion to exclude the evidence. Had the evidence of crime
which was legally admissible, been excluded, the judge would have made
prosecution impossible for a crime of which there was available unimpeachable
and credible evidence: and, in so doing, would have exceeded his function.
I would, therefore, dismiss the appeal.
My Lords, I am acutely aware that the rest of my speech is "obiter". I
trespass upon your Lordships' time only because unless I do so, I am unable
to answer the question certified by the Court of Appeal.
The development of the discretion has, of necessity, been largely associated
with jury trial. In the result, legal discussion of it is apt to proceed in terms of
the distinctive functions of judge and jury. No harm arises from such traditional
habits of thought, provided always it be borne in mind that the principles of the
criminal law and its administration are the same, whether trial be (as in more
than 90 per cent of the cases it is) in the magistrates' court or upon indictment
before judge and jury. The magistrates are bound, as is the judge in a jury trial,
to ensure that the accused has a fair trial according to law; and have the same
discretion as he has in the interests of a fair trial to exclude legally admissible
evidence. No doubt, it will be rarely exercised. And certainly magistrates
would be wise not to rule until the evidence is tendered and objection is taken.
Assumptions, such as Judge Buzzard made in this case, should never be made
by magistrates. They must wait and see what is tendered; and only then, if
objection be taken, rule. When asked to rule, they should bear in mind that
it is their duty to have regard to legally admissible evidence, unless in their
judgment the use of the evidence would make the trial unfair. The test of
unfairness is not that of a game: it is whether in the light of the considerations
to which I have referred the evidence, if admitted, would undermine the justice
of the trial. Any closer definition would fetter the sense of justice, upon which
in the last resort all judges have to rely: but any extension of the discretion,
such as occurred in Reg. v. Ameer and Lucas [1977] Crim.L.R. 104—to which
22
my noble and learned friends, Lord Diplock and Viscount Dilhorne, have
referred with disapproval—would also undermine the justice of the trial.
For the conviction of the guilty is a public interest, as is the acquittal of the
innocent. In a just society both are needed.
The question remains whether evidence obtained from an accused by
deception, or a trick, may be excluded at the discretion of the trial judge.
Lord Goddard C.J. thought it could be: Kuruma at p.204. Lord Parker CJ.
and Lord Widgery C.J. thought so too: see Callis v. Gunn [1964] 1 Q.B. 495 at
502 and Jeffrey v. Black [1977] 3 W.L.R. 895. The dicta of three successive
Lord Chief Justices are not to be lightly rejected. It is unnecessary, for the
purposes of this appeal, to express a conclusion upon them. But, always
provided that these dicta are treated as relating exclusively to the obtaining of
evidence from the accused, I would not necessarily dissent from them. If an
accused is misled or tricked into providing evidence (whether it be an admission
or the provision of finger-prints or medical evidence or some other evidence),
the rule against self-incrimination—"nemo tenetur se ipsum prodere"—is likely
to be infringed. Each case must, of course, depend on its circumstances. All
I would say is that the principle of fairness, though concerned exclusively with
the use of evidence at trial, is not susceptible to categorisation or classification,
and is wide enough in some circumstances to embrace the way in which, after
the crime, evidence has been obtained from the accused.
For these reasons I agree with the answer to the certified question in the
terms proposed by my noble and learned friends. Lord Diplock and Viscount
Dilhorne.
In reaching my conclusion that the discretion is a general one designed to
ensure the accused a fair trial, I am encouraged by what I understand to be
the Scots law. Such research as I have been able to make makes clear that the
Scots judges recognise such a discretion. Indeed, I think they go further than
the English law, the Scots principle being that evidence illegally or unfairly
obtained is inadmissible unless in the exercise of its discretion the court allows
it to be given. Sheriff MacPhail in his massive research paper on the law of
evidence in Scotland (April 1979) describes the existing law in these terms
(para. 21.01):-
"In criminal cases, on the other hand, evidence illegally or irregularly
obtained is inadmissible unless the illegality or irregularity associated
with its procurement can be excused by the court".
It would appear that the principal authority is the Full Bench case of Lawrie v.
Muir [1950] JC 19, in which Lord Justice-General Cooper, after saying that
irregularities require to be excused, continued (pages 26-27):—
"In particular, the case may bring into play the discretionary principle of
fairness to the accused which has been developed so fully in our law
in relation to the admission in evidence of confessions or admissions by
a person suspected or charged with crime. That principle would obviously
require consideration in any case in which the departure from the strict
procedure had been adopted deliberately with a view to securing the
admission of evidence obtained by an unfair trick".
How far the Scots judges have extended "the discretionary principle of
"fairness to the accused" I am not qualified to say. It is, however, plain that
by the law of Scotland it may be invoked in a case where, after the commission
of the crime, illegal or irregular methods have been used to obtain evidence
from the accused: see also H.M. Advocate v. Turnbull [1951] J.C. 96 per Lord
Guthrie at p. 103. Though differences of emphasis and scope are acceptable,
it would be, I think, unfortunate if the "discretionary principle of fairness to
the accused" was not recognised in all the criminal jurisdictions of the United
Kingdom. Indeed, it must be a fundamental principle in all British criminal
jurisdictions that the court is under the duty to ensure the accused a fair trial:
and I do not believe that a judge can effectually discharge his duty without,
at the very least, the availability of the discretion I have endeavoured to describe.