. 28
( 34)


under the European Convention on Human Rights challenging the legality of
the power was rejected by the Strasbourg Court in March 1987. The European
Court of Human Rights was told that, although there were no statistics, loss of
time was ordered in some sixty or so cases per year by the single judge or the full
court. The normal order was for twenty-eight days to be added on, though such
orders ranged from seven days to sixty-four days.103 (In later years the number
of such orders is much lower still. In 1998 there were four, in 1999 two, in 2000
none and in 2001 two. In April 2005 the Court of Appeal said it was a power
that had been too little used.104)
Research conducted for the Runciman Royal Commission on Criminal
Justice showed that there was a great deal of misinformation in the prisons
about the time loss rules. Many prisoners were under the erroneous impression
that all the time spent appealing could be added on by the Court of Appeal.
(This error was less surprising when seen against the fact that many solicitors
appeared to share the same misapprehension and that over half of all solicitors
responding to the survey thought that the Court of Appeal still had the power
to increase sentences which had been abolished twenty-¬ve years earlier!) A
third of the sample of prisoners who did not appeal said the threat of time being
added on had been the reason.105
The Runciman Commission recommended that prisoners “ and lawyers “ be
made aware of the true position. (˜We think it wrong that appellants who spend
several months awaiting appeal should be left with the impression that if they
fail, those months will be added to their sentences. Nor should they have reason
to fear that the Court of Appeal will increase their sentence™.106) It recom-
mended that the Court of Appeal issue a new Practice Direction dealing with
the issue and that the o¬cial guides issued by the Criminal Appeal O¬ce, the
Bar Council and the Law Society make matters clear, even though the result
would be likely to be an increase in the number of applications for leave to
appeal. (˜We would regard it as an unavoidable result of correcting an impor-
tant piece of misinformation common among prisoners™ (ibid).) This recom-
mendation, however, was not acted upon.
The great majority of appeals are against sentence. In the ten years 1995“2005
the number of applications for leave to appeal against conviction ¬‚uctuated

Monnell and Morris v. United Kingdom (1987) 10 EHRR 205.
Kuimba (2005) Times, 17 May. See also Hart, George, Clarke, Brown (2007) Times, 16 February.
J. Plotniko¬ and R. Woolfson, Information and Advice for Prisoners about Grounds for Appeal
and the Appeals Process (Royal Commission Research Study No. 18, 1993) pp. 79“82.
Runciman, pp. 165“6, para. 19.
676 Appeals

from a high of 2,393 (in 1995) to a low of 1,661 (in 2005). The number of appli-
cations for leave to appeal against sentence ¬‚uctuated in the same period
between a high of 7,160 (in 1997) and a low of 5,178 (in 2005).107
In 2005, of the applications for leave to appeal considered by a single judge,
24 per cent were granted against conviction and 33 per cent against sentence.108
Of those applications which were refused by the single judge, 50 per cent were
renewed to the full court against conviction and 27 per cent against sentence.109
A quarter (25 per cent) of the renewed applications against conviction and two-
¬fths (40 per cent) of those against sentence were successful.110
Of the appeals heard by the full court during 2005, 37 per cent (twenty-two)
against conviction were allowed and 71 per cent (1,534) against sentence were

Legal advice for appellants in criminal cases
The Access to Justice Act 1999, s. 26 states that ˜representation™ includes advice
and assistance as to any appeal. The o¬cial Guide to practitioners on criminal
appeals states in para 1.1 that no one convicted or sentenced in the Crown Court
should be without advice or assistance on appeal.112 The solicitor™s brief to a
barrister must include instructions to the barrister to give advice and assistance
as to the prospects of an appeal in the event of a conviction or sentence. The
advice must be whether there are grounds of appeal. If so, the grounds must be
drafted. The procedure requires that counsel ¬ll out a form right away at court
telling the client whether it is thought that there are grounds of appeal or
whether counsel needs time to consider the matter. He is required then to
deliver written advice to the solicitor within fourteen days, including, where
appropriate, signed grounds of appeal. The solicitors should send it on to the
client so that he receives it within twenty-one days measured from his convic-
tion or sentence.
Research done for the Runciman Royal Commission showed that in various
respects this system was not at that time functioning as it should. Thus 9 per
cent of prisoners said they had not been visited in the cells at the end of the case
and 23 per cent said they had been visited but an appeal had not been discussed.
The Royal Commission said it regarded these as serious matters and called on
both branches of the profession to ˜take all necessary steps to ensure that prac-
titioners not only perform their duty to see the client at the end of the case, as
most do, but also give preliminary advice both orally and in writing™.113 No later
research has inquired into whether this exhortation has been heeded.

107 108 109 110
Judicial Statistics, 2005, Table 1.7. Ibid. Ibid. Ibid.
Ibid, Table 1.8.
A Guide to Proceedings in the Court of Appeal, Criminal Division [1997] 2 Cr App Rep 459 and
the current edition of Archbold Criminal Pleading Evidence & Practice. The Guide is also
accessible at www.courtservice.gov.uk “ Publications “ Guidance.
J. Plotniko¬ and R. Woolfson, Information and Advice for Prisoners about Grounds for Appeal
and the Appeals Process (Royal Commission Research Study No. 18, 1993) pp. 164“5, para. 14.
677 The practice and procedure of appeals

Where it appears that the defendant has submitted his own grounds, the
Criminal Appeal O¬ce writes to the solicitors who acted at the trial to ask if
advice was given. If the reply is a¬rmative, no doubt the Criminal Appeal O¬ce
assumes that the solicitors were discouraging about the prospects of an appeal.

Appeals by way of case stated
An appeal may be brought against a decision of the magistrates™ court on the
ground that it is wrong in law or in excess of jurisdiction by asking the magis-
trates to state a case to the Divisional Court (Magistrates™ Courts Act 1980,
s. 111(1)). This must be done within twenty-one days. There is no power to give
an extension of time. In a criminal case the prosecution may ask for a case to be
stated, as can the defence. The magistrates draw up a statement of the facts
found, the cases cited, the decision and the issue for the consideration of the
Divisional Court. If the appeal is based on the argument that there was no evi-
dence on which the magistrates could have reached their decision, the case
stated also includes a resum© of the evidence. The court supplies the parties with
a draft of the case to be stated and invites their comments. In the event that a
party is dissatis¬ed with the way in which the case has been put, he can apply to
the Divisional Court asking for the case to be remitted to the magistrates for
restatement of the facts. The magistrates can refuse to state a case on the
grounds that it is a frivolous request. However, an unreasonable refusal to state
a case can be the subject of an application for judicial review.114
As noted above (p. 662) Lord Justice Auld recommended that the right to
appeal from the magistrates™ court to the High Court by way of case stated
should be abolished,115 but this recommendation was not acted upon.

Leapfrog appeals
In 1969 a new procedure was devised to enable appeals to go direct from the
High Court to the House of Lords in certain limited circumstances.

To the House of Lords
Administration of Justice Act 1969
Section 12(3) “ that a point of law of general public importance is involved in
that decision and that that point of law either:
(a) related wholly or mainly to the construction of an enactment or of a statu-
tory instrument, and has been fully argued in the proceedings and fully con-
sidered in the judgment of the judge in the proceedings, or
(b) is one in respect of which the judge is bound by a decision of the Court
of Appeal or of the House of Lords in previous proceedings, and was fully

Sunworld Ltd v. Hammersmith and Fulham London Borough Council [2000] 2 All ER 837, Div
Court. See generally A. Murdie, ˜Appeals by Case Stated from the Magistrates™ Court™,
Solicitors™ Journal, 6 October 1995, p. 984; J. A. Backhouse, ˜Right of Appeal by way of Case
Stated “ Should it be Simpli¬ed?™, 156 Justice of the Peace, 16 May 1992, p. 310.
Auld, p. 623.
678 Appeals

considered in the judgments given by the Court of Appeal or the House of
Lords (as the case may be) in those previous proceedings . . .
The power has been used very little.

To the Court of Appeal Civil Division
Where an appeal would otherwise be heard on appeal by the county court or
the High Court it can be transferred direct to the Court of Appeal if the Master
of the Rolls or the court from which the appeal is taken or the court to which
the appeal is going considers that it raises an important point of principle or
practice or ˜there is some other compelling reason for the Court of Appeal to
hear the case™.116

To the Court of Appeal Criminal Division
Lord Justice Auld recommended in his report that the leapfrog appeal be
extended to criminal cases for use where there are con¬‚icting decisions of the
Court of Appeal that can only be resolved by the House of Lords.117 This has not
been implemented.

In the early 1960s a team of eminent English and American judges and lawyers
spent a period in each other™s countries studying the appeal system. The object
was for each to assess the strengths and weaknesses of both systems. A member
of the American team reported on the meeting.118 What follows distils the main
points of comparison at that time and (in editorial square brackets or footnotes)
what has happened to the English system in the intervening forty or more years.
As will appear, although the English system has moved somewhat in the direc-
tion of the American system, many of the di¬erences identi¬ed then are still

The decision
In the United States, almost all decisions are reserved and rendered in
written form. Rarely is one pronounced from the bench. Furthermore, an
attempt is always made to have the judges agree upon an opinion for the court
as a whole, or, if that cannot be done, to secure as broad a base of agreement
as possible.
While concurring opinions are not unusual and even multiple separate dis-
sents not unknown, it is not expected that each judge will express his own views.
The ideal is a unanimous opinion for the court, or, failing that, one majority
opinion and one dissent.
[The English system is moving strongly in that direction. In his 2001“2 annual
report as Master of the Rolls Lord Phillips said: ˜it is now more common for a

116 117
Access to Justice Act 1999, s. 57 and CPR 52.14. Auld, Ch. 12, para. 117, p. 657.
D. Karlen, ˜Appeal in England and the United States™, 78 Law Quarterly Review, 1962, p. 371.
679 The practice and procedure of appeals

constitution of the court to deliver a single judgment to which all members of
the court have contributed. This is a trend which has my support. Profusion of
precedent is the bain of judges and practitioners alike. A single judgment
reduces the material that has to be read, avoids the opportunity for di¬erences
of interpretation and provides greater clarity™ (ed.).119]
In England, few decisions are either reserved or written. In the Court of Appeal,
the practice is for each judge to express his individual views orally and extem-
poraneously immediately upon the close of argument. In the Court of Criminal
Appeal a single opinion for the court is customarily expressed, but almost always
orally and extemporaneously. Only in the House of Lords and the Privy Council
are decisions customarily reserved and written.
[This was the position in the Civil Division and remains the position in the
Criminal Division, but the proportion of cases in which the decision is reserved
in the Civil Division is now very considerable. The Bowman report in 1997 said
that the Court of Appeal reserved judgment in a quarter to a third of cases. The
Master of the Rolls™ annual reports show that the proportion has now risen to
over half.120 As the simpler appeals were dealt with by lower courts, the pro-
portion of complex cases heard by the Court of Appeal in which it was neces-
sary to reserve judgment was inevitably rising (ed.).]
The American approach entails di¬erent internal operating procedures than
are usual in England. Conferences, both formal and informal, are a prominent
feature of American practice. So are exchanges of memoranda and draft opin-
ions. On the other hand, since reading and writing are by their nature solitary
operations, American judges, who are compelled to do much of both, spend
many, if not most, of their working hours alone. They are frequently required
to shift their attention from one case to another and then back again, because,
with cases being heard in batches, several are awaiting decision at any given
To the limited extent that the English practice conforms to the American
pattern, the same internal procedures doubtless apply. In the great majority of
English appeals, however, the judges follow a vastly di¬erent routine. Most of
their working time is spent together sitting on the bench, listening and talking
rather than reading and writing. The discussions they hold are brief and seem-
ingly casual, although highly economical, by reason of the fact that cases are
heard and decided one at a time. The judges™ minds are already focused on the
problems at hand and not distracted by other cases which have been heard and
are awaiting decision. They whisper between themselves on the bench; they con-
verse as they walk to and from the courtroom; and they indirectly make com-
ments to each other as they carry on Socratic dialogues with counsel, but they

For a critical discussion of the trend toward single judgments in the Court of Appeal see R.
Munday, ˜“All for One and One for All” “ The Rise to Prominence of the Composite Judgment
within the Civil Division of the Court of Appeal™, 61 Cambridge Law Journal, 2002,
pp. 321“50.
In the ¬ve years from 2001“2 to 2004“5 it was respectively 52 per cent, 54 per cent, 55 per
cent, 61 per cent and 55 per cent “ Graph 4.
680 Appeals

do not ordinarily exchange memoranda or draft opinions or engage in full scale
In short, the appellate judge in England spends most of his working time in
open court, relatively little in chambers, whereas his counterpart in America
spends most of his working time in chambers, and relatively little in open
court. This is neatly illustrated by the times of sitting for comparable courts in
the two nations. In the United States Court of Appeals for the Second Circuit,
each judge hears arguments one week out of four, and uses the other three for
studying written briefs and records on appeal, conferring with his brother
judges, and writing opinions. By way of contrast, each judge on the English
Court of Appeal hears arguments, day after day, ¬ve days a week, throughout
each term.
[Following Bowman, the judges in the Court of Appeal Civil Division nowadays
have a signi¬cant proportion of working time for writing but the difference
between the two systems remains (ed.).]

Supporting personnel
In the United States, most appellate judges have law clerks, sometimes more
than one. These typically are young men, recently graduated from law school
with ¬ne academic records, who serve for a period of a year or two. They are
chosen by and answerable to the judges, although paid out of public funds. The
services they perform vary greatly from one judge to another, but in general they
carry on research, prepare memoranda, discuss the cases to be decided with the
judges for whom they work, and sometimes even draft opinions or parts of
opinions to be rendered. They participate in the decisional process to the extent
that their judges wish them to participate . . .
In England there are no law clerks.122
[Professor Karlen made the point that since in the English system written briefs
are not used and most opinions are given extemporaneously at the close of oral
argument, it was di¬cult to see what use law clerks would be in most English
appellate courts. As will be seen, skeleton arguments “ mini versions of the fully
argued American written brief “ are now an established part of the English
system, but about half the decisions of the Court of Appeal Civil Division and
the overwhelming majority of the decisions of the Court of Appeal Criminal
Division are still rendered extemporaneously at the end of the case. The young
judicial assistants now employed by the Court of Appeal Civil Division are,
so far at least, a pale re¬‚ection of the US style law clerk. Their work consists

See to like e¬ect K. Malleson, ˜Decision-making in the Court of Appeal: the Burden of Proof
in an Inquisitorial Process™, 1 International Journal of Evidence and Proof, 1997, pp. 175“86
As has been seen, the Court of Appeal Civil Division does now employ young judicial
assistants. For a fascinating, and perhaps disturbing, insider™s description (and critique) of the
way the justices of the US Supreme Court and their clerks operate as seen by a former clerk
see E. Lazarus, Closed Chambers (1999, Penguin). See also G. R. Smith, ˜A Primer of Opinion
Writing for Law Clerks™, 26 Vanderbilt Law Review, 1973, p. 1203 (ed.).
681 The practice and procedure of appeals

typically in preparing summaries and analysis of the issues in a case. They are
not involved in the writing of a judge™s written judgments (ed.).]

In the United States, appellate decisions possess less ¬nality. New trials can be
granted in all types of cases, criminal as well as civil. Rehearings are frequently
asked for and occasionally allowed. Existing side by side with appeals are a variety
of methods of collateral attack, including habeas corpus, sometimes entailing
successive re-examination of a single case by courts of co-ordinate jurisdiction.
Finally, the American doctrine of precedent is such that a decision is never
beyond the reach of challenge in a new lawsuit. If conditions or thinking have
changed, sometimes if only the personnel of the court has changed, there is
always the possibility that the unwanted decision may be overruled.
[The di¬erence between the two civil systems in this respect is even greater today
than it was then, since the Court of Appeal has taken such strong action post-
Bowman to reduce second appeals. As will be seen, retrials in criminal cases are
not quite so rare today as they were then but they are still very rare (ed.).]

Oral argument
In the United States, oral arguments are secondary in importance to the briefs,
and are rigidly limited in duration. In the United States Supreme Court, one
hour is allowed to each side, but in many appellate courts, less time than that is
permitted, frequently no more than ¬fteen minutes or a half-hour for each side.
Reading by counsel is frowned upon. The judges do not wish to hear what they
can read for themselves. They expect to get all the information they need about
the judgment below, the evidence, and the authorities relied upon from study-
ing the briefs and record on appeal. They do not even encourage counsel to
discuss in detail the precedents claimed to govern the decision, preferring to do
that job by themselves in the relative privacy of their chambers, with or without
the assistance of law clerks.
In England, where there are no written briefs,123 oral arguments are all-
important. They are never arbitrarily limited in duration. While some last for
only a few minutes, others go on for many days, even weeks. The only controls
ordinarily exercised over the time of oral arguments are informal, ad hoc sugges-
tions from the judges. Thus when counsel wishes to cite a case as authority, the
presiding judge may ask him: for what proposition? If the judges indicate that
they accept the proposition as stated, there is no need to read the case. Similarly
if counsel has persuaded the judges on a certain point, they may indicate that it
is unnecessary for him to pursue it further. If counsel for the appellant, by the
time he ¬nishes his argument, has failed to persuade the court that the decision
below should be reversed or modi¬ed, the court informs counsel for the respon-
dent that it does not wish to hear from him at all, and proceeds forthwith to
deliver judgment. Despite such controls as these, the time spent in England in
oral arguments tends to be very much greater than that spent in the United States.

As already noted, written ˜skeleton arguments™ are, however, now required “ see below (ed.).
682 Appeals

[As will be seen below, the English system does now make some attempts to
restrict oral argument but so far at least they have not gone far. The basic
di¬erence between the system is still very great indeed (ed.).]
Various steps have been taken to improve the e¬ciency of the Court of
Appeal. One is to pay vastly more attention to getting the parties to prepare the
bundle of documents for the court in proper form.124 Another is to have the
judges pre-read so that when the oral argument commences the judges will be
able to focus on the important issues.

Skeleton arguments
The beginnings of the skeleton argument were in a Practice Note issued in 1983
by Lord Donaldson, the then Master of the Rolls.125 The Practice Note said that
the points which counsel intended to argue should be set out in not more than
one or two sentences together with full references to be used in support of each
point. The skeleton should also contain anything that would otherwise have to
be dictated to the bench such as propositions of law, chronologies of events, lists
of dramatis personae or, where necessary, glossaries of terms. No one would be
held to the contents of such a document. The document should, however, be
sent to the court (and the other side) well before the hearing or, at the latest,
when counsel rose.126
A somewhat similar development had already taken place in the House of
Lords. In 1982 in MV Yorke Motors v. Edwards 127 Lord Diplock set out what the
House of Lords would in future require by way of written documents in a case.
Previously the case presented by the parties would contain a summary of the
facts, the proceedings in the courts below, the judgments and the arguments on
appeal. Now, Lord Diplock said, the case should start ˜with a statement of what
the party conceives to be the issues that arise on the appeal™.128 Counsel should
bear in mind that the members of the appellate committee would have read the
judgments below. Each issue should be mentioned in a sentence or two. If there
were points that it was not intended to pursue, this should be stated; conversely,
if it was intended to take a point that was not argued below, the case should
mention the fact. If there was an intention to ask the House of Lords not to
follow one of its own previous decisions, this should be made clear. Heads of
argument should be prepared setting out the chief authorities to be relied on.
Lord Diplock said that it was not intended to move towards the American
written brief. Counsel for one side had put in a document of thirty-nine pages,
which was far too long. Counsel for the other side had put in one a sixth of that
length, which was perfectly adequate.
What started as an experiment with a voluntary system became mandatory
in 1989. The rule is now stated in the Practice Direction accompanying CPR,
124 125
See Practice Direction 39 PD 3, para. 3.2. [1983] 2 All ER 34.
For a powerful critique of the innovation of skeleton arguments see F. A. Mann, ˜Re¬‚ections
on English Civil Justice and the Rule of Law™, 2 Civil Justice Quarterly, 1983, p. 320.
127 128
[1982] 1 All ER 1024. At 1025.
683 The practice and procedure of appeals

r. 52.4. The post-CPR White Book states that paras. 5.10 and 5.11 of the Practice
Direction ˜replace voluminous earlier practice directions concerning skeleton
arguments™. They should be succinct: ˜The practice of drafting di¬use skeleton
arguments (which some advocates favoured under the former regime) is not
Since 1999 the rule has been that the skeleton argument must be presented
with the application for permission to appeal, failing which within fourteen
days thereafter. Lists of authorities (with the relevant passages marked up) are
supposed to be handed in not less than seven days before the hearing.130
The new approach was criticised by David Pannick QC. He argued that it
would be wasteful for counsel to prepare a skeleton argument many months
before the hearing of the appeal. A skeleton prepared so long in advance
would lack quality and focus. Also, the lawyers would have to prepare the case
twice “ once to draft the skeleton argument and again for the actual appeal
hearing.131 Lord Woolf, replying, said early presentation of the skeleton was
vital if the court was to be able to take the necessary case management

Restrictions on oral argument?
One of the features of American appellate practice, as has been seen, is drastic
restriction of oral argument. This has not yet come to the English system.
Despite the new emphasis on case management by the courts, counsel are still
permitted to argue their case at length “ indeed generally at the length that they
think appropriate.
In 1991 an American scholar, Professor Robert Martineau, spent three
months in the Court of Appeal Civil Division to study the English oral tradi-
tion. He started with the hypothesis that the American system could probably
learn much from the English. He ended with the opposite conclusion.
Moreover, surprisingly, he was not overly impressed with the quality of the oral
advocacy he observed. (˜Most English barristers are not e¬ective appellate advo-
cates™.) The situation in England seemed to him to be pretty much the same as
in the USA. In both countries, he thought, 15 per cent of appellate advocates
were highly competent, 30“40 per cent were competent and 50“60 per cent were

Civil Procedure, 2003, 52.4.3.
Note Haggis v. DPP [2003] EWHC 2481 (Admin), [2004] 2 All ER 382. Lord Justice Brooke
set out the rules and added: ˜It is now high time that practice in this respect is tightened up so
that unnecessary time is not wasted either by the parties waiting for the other side to ¬le
skeleton arguments in accordance with the rules, or by the court in being bombarded at a very
late stage, sometimes after it has already done its pre-reading, with the late arrival of skeleton
arguments and important authorities™ ([34]).
131 132
The Times, 26 January 1999. The Times, 2 February 1999.
W. Hein, Appellate Justice in England and the United States, 1991; and see an article based on
the book by the present writer “ ˜A Brief Encounter™, 141 New Law Journal, 12 April 1991,
p. 491.
684 Appeals

The chief step taken so far towards limited oral argument in the Court of
Appeal has been a rule that counsel is required to give an estimate of time for
the case. There are no penalties as yet for overrunning. Martineau, whose
research was pre-Woolf reform, found that even judges who pre-read the
papers generally left counsel to develop his oral argument in his own way and
at his own length, out of belief in the virtues of the oral tradition. This was
con¬rmed in a paper by Lord Justice Leggatt written for the Anglo-American
judicial exchange in 1994.134 He acknowledged that skeleton arguments help
by telling the judges what appeals are about before they start but ˜it sometimes
e¬ects little perceptible saving of time because counsel are su¬ered to repeat
orally what they have already rendered in writing™. That some presiding judges
allowed that to happen was ˜another example of the oral tradition dying hard™.
He suggested that the court was in that respect falling between two stools
because skeleton arguments (which were sometimes of inordinate length)
were required, yet oral argument essentially was open-ended. The Practice
Direction required counsel to open his appeal by going directly to the ground
of appeal in the forefront of the appellant™s case but this enjoinder was not
always obeyed.
Lord Justice Leggatt said that ˜immoderate periods of time are spent in
informing the courts about the facts and the law, as distinct from presenting the
reasons why they support the cause of the one side or the other™. In 1954 there
had been eight Lord Justices, in 1974 sixteen and in 1994 there were twenty-
nine. (In 2006 there were thirty-seven.) Yet the delays increased. The average
time taken from setting down to judgment had risen to an average of 8.4 months
from an average between 1985 and 1994 of 7.3 months. It was clear, he sug-
gested, that the only alternative to increasing the number of judges was to
reduce the time taken to resolve appeals. ˜That can only be done by reversing
the traditional practice of allowing counsel to state how long they want and sub-
stituting a system whereby the court stipulates the length of time for which
counsel shall be permitted to address the court™.
In 1986 the Commercial Court introduced a table of the periods for which
particular kinds of oral application would be allowed to last, unless counsel had
previously obtained permission to take more time.135 This, Lord Justice Leggatt
said, worked well and more time was only rarely sought. It was the experience
of commercial judges that ˜competent counsel can on demand tailor their sub-
missions to take no longer than a stipulated period of time, however short™:
Not only can counsel adapt to the time available, but unless the curtailment is
too drastic, the quality of the argument will almost always be improved. Increase
in the intensity of oral argument may reasonably be expected to increase its
quality. The best counsel are invariably concise; lesser counsel would usually be

˜The Future of the Oral Tradition in the Court of Appeal™, 14 Civil Justice Quarterly ,1995,
p. 11.
Admiralty and Commercial Courts Guide, F10. Civil Procedure, 2006, vol. 2, 2A-93.
685 The grounds of appeal

better if they were so. That they are not concise is mainly due to lack of the dis-
cipline that limitations of time impose.136
Lord Justice Leggatt said that, although he had no statistics on the matter, it was
comparatively uncommon for members of the court to change their minds
about whether to allow or dismiss an appeal once they had read the skeleton
arguments. There was no reason to suppose that the judges would change their
minds less often if speeches were shorter.
In ordinary cases the appellant™s solicitors must lodge an estimate of time
needed for the hearing, signed by counsel. A copy must be sent to the respon-
dent who then has the opportunity of disagreeing the time estimate. Failure to
do so is taken as acceptance of the proposed time limit. Any revised time esti-
mate must be lodged with the court, signed by the advocate concerned.
Since 1991 the House of Lords too has required that counsel should notify
the Judicial O¬ce how many hours were needed for argument and broadly
expects them to keep within that estimate.137
The Bowman Committee said in its 1997 report that although it did not
favour the drastic American approach to time limits for oral argument, it did
think that ˜there is a greater need to impose appropriate time limits for
individual appeals™.138 But it did not wish ˜to see counsel being prevented
from making relevant submissions because they are abruptly cut o¬ in mid-

3. Appeal decisions

The grounds of appeal
An appeal can be brought on a variety of grounds. In a civil case it can be on
fact or law, on the amount of damages, on the wrong exercise by the trial court
of a discretion or an allegation that the court exceeded its jurisdiction.
In a criminal case the appeal can be against conviction or sentence. If the
appeal is against conviction, it can be either on the facts (that the court or the
jury reached the wrong result), on a point of law, on a question of mixed fact
and law or on any other ground which appears su¬cient (Criminal Appeal Act
1968, s. 1(2)(b)).
Mistakes of counsel Incompetent representation by counsel at the trial is not
in itself a ground of appeal140 though if the advocate was ¬‚agrantly incompe-

Note 134 above at p. 15.
Procedure Direction [1991] 3 All ER 608. See also the Patent Court Practice Direction [1998]
3 All ER 372 which required the parties to give estimates of time needed for the trial and for
the judge to read the papers before the hearing. Parties were reminded of ˜the court™s power to
impose guillotines on the duration of submissions and cross-examination™. This power, the
Practice Direction warned, ˜will be exercised in any case where it is of the view that a case is
not being conducted with reasonable expedition™ (para. 21). Bowman, p. 88.
139 140
Ibid, p. 99. Gautam [1988] Crim LR 109; Day [2003] EWCA Crim 1060.
686 Appeals

tent it used to be said that that might be a ground of appeal.141 However, in 1993
in Clinton142 the Court of Appeal took a di¬erent approach. It held that the test
was whether the conviction was safe. If counsel™s conduct rendered the verdict
unsafe (or unsatisfactory)143 the court would not seek to assess the qualitative
value of counsel™s alleged incompetence but would seek to assess its e¬ect on the
trial and the verdict.144
The Runciman Royal Commission (1993) recommended that the Court of
Appeal™s attitude to errors by counsel be based (as suggested in Clinton) by its
e¬ect rather than on the degree of incompetence. (˜It cannot possibly be right
that there should be defendants serving prison sentences for no other reason
than that their lawyers made a decision which later turns out to have been mis-
taken. What matters is not the degree to which the lawyers were at fault but
whether the particular decision, whether reasonable or unreasonable, caused a
miscarriage of justice™.145)
This now represents the Court of Appeal™s basic approach,146 though
mention of ˜¬‚agrant incompetence™ still occur.147
The approach adopted by the Strasbourg Court is based on the question “
was the defendant deprived of his right to a fair trial?148 In Thakrar149 the Court
of Appeal said: ˜The test is whether in all the circumstances, the conviction is
safe. Nonetheless, if such failures have prevented an appellant from having a fair
trial, within the meaning of Article 6 of the European Convention on Human
Rights, that will normally mean that the conviction is unsafe and should be

Only one appeal
In Pinfold150 it was held that an appellant only has a right to appeal once. The
court had no jurisdiction to hear a second appeal, even on the grounds of fresh
evidence. The only recourse for the defendant then was to ask the Home
Secretary to refer the case back to the Court of Appeal under his powers under
s. 17 of the Criminal Appeal Act and is now to try to get the Criminal Cases
Review Commission (CCRC) (see pp. 725“29 below) to do so.

Ensor [1989] 1 WLR 497; Crabtree, Foley, McCann [1992] Crim LR 65.
[1993] 2 All ER 998.
As will be seen below, the statutory formula ˜unsafe or unsatisfactory™ was changed to ˜unsafe™.
Followed in Boal [1992] 3 All ER 177; Irwin [1987] 2 All ER 1085; Ahluwalia [1992] 4 All ER
889. Runciman, p. 174, para. 59.
See for instance Thakrar [2001] EWCA Crim 1096; Harrison [2002] EWCA Crim 2309; Day
[2003] EWCA Crim 1060.
Boodram [2001] UKPC 20, [2002] 1 Cr App Rep 103. The Judicial Committee quashed a
conviction on account of counsel™s ˜multiple failures™ “ ˜the worst case of the failure of counsel
to carry out his duties in a criminal case that their Lordships have come across™ (at [39]). The
barrister who appeared at the retrial had for instance not been aware that it was a retrial and
when he became aware he did not inquire as to what had happened at the ¬rst trial. There was
no need to embark on an inquiry as to the impact of the failures. The defendant had not had a
fair trial. See the commentary on Nangle [2002] Crim LR 506 at 507.
149 150
[2001] EWCA Crim 1096, para. 35. [1988] 2 All ER 217.
687 The grounds for allowing appeals

It has been held that when the CCRC refers a case back to the Court of
Appeal, the court is not bound by the decision in Pinfold and in exceptional
circumstances it can therefore reconsider an issue that it has previously deter-

Powers of the Court of Appeal
Court of Appeal Civil Division
The Court of Appeal can make any order which could have been made in the
court below and substitute its own decision as to liability, quantum of damages
or costs. It is not limited to points raised in the notice of appeal. It can, though it
rarely does, take further points itself, for instance, as to the illegality of a contract.
The court can order a retrial. However, where the court was considering an
award of damages by a jury, it had no power to substitute its own award for that
of the jury unless the parties consented (which generally they did). Absent such
consent, it had to order a retrial. The Courts and Legal Services Act 1990, s. 8
gave a power for rules to be made to permit the court to change the amount of
damages.152 Usually, the Court of Appeal intervenes to reduce damages but it
can increase them.153

Court of Appeal Criminal Division
The Court of Appeal Criminal Division can quash a conviction or reduce a sen-
tence. Since 1966 it has not had the power to increase sentences, though, curiously,
this power is still exercisable by the Crown Court when it hears appeals from the
magistrates™ courts. The Court of Appeal also has a right to order a retrial.

The grounds for allowing appeals
Civil cases
Under the former system there was no rule in either statute or the rules of court as
to the grounds for allowing an appeal. The Rules of the Supreme Court simply said
that the Court of Appeal ˜shall have power to draw inferences of fact and to give
any judgment and make any order which ought to have been given or made, and
to make such further or other order as the case may require™.154 Case law and com-
mentaries, such as the White Book for the High Court and the equivalent Green
Book for the county court, established the principles on which the courts acted.

Thomas [2002] EWCA Crim 941, [2002] Crim LR 912; Wallace Duncan Smith (No 3) [2002]
EWCA Crim 2907, [2003] 1 Cr App Rep 648; Mills (No 2), Poole (No 2) (2003) Times, 26 June.
See RSC (Amendment No 3) 1990, SI 1990/2599, reg. 13, amending CPR Sch. 1, RSC Order
59, r. 11.
For a rare example see Clark v. Chief Constable of Cleveland Constabulary (1999) 21 LS Gaz 38
when the court increased an award of damages for malicious prosecution from £500 to
£2,000. RSC Order 59, r. 10(3).
688 Appeals

The rules on the hearing of appeals now provide that the appeal court will
allow an appeal ˜where the decision of the lower court was (a) wrong; or (b)
unjust because of a serious procedural or other irregularity in the proceedings
in the lower court™ (CPR 52.11(3)). These rules apply not just to the Court of
Appeal; they apply to all civil courts exercising appellate functions.
The editors of the White Book suggest that ˜wrong™ presumably means that the
court below erred in law, erred in fact or erred in the exercise of its discretion.
As regards errors of fact, the Court of Appeal has always been chary of taking
a di¬erent view of the facts from that taken by the trial court, especially where
the ¬ndings of fact were based on testimony given by witnesses. When the deci-
sion was that of a jury the reluctance was even greater. The position was
described by the House of Lords in a case in 1927.

SS Hontestroom (Owners) v. SS Sagaporack (Owners) [1927] AC 37, HL
[In actions arising out of a collision between two ships the trial judge found that
the Sagaporack was wholly to blame. His decision was reversed by the Court of
Appeal which found the other ship was wholly to blame.
On appeal to the House of Lords, Lord Sumner, giving the judgment for the
majority, said:]
The learned President, after seeing both pilots, accepted the story of the
Hontestroom. Though he does not expressly say so, it is evident that he regarded
the Hontestroom™s pilot as an honest and a credible witness and, conversely, that
he did not accept the story of the pilot of the Sagaporack, not thinking that his
memory could be trusted . . .
What then is the real e¬ect on the hearing in a Court of Appeal of the fact that
the trial judge saw and heard the witnesses? I think it has been somewhat lost sight
of. Of course, there is jurisdiction to retry the case on the shorthand note, includ-
ing in such retrial the appreciation of the relative values of the witnesses, for the
appeal is made a rehearing by rules which have the force of statute: Order 68, r. 1.
It is not, however, a mere matter of discretion to remember and take account of this
fact; it is a matter of justice and of judicial obligation. None the less, not to have
seen the witnesses puts appellate judges in a permanent position of disadvantage
as against the trial judge, and, unless it can be shown that he has failed to use or has
palpably misused his advantage, the higher court ought not to take the responsi-
bility of reversing conclusions so arrived at, merely on the result of their own com-
parisons and criticisms of the witnesses and of their own view of the probabilities
of the case. The course of the trial and the whole substance of the judgment must
be looked at, and the matter does not depend on the question whether a witness has
been cross-examined to credit or has been pronounced by the judge in terms to be
unworthy of it. If his estimate of the man forms any substantial part of his reasons
for his judgment the trial judge™s conclusion of fact should, as I understand the
decisions, be let alone. In The Julia (1860) 14 Moo PC 210 at 235 Lord Kingsdown
says: ˜they, who require this Board, under such circumstances, to reverse a decision
of the court below upon a point of this description, undertake a task of great and
almost insuperable di¬culty . . . We must, in order to reverse, not merely entertain
doubts whether the decision below is right, but be convinced that it is wrong™. . ..
689 The grounds for allowing appeals

My Lords, for these reasons I do not propose to retry this case, nor do I think
that the Court of Appeal should have done so.
For rarely expressed scepticism about the value to trial courts of observing the
demeanour of the witnesses, see, however, a lecture given by Sir Thomas
Bingham (as he then was) given at an early stage of his illustrious judicial
career “ ˜The Judge as Juror™, Current Legal Problems, 1985, p. 1 at 6“13.
Sometimes, the court would reverse a judge™s ¬nding of fact on the ground
that it was plainly wrong155 and, very exceptionally, the Court of Appeal was
prepared to reverse even a jury™s decision if it found it to be perverse. An
example was Grobbelaar v. News Group Newspapers Ltd.156 The Court of Appeal
set aside a jury™s award of £85,000 libel damages awarded to Bruce Grobbelaar,
the famous goalkeeper. He sued the Sun newspaper which, in a series of sensa-
tional articles published over seven days, had accused him of taking bribes to ¬x
games. The Court of Appeal said it had a duty to intervene where the verdict
was so plainly wrong that no jury acting reasonably could have reached such a
decision on a balance of probabilities. Having regard to the evidence,
Grobbelaar™s story was simply incredible and he should not be permitted to
retain an unmerited award of damages.157 On appeal, the House of Lords
reversed the Court of Appeal™s decision on the ground that although the Court
of Appeal could and should quash a perverse jury decision, it did not agree that
the jury™s decision was perverse. (Any satisfaction Mr Grobbelaar might have
taken in the Law Lords™ decision will have been considerably diminished by their
decision to reduce his award of damages to a nominal £1 and to order that he
pay two-thirds of the Sun newspaper™s costs.158)
The position was somewhat di¬erent when the appeal court was asked to
review the drawing of inferences from facts by the trial judge. In such cases the
appeal court regarded itself as permitted to draw di¬erent inferences even
though it had not seen the witnesses (Benmax v. Austin Motor Co Ltd159), but in
Biogen Inc v. Medeva Plc160 Lord Ho¬mann warned against treating Benmax as
authorising an appellate court to undertake a fresh evaluation of the evidence
where there was no question of the credibility of witnesses. The need for judi-
cial caution in reversing the judge™s evaluation of the fact was based on much
more than professional courtesy:

See for instance The Ikarian Reefer [1995] Lloyd™s Rep 455 reversing the ¬nding that insured
shipowners had not intentionally scuttled their vessel.
[2001] EWCA Civ 33, [2001] 2 All ER 437.
For critical commentary on the decision see P. Robertshaw, ˜The Review Roles of the Court of
Appeal: Grobbelaar v. News International™, 64 Modern Law Review, 2001, pp. 923“32.
[2002] UKHL 40, [2002] 4 All ER 732. The Law Lords said that although the jury must have
accepted that G had corruptly accepted bribes, it must also have found that he did not in fact
˜throw™ matches. To that extent, on the evidence, the jury™s decision was not perverse, but the
Law Lords held that it would be an a¬ront to justice if a court of law were to award substantial
damages to a man shown to have acted in such ¬‚agrant breach of his legal and moral
obligations. [1955] AC 370 or Whitehouse v. Jordan [1981] 1 All ER 267.
[1997] RPC 1 at 45.
690 Appeals

It is because speci¬c ¬ndings of fact, even by the most meticulous judge, are
inherently an incomplete statement of the impression which was made upon
him by the primary evidence. The expressed ¬ndings are always surrounded by
a penumbra of imprecision as to emphasis, relative weight, minor quali¬cation
and nuance . . . of which time and language do not permit exact expression but
which play an important part in the judge™s overall evaluation.
An appellate court, Lord Ho¬mann said, should be very cautious in di¬ering
from the judge™s evaluation.
The Court of Appeal™s attitude to altering awards of damages was similar. It
was more reluctant to interfere with an award by a jury than a judge though it
would alter an award even of a jury if it thought it to be wholly wrong,161 but
the court would uphold an award of damages even if it thought it was consid-
erably more than it would itself have awarded.162
With regard to review of discretionary decisions, the classic rule was that the
appeal court would not interfere, even if it disagreed with the decision, unless
it could be shown that the judge below had erred in law or had acted on wrong
principles “ such as taking into account irrelevant matters, acting under a mis-
apprehension of fact or failing to exercise the discretion.163

The position since the introduction of the CPR is much the same. In Designers
Guild Ltd v. Russell Williams (Textiles) Ltd164 the House of Lords held that the
Court of Appeal had been wrong to substitute its own assessment of the evi-
dence for that made by the trial judge. The question at issue was whether there
had been infringement of copyright. Lord Bingham said that the Court of
Appeal had approached this issue ˜more in the manner of a ¬rst instance court
making original ¬ndings of fact than as an appellate court reviewing ¬ndings
already made . . . It was not for the Court of Appeal to embark on the issue of
substantiality afresh, unless the judge had misdirected himself, which in my
opinion he had not™.165 Lord Ho¬mann said that although the issue had not
involved assessment of the credibility of witnesses, nevertheless the trial court
had had the bene¬t of expert testimony. The court™s decision involved the
application of a not altogether precise legal standard to a combination of fea-
tures of varying importance. The case fell into a class of case in which an

For examples see Lewis v. Daily Telegraph Ltd [1964] AC 234 (damages manifestly too high) or
English and Scottish Co-op Properties Mortgage and Investment Society Ltd v. Odhams Press Ltd
[1940] 1 KB 440 (damages manifestly too low).
In Blackshaw v. Lord [1983] 2 All ER 311 the court refused to interfere with libel
damages of £45,000 awarded by a jury even though all three judges thought it was far too high.
See for instance Culver v. Beard [1937] 1 All ER 301 (court refused to change allocation of case
from county court to High Court); Stevens v. Walker [1936] 2 KB 215 (court interfered
because the judge had not considered matters he should have taken into account); Eagil Trust
Co Ltd v. Pigott-Brown [1985] 3 All ER 119 (dismissal for want of prosecution “ court™s role is
to review the exercise of discretion, not to substitute its own decision).
164 165
[2001] 1 All ER 700. At 702.
691 The grounds for allowing appeals

appellate court should not reverse a judge™s decision unless he has erred in
In Assicurazioni Generali SpA v. Arab Insurance Group (BSC)166 Lord Justice
Ward said that two factors led appellate judges to be cautious about interfering:
First, the appellate court recognises that judging the witness is a more complex
task than merely judging the transcript. Each may have its intellectual compo-
nent but the former can also crucially rely on intuition. That gives the trial judge
the advantage over us in assessing the witness™s demeanour, so often a vital factor
in deciding where the truth lies. Secondly, judging is an art not a science. So the
more complex the question, the more likely it is that di¬erent judges will come
to di¬erent conclusions and the harder it is to determine right from wrong.
Borrowing language from other jurisprudence, the trial judge is entitled to ˜a
margin of appreciation™ (at [196]).
This is familiar language which could equally have come from the pre-CPR era.
As to what constitutes su¬cient error in the exercise of discretion to justify
interference by the appeal court, in Tanfern Ltd (n. 16 above) Lord Justice
Brooke referred to Lord Fraser™s speech in G v. G (Minors: Custody Appeals):167
The appellate court should only interfere when they consider that the judge of
¬rst instance has not merely preferred an imperfect solution which is di¬erent
from an alternative imperfect solution which the Court of Appeal might or
would have adopted, but has exceeded the generous ambit within which a rea-
sonable disagreement is possible.168
Again, that represents business as usual.

Criminal cases
The conditions for the court to allow an appeal were ¬rst laid down in s. 4 of
the Criminal Appeal Act 1907. (As will be seen, this was replaced by s. 2 of the
Criminal Appeal Act 1966, which became s. 2 of the Criminal Appeal Act 1968,
which in its turn was replaced by s. 2 of the Criminal Appeal Act 1995.)

Criminal Appeal Act 1907
4.“(1) The Court of Criminal Appeal on any such appeal against conviction shall
allow the appeal if they think that the verdict of the jury should be set aside on
the ground that it is unreasonable or cannot be supported having regard to the
evidence, or that the judgment of the court before whom the appellant was con-
victed should be set aside on the ground of a wrong decision of any question of
law or that on any ground there was a miscarriage of justice, and in any other
case shall dismiss the appeal:
Provided that the court may, notwithstanding that they are of opinion that
the point raised in the appeal might be decided in favour of the appellant,
dismiss the appeal if they consider that no substantial miscarriage of justice has
actually occurred.

166 167 168
[2002] EWCA Civ 1642. [1985] 1 WLR 647. At 652.
692 Appeals

The Court of Appeal™s attitude to jury verdicts in criminal cases was, if anything,
even more deferential than that it adopted in civil cases. The case that follows
was typical:

R v. Hopkins-Husson (1949) 34 Cr App Rep 47, CA
[Lord Chief Justice Goddard, giving the judgment of the court, said:]
With regard to the other six cases, the jury found a verdict of not guilty in ¬ve
of them, and in the case of one boy, a boy called Allan Simpson, they found the
appellant guilty. It is fair and right to say that the learned judge said in terms that
he was surprised at the verdict, and he himself would obviously have preferred
a verdict of acquittal; but it is also right to say that from a very early period in
the history of this court it has been laid down, and has been laid down frequently
since, that the fact that the trial judge was dissatis¬ed with the verdict, although
it is a matter to be taken into account in this court, must not be taken as a ground
by itself for quashing the conviction. If it were, it would mean that we should be
substituting the opinion of the judge for the opinion of the jury, and that is one
of the things which this court will never do.
In just the same way it has been held from an equally early period in the
history of this court that the fact that some members or all the members of the
court think that they themselves would have returned a di¬erent verdict is again
no ground for refusing to accept the verdict of the jury, which is the constitu-
tional method of trial in this country. If there is evidence to go to the jury, and
there has been no misdirection, and it cannot be said that the verdict is one
which a reasonable jury could not arrive at, this court will not set aside the
verdict of guilty which has been found by the jury.
A commentator in 1966, describing the attitude of the court to its powers,
The broad picture that emerged was a court concerned in appeals against con-
viction, with the judge™s direction, evidence and procedure and the occasional
point of substantive law rather than the ˜merits™ of the case. An appellant who
could point to a clear misdirection, the wrongful admission or exclusion of evi-
dence or some procedural irregularity, had better prospects of success than the
appellant who simply claimed that he was innocent and that the jury had come
to the wrong decision.169
A JUSTICE Committee in 1964 thought ˜it seems absurd and unjust that ver-
dicts which experienced judges would have thought surprising and not sup-
ported by really adequate evidence, should be allowed to stand for no other
reason than that they were arrived at by a jury™.170 In 1965 the Donovan
Committee took a similar view:
Under the terms of s. 4(1), if it is strictly construed, there is, in the case of an
innocent person who has been wrongly identi¬ed and in consequence wrongly

M. Dean, ˜Criminal Appeal Act 1966™, Criminal Law Review, 1966, pp. 535 and 539.
JUSTICE, Criminal Appeals, 1964, para. 59.
693 The grounds for allowing appeals

convicted, virtually no protection conferred by his right to appeal . . . provided
that the evidence of identi¬cation was, on the face of it, credible. We think that
this defect should be remedied.171
It recommended the adoption of a broader formula (one originally proposed in
1907 by F.E. Smith, the later Lord Birkenhead, Lord Chancellor, during the
debates on the Criminal Appeal Bill), that the court should quash a conviction
where the verdict in the opinion of the court was ˜under all the circumstances
of the case unsafe or unsatisfactory™. This was duly achieved in the Criminal
Appeal Act 1966, which was then incorporated into the 1968 Act and became
s. 2(1)(a) of that Act.

Criminal Appeal Act 1968
2.“(1) Except as provided by this Act, the Court of Appeal shall allow an appeal
against conviction if they think:
(a) that the verdict of the jury should be set aside on the ground that under all
the circumstances of the case it is unsafe or unsatisfactory; or
(b) that the judgment of the court of trial should be set aside on the ground of
a wrong decision of any question of law; or
(c) that there was a material irregularity in the course of the trial, and in any
other case shall dismiss the appeal:
Provided that the court may, notwithstanding that they are of opinion that the
point raised in the appeal might be decided in favour of the appellant, dismiss
the appeal if they consider that no miscarriage of justice has actually occurred.
In the case of an appeal against conviction the court shall, if they allow the
appeal, quash the conviction.
In the ¬nal appeal of the Birmingham Six the prosecution argued that the two
words ˜unsafe™ and ˜unsatisfactory™ in s. 2(1)(a) had separate meanings and that
therefore convictions could be unsatisfactory but not unsafe. The Court of
Appeal rejected this view. The two words, it said, were indistinguishable. (As
will be seen, this re-emerged as a question in interpretation of the 1995 Act “
see pp. 701“04 below.)

Cooper and the ˜lurking doubt™ test
In 1969 the Court of Appeal Criminal Division decided the Cooper case, in
which it pronounced a philosophy with regard to the way in which the court
should approach jury verdicts that was very di¬erent from the approach shown
in the Hopkins-Husson decision.

R v. Cooper [1969] 1 QB 267, CA
[The defendant was convicted of assault occasioning actual bodily harm after an
incident in which a twenty-two year old girl was attacked by one of a group of

Report of the Interdepartmental Committee on the Court of Criminal Appeal, 1965, Cmnd.
2755, p. 33, para. 145.
694 Appeals

three drunken youths. At an identi¬cation parade six weeks after the o¬ence she
picked out the defendant. In his own words: ˜She never looked at anyone else™
and according to the court she clearly had no doubt at all. The question for the
court was whether the conviction was unsafe by reason of the evidence at the
trial that B had told D that he rather than the defendant had committed the
attack. There was close physical similarity between the defendant and B.
Nevertheless the jury convicted.
Lord Justice Widgery, giving the judgment of the court, said:]
It has been said over and over again throughout the years that this court must
recognise the advantage which a jury has in seeing and hearing the witness, and
if all the material was before the jury and the summing-up was impeccable, this
court should not lightly interfere. Indeed, until the passing of the Criminal
Appeal Act 1966, provisions which are now to be found in s. 2 of the Criminal
Appeal Act 1968, it was almost unheard of for this court to interfere in such a
However, now our powers are somewhat di¬erent, and we are indeed charged
to allow an appeal against conviction if we think that the verdict of the jury
should be set aside on the ground that under all the circumstances of the case it
is unsafe or unsatisfactory. That means that in cases of this kind the court must
in the end ask itself a subjective question, whether we are content to let the
matter stand as it is, or whether there is not some lurking doubt in our minds
which makes us wonder whether an injustice has been done. This is a reaction
which may not be based strictly on the evidence as such; it is a reaction which
can be produced by the general feel of the case as the court experiences it.
The court said that after due consideration it had decided that the conviction
should be quashed as unsafe.
If the very broad ˜lurking doubt™ test as formulated in Cooper re¬‚ected the
Court of Appeal™s normal attitude, a high proportion of appellants against con-
viction might stand a reasonable chance of getting their convictions over-
turned. In fact, however, the Court of Appeal was not easily persuaded to adopt
the ˜lurking doubt™ test. One expert stated in 1983: ˜the “lurking doubt” test,
enunciated by Lord Widgery when he was ¬rst appointed, has been quietly
buried™.172 Research carried out for JUSTICE almost twenty years after the deci-
sion stated that only six reported cases had been found where the court had
quashed a conviction on the grounds that there was a lurking doubt about the
conviction and there was nothing new to throw doubt on it.
An important insight into the Court of Appeal™s marked reluctance to use
the ˜lurking doubt™ test was supplied by former Lord Justice Lawton, a vastly
experienced criminal appeal judge, in his evidence to the Runciman Royal
Until the decision of the Court of Appeal in R v. Cooper it had been assumed that

T. Sargant, More Law Reform Now (Barry Rose, 1983) p. 91. Sargant was the Director of
695 The grounds for allowing appeals

a conviction should not be quashed unless there was some reason in law for
doing so. In that case however it was adjudged that the court could apply a sub-
jective test had it a lurking doubt or reasoned unease which made it wonder
whether an injustice had been done. In simpler terms this means that the court
can quash a conviction if it has a hunch that there has been an injustice. This
cannot be a sound way of administering criminal justice; and since 1969 the
judges seem to have appreciated that it was not because only six appeals have
been allowed on this ground.
In other words, the judges did not apply the ˜lurking doubt™ test because they
did not like it. They believed that to apply it would be to usurp the function of
the jury.

Report of the Runciman Royal Commission
The Runciman Royal Commission said it had received con¬‚icting evidence
about the ˜lurking doubt™ test. On the one hand, there were those who pointed
out that the Court of Appeal had only very rarely acknowledged that it was
applying this test. On the other hand, it had been suggested to the Royal
Commission that the Court of Appeal had not infrequently allowed appeals on
what had in truth been the ˜lurking doubt™ principle, even though there had
been no reference to the phrase. These were cases where there was no error at
the trial nor any error in law, ˜but nevertheless the combined experience of the
three members of the court leads them to conclude that there may have been an
injustice in the trial and in the jury™s verdict™. They consequently allowed the
appeal on the ground that, at the least, the jury™s verdict was unsatisfactory.
˜There is no real di¬erence between this approach and an application of the
“lurking doubt” principle™.173
The Royal Commission™s conclusion on the matter was to encourage the
court to use this power when it felt it right to do so:
We fully appreciate the reluctance felt by judges sitting in the Court of Appeal
about quashing a jury™s verdict. The jury has seen the witnesses and heard
their evidence; the Court of Appeal has not. Where, however, on reading the
transcript and hearing argument the Court of Appeal has a serious doubt
about the verdict, it should exercise its power to quash. We do not think that
quashing the jury™s verdict where the court believes it to be unsafe undermines
the system of jury trial. We therefore recommend that, as part of the redraft-
ing of s. 2, it be made clear that the Court of Appeal should quash a conviction
notwithstanding that the jury reached their verdict having heard all the rele-
vant evidence and without any error of law or material irregularity having
occurred if after reviewing the case, the court concludes that the verdict is or
may be unsafe.174
For an example of the Court of Appeal adopting this approach see R v.
Haughton.175 The case was referred back to the Court of Appeal by the Home
173 174
Runciman, p. 171, para. 45. Runciman, pp. 171“2, para. 46.
Unreported, Case No. 589/SI/91, 21 May 1992.
696 Appeals

Secretary on the ground that the ESDA test appeared to show that police o¬cers
had fabricated the appellant™s confession. The Court of Appeal rejected that
argument, but it said that its duty was ˜to review the case generally™. Having
done that, it found that the verdict was unsafe and unsatisfactory even though
there was nothing new that had not been before the jury.
As will be seen below, the Government accepted the Royal Commission™s pro-
posal that s. 2 be redrafted but it did not adopt the proposal in the form sug-
gested by the Commission. The status of the ˜lurking doubt™ test was uncertain.

Do appeal judges have the time it takes?
A practical point made in a powerful lecture on the problem of miscarriages of
justice by the distinguished Australian judge, Justice Michael Kirby, is that
appeal judges do not have the time to consider the trial evidence properly. Nor,
typically, do they have the time, all of them, to read the entirety of the transcript
of what may have been a trial lasting many days or even weeks. ˜They visit the
evidence, on the invitation of counsel, skipping from one passage to another.
Rarely do they capture the subtle atmosphere of the trial, for such things do not
readily emerge from cold pages. These are the reasons why so much deference
is paid to the advantages of the trial judge or jury, who see the evidence unfold
in sequence and observe the witnesses giving their testimony.™176
Lord Justice Auld™s report painted a picture of a court that was under enor-
mous pressure:
It is no secret that the judge allotted the task of giving the judgment of the court
in each case will often need to prepare in advance some provisional notes of the
relevant facts, issues and law as a reference for his judgment. The volume and
speed of the work is such that the judges could not cope if they did not do
that . . . Working at such speed gives the judges of the court little time to focus
on anything but the application of the law to the particular facts before them . . .
It is thus di¬cult for them to apply and develop the law in a principled and con-
sistent manner . . . This is a serious shortcoming in the main judicial institution
in this country responsible for declaring and developing the criminal law as well
as for applying it.177

Quashing the jury™s verdict on account of error at trial
Research by Professor Kate Malleson shows that by far the most frequent reason
for the Court of Appeal to quash a conviction is because of some error at trial,
usually error by the trial judge in the form of misdirection of the jury on the law

˜Miscarriages of Justice™, The Child and Co Lecture, London, 1991, p. 26. The evidence of the
Lord Chancellor™s Department to the Runciman Royal Commission stated that in a typical
week in September 1991, the one Division of the Court of Appeal Criminal Division that was
then sitting was provided with 4,800 pages of documentation. About half the Court of
Appeal™s time was spent on sentence appeals. Such appeals were declining while the
proportion of appeals against conviction was increasing. In a normal sitting day, each
Division of the court could deal with up to ten sentence cases and one or two conviction
cases. Auld, Ch. 12, pp. 643“4, para. 000.
697 The grounds for allowing appeals

or some other defect in the summing up, or a wrong decision to allow or to
exclude evidence.178
Under the Criminal Appeal Act 1968, errors at trial could be dealt with in
three ways. One was to treat the error as inconsequential by applying ˜the
proviso™ (see pp. 699“700 below). The second was to quash the conviction and
order a retrial (see pp. 714“15 below). The third and, according to Malleson™s
research, by far the most common was to quash the conviction.
A majority of the Runciman Royal Commission proposed a di¬erent approach:
(1) If the court believes that the conviction is safe despite the error, the appeal
should be dismissed.
(2) If the court believes the error has rendered the verdict unsafe, the appeal
should be allowed and the conviction quashed.
(3) If it believes the conviction may be unsafe as a result of the error, it should
quash the conviction and order a retrial.179
Three of the Commission™s members wished to add a further category for cases
where there is an error at trial su¬ciently serious to a¬ect the trial materially
but not su¬ciently serious to make the conviction unsafe. In such a case they
thought the court should order a retrial. The majority disagreed: ˜The majority
of us do not believe that a person who is clearly guilty should be accorded a
retrial merely because there has been some error at the trial™.180
As will be seen, the minority™s approach eventually prevailed in the judicial
interpretation of the new statutory formula but in 2006 the Government sig-
nalled that it wanted the majority™s view to be adopted.

Quashing the jury™s verdict on account of pre-trial malpractice or procedural
Where an appeal is based on some pre-trial matter (which might be anything
from fabrication of evidence to some serious irregularity in the implementation
of PACE) a majority of nine out of eleven members of the Runciman Royal
Commission thought the Court of Appeal should only act if it thought the
matter was such as to make, or maybe make, the conviction unsafe. If there was
plenty of other, untainted evidence showing the defendant to be guilty, his con-
viction should not be quashed even if there were some gross impropriety in the
pre-trial handling of the case. The minority of two (which included the writer)
thought that there could be occasions when the court should quash a convic-
tion even though there was clear evidence of guilt.

The majority view
49. In the view of the majority, even if they believed that quashing the convictions
of criminals was an appropriate way of punishing police malpractice, it would be
In the 1990 and 1992 samples in her research for the Runciman Royal Commission, trial
errors were involved in 83 per cent and 82 per cent of successful appeals. (K. Malleson, Review
of the Appeal Process, Royal Commission on Criminal Justice, Research Report No. 17, 1993,
179 180
p. 22, Table 1.2.) Auld, p. 170, para. 38. Ibid.
698 Appeals

naive to suppose that this would have any practical e¬ect on police behaviour. In
any case it cannot in their view be morally right that a person who has been con-
victed on abundant other evidence and may be a danger to the public should walk
free because of what may be a criminal o¬ence by someone else. Such an o¬ence
should be separately prosecuted within the system. It is also essential, if con¬-
dence in the criminal justice system is to be maintained, that police o¬cers
involved in malpractice should be disciplined, and in this connection we attach
great importance to the recommendations in chapter three, which should lead to
more e¬ective police disciplinary procedures. The Court of Appeal must report
any cases of malpractice by police o¬cers which come to their attention to chief
o¬cers of police. We also envisage that the more serious the malpractice the less
likely it is that the court would conclude that the verdict could be safe.
50. In the view of the majority, the minority view is illogical. It would only be
e¬ective if the judge at ¬rst instance had allowed the tainted evidence to be heard
by the jury. If the judge had properly excluded the evidence then the verdict
would be unassailable. The minority view must logically involve the trial judge
in stopping a case on the basis of tainted evidence which he or she nevertheless
proposed to exclude. The majority believe this to be unacceptable precluding as
it must the jury from returning a verdict on the basis of evidence which was safe,
admissible, and probative. It is only the tainted evidence which is excluded by
s. 78 of PACE. That section does not allow the court to stop the case if there
remains admissible probative evidence to support it [p. 23].

The minority view
[The minority view was expressed in the writer™s dissent:]

68. The moral foundation of the criminal justice system requires that if the pros-
ecution has employed foul means, the defendant must go free even though he is
plainly guilty. Where the integrity of the process is fatally ¬‚awed, the conviction
should be quashed as an expression of the system™s repugnance at the methods
used by those acting for the prosecution.
69. The majority™s position would I believe encourage serious wrongdoing
from some police o¬cers who might be tempted to exert force or fabricate or
suppress evidence in the hope of establishing the guilt of the suspect, especially
in a serious case when they believe him to be guilty. There have unfortunately
been some gross examples of such conduct.
70. The position adopted by the majority also seems to me to risk undermin-
ing the principle at the heart of s. 78 of PACE which explicitly gives the court the
power to exclude evidence on the ground that it renders the proceedings ˜unfair™.
The word ˜unfair™ expresses the underlying moral principle and the Court of
Appeal has repeatedly used this new statutory power very broadly to express its
refusal to uphold convictions based on unacceptable police practices even when
it could not be said that the misconduct had any impact on the jury™s verdict.
71. Section 78 would of course remain “ but the majority would in e¬ect be
encouraging the Court of Appeal to undercut a part of its moral force by saying
that the issue of ˜unfairness™ can be ignored where there is su¬cient evidence to
show that the defendant is actually guilty. Any judge concerned to discourage
699 The grounds for allowing appeals

prosecution malpractice would I believe be dismayed by the majority™s position.
In terms of the message sent to the police service and other prosecution agen-
cies it could undo much of the good e¬ect being achieved by the attitude of the
judges to s. 78 of PACE.
72. But the matter goes beyond discouraging prosecution malpractice. At the
heart of the criminal justice system there is a fundamental principle that the
process must itself have integrity. The majority suggest that the answer to pros-
ecution wrongdoing in the investigation of crime is to deal with the wrongdo-
ers through prosecution or disciplinary proceedings. Even were this to happen
(and often in practice it would not), the approach is not merely insu¬cient, it
is irrelevant to the point of principle. The more serious the case, the greater the
need that the system upholds the values in the name of which it claims to act. If
the behaviour of the prosecution agencies has deprived a guilty verdict of its
moral legitimacy the Court of Appeal must have a residual power to quash the
verdict no matter how strong the evidence of guilt. The integrity of the criminal
justice system is a higher objective than the conviction of any individual.181

Applying ˜the proviso™
No one suggests that a conviction should be quashed, or even a retrial ordered,
where the matter complained of by the appellant is trivial. (In the United States
this is known as ˜harmless error™.) Here the matter was previously dealt with by
what was called ˜the proviso™.
The proviso referred to here is that at the end of the Criminal Appeal Act
1968, s. 2 “ p. 693 above “ ˜Provided that the court may, notwithstanding that
they are of opinion that the point raised in the appeal might be decided in
favour of the appellant, dismiss the appeal if they consider that no miscarriage
of justice has actually occurred™.
The application of the proviso was explored by Michael Knight in his book
on criminal appeals. In this he showed that, contrary to what was often main-
tained, the great majority of cases where the power of the proviso had been
exercised were cases of serious error in the trial. Before substantiating this con-
troversial assertion, he set out the test which the court had developed for the
application of the proviso.

Michael Knight, Criminal Appeals, 1970, pp. 9“53
The test which the appellate court goes by is not the degree of error but whether
there is, despite the fault, su¬cient evidence and a su¬cient direction for a rea-
sonable jury inevitably to convict for, if so, there is no substantial miscarriage of
justice.182 However, if it is correct to say that the error can have had any crucial
in¬‚uence on a reasonable jury the conviction must be quashed, for to uphold it
then would be a miscarriage of justice. The court metaphorically blots out the
fault “ the error in the direction, the piece of inadmissible evidence, the impact
of the wrongly drafted indictment “ and asks if, without it, there is a strong

Runciman, dissent, pp. 234“5.
The test was laid down in Stirland v. DPP [1944] AC 315 (ed.).
700 Appeals

enough case for an inevitable conviction. And if they can answer ˜yes™ to this
question, they show the Nelson Touch by turning a blind eye to the fault (p. 16).
Knight gave numerous examples of cases where the proviso was applied in spite
of serious errors in the trial, e.g.:
• Haddy183 “ jury wrongly invited to infer guilt from the accused™s silence.
• Farid184 “ jury not warned by the judge that corroboration is desirable for the
evidence of accomplices.
• Whybrow185 “ misdirection as to intent in attempted murder.
• Slinger186 “ judge did not tell the jury that the onus of proof lay on the pros-
Knight also produced sixteen examples of cases where the proviso was applied
although the jury had wrongly been informed of the defendant™s previous con-
victions (pp. 19“21). He continued:
Certainly in recent years the appellate court in their judgments go extremely
carefully through the evidence other than the inadmissible evidence wrongly let
in plus the direction, or the direction minus the o¬ending portion plus the evi-
dence, to show that it is fair to say that a reasonable jury would inevitably have
convicted. This de¬nite and very often scrupulous care betrays a sense of uneasi-
ness and dislike which can be taken as further recognition of the regularity of
use “ in serious fault cases [of the proviso] (p. 21).
The line between some of the cases where the proviso has been exercised and
some where it has not is sometimes so narrow as to be almost non-existent, and
the answer to this conundrum lies in the amount of evidence and the standard
of the direction outside of the fault.
Occasionally, use of the proviso is declined because a particular fault is of its
nature so serious that, even though the appellate court would like to uphold the
conviction, and, even though there probably would be su¬cient evidence and
direction apart from the fault to justify in their opinion an inevitable ¬nding of
guilty by a reasonable jury, their desire to have a deserved conviction must be
sacri¬ced to the general principle of fairness in our criminal trial. This is the
principle stated in Maxwell v. DPP.187 It is often better that one guilty man
should escape than that the general rules evolved by the dictates of justice for the
conduct of criminal prosecutions should be disregarded or discredited . . .

The redrafting of s. 2
The Runciman Royal Commission The Royal Commission unanimously
agreed that the Criminal Appeal Act 1968, s. 2 needed to be redrafted,188 but the
Commission was not agreed as to how it should be redrafted. The majority of
eight recommended that the di¬erent grounds of appeal set out in s. 2(1)(a),

183 184
(1944) 29 Cr App Rep 182. (1945) 30 Cr App Rep 168.
185 186 187
(1951) 35 Cr App Rep 141. (1961) 46 Cr App Rep 244. [1935] AC 309.
For an article detailing the drafting defects of the section see R. Buxton, ˜Miscarriages of
Justice and the Court of Appeal™, 109 Law Quarterly Review, 1993, p. 66.
701 The grounds for allowing appeals

(b) and (c) (p. 693 above) should be replaced by a single new ground “ that the
conviction ˜is or may be unsafe™. If the court is satis¬ed that the conviction is
unsafe it should quash the conviction; if the court is satis¬ed that the convic-
tion may be unsafe it should quash the conviction and order a retrial unless
there are reasons which make a retrial impracticable or undesirable.189 Under
that scheme the proviso would be redundant.
The minority of three argued that it would be confusing to wrap up all pos-
sible grounds of appeal in the one word ˜unsafe™. That word implied that there
was something wrong with the jury™s verdict whereas the defect might be ˜some
irregularities or errors of law or procedure which did not necessarily a¬ect the
jury™s verdict but were so serious that the conviction should not stand™.190
Furthermore, in the view of the minority, an umbrella formula would not give
the Court of Appeal su¬cient guidance. In the view of the minority the grounds
of appeal should distinguish between appeals claiming that the jury reached the
wrong result and those alleging material irregularities or errors of law or pro-
cedure in or before the trial.191

The Criminal Appeal Act 1995 The Government did not accept the Royal
Commission™s recommendation that the formula should distinguish between ˜is
unsafe™ and ˜may be unsafe™. The formula in the new Act is simply whether the
conviction is unsafe. The Criminal Appeal Act 1995, s. 2 replaced the Criminal
Appeal Act 1968, s. 2 (including the proviso) with the following new provision:
˜subject to the provisions of this Act, the Court of Appeal (a) shall allow an
appeal against conviction if they think that the conviction is unsafe; and (b)
shall dismiss such an appeal in any other case™.
The Government therefore rejected the view of the minority but it also
rejected the majority™s view that the formula should include the words ˜or may
be unsafe™. The Home O¬ce minister speaking in the Committee stage of the
Bill said: ˜The di¬culty with the phrase “may be unsafe” is that it is inherently
uncertain. Almost any conviction may be unsafe. The test might well result in
the Court of Appeal having to allow a considerably greater number of appeals
than at present, simply because it did not know for certain that the conviction
was safe™.192 Also, ˜may be unsafe™ had about it a suggestion of subjectivity on the
part of someone other than the Court of Appeal. ˜That would go far broader
than current practice and far broader than the Committee would wish . . . We
do not intend it to result in fewer convictions being overturned than at present.
We want to consolidate the existing practice of the Court of Appeal™ (ibid).
The late Professor Sir John Smith, addressing this issue, basically agreed with
the Government™s view that the words ˜may be unsafe™ added nothing:
A conviction is unsafe if the court has nothing more than a lurking doubt
whether the appellant is guilty “ that is the court thinks that he may have been

189 190 191
Runciman, p. 170, para. 38. Ibid, p. 169, para. 3. Ibid.
House of Commons, Standing Committee B, 21 March 1995, col. 27.
702 Appeals

wrongly convicted. What then is the di¬erence between ˜we think that the appel-
lant may have been wrongly convicted?™ and ˜we think that it may be that he may
have been wrongly convicted?™ Surely there is no di¬erence. Either the court has
a lurking (or greater) doubt, or it does not. It is submitted that the Government
was right to insist on the exclusion of the words, ˜or may be™, which could have
led only to confusion, and possibly, to the court feeling obliged to give a narrow
meaning to ˜unsafe™.193
The Court of Appeal had previously quite often quashed a conviction on the
ground that there was an error of law or a material irregularity even though it
probably had no doubt that the defendant was guilty. If the court were to hold
that a conviction was only ˜unsafe™ if the court had a lurking (or greater) doubt
about the defendant™s conviction, that would be a drastic restriction of the
court™s power, but the parliamentary debates make it clear that this was not the
Government™s intention. In moving the Second Reading of the Bill, the Home
Secretary said of this section: ˜in substance, it restates the existing practice of the
Court of Appeal . . .™194
The Home O¬ce minister rejected an amendment to retain the words ˜or
unsatisfactory™. The Government, he said, agreed with the majority of the Royal
Commission that there was no real di¬erence between ˜unsafe™ and ˜unsatisfac-
tory™. It had been argued by some that ˜unsafe™ referred to evidential ¬‚aws whilst
˜unsatisfactory™ connoted procedural ¬‚aws, but in the Government™s view ˜the
word “unsafe” is su¬cient to deal with convictions which are unacceptable
because of ¬‚aws in the manner in which a case is prosecuted or tried, and
because of evidence which undermines the prosecution case. If a procedural
¬‚aw is su¬ciently serious to cast doubt on the safety of a conviction, the court
will allow the appeal™.195
Speaking in the Second Reading debate in the House of Lords, the Lord Chief
Justice, Lord Taylor, said the new formula “ whether the conviction is unsafe “
˜will in my view be concise, just and comprehensible to the ordinary citizen
without narrowing the present grounds of appeal™.196
On the redrafting of s. 2 see D. Schi¬ and R. Nobles, ˜Criminal Appeal Act
1995: the Semantics of Jurisdiction™, 58 Modern Law Review, 1996, pp. 299“320.
See also A. Clarke, ˜Safety or Supervision™, Criminal Law Review, 1999, p. 108;
V. Tunkel, ˜When Safe Convictions are Unsafely Quashed™, 149 New Law


. 28
( 34)