[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	The Law Commission
You are here: BAILII >> Databases >> The Law Commission >> Evidence of Bad Character in Criminal Proceedings (Report) [2001] EWLC 273(11) (October 2001) URL: http://www.bailii.org/ew/other/EWLC/2001/273(11).html Cite as: [2001] EWLC 273(11)

[New search] [Help]

 


PART XI THE INCRIMINATORY EXCEPTION

INTRODUCTION

11.1      In Part VIII we explained our recommendation that leave should be required for evidence of bad character which falls outside the central set of facts. This Part describes the second of the exceptions for evidence of the bad character of a defendant. This exception is for evidence of substantial probative value which goes to a matter in issue. It will enable the prosecution, in certain circumstances, to adduce evidence of the defendant's bad character as part of the prosecution's case.

11.2      In Part X of the consultation paper we took the view that the present law, based on the concept of similar fact, is unsatisfactory, and set out six options for consideration. We start by explaining what the responses were to those options, and how they have informed our recommendation. At the end of this Part, we address three special cases where evidence of the defendant's bad character might form part of the prosecution case.

THE OPTIONS FOR REFORM AND THE RESPONSE ON CONSULTATION

11.3      The first option considered was for no change. We rejected that option in the light of the problems we thought exist in the current law. ^[1] A substantial number of respondents addressed this, but only a small minority thought the law should be left as it is. Most of those thought that the difficulties are not attributable to the law itself but are inherent in the nature of the problem.

Option 2: adducing bad character evidence to prove mens rea^[2]

11.4      This option would allow evidence of an accused's previous convictions to be adduced where the conduct is admitted but there is an issue whether it was performed with any criminal knowledge or intent. For example, where a person with convictions for burglary is charged with having entered a building as a trespasser with intent to steal^[3] and his defence is that he entered the building only to sleep,^[4] evidence of previous convictions for burglary would be admitted, even without particular similarities between the past conduct and the charge.

11.5      We pointed out in the consultation paper that, in practice, this option would not in fact lead to the automatic admission of previous convictions for the same offence. The bare fact of the conviction would have little probative value but its prejudicial effect could easily be more than trivial, and so the defence would, no doubt, apply for the evidence to be excluded by the judge or magistrates, pursuant to the common law rule^[5] or section 78(1) of PACE. The net result would be that probative value would have to outweigh prejudicial effect.

11.6      We identified the disadvantages of this option: a significant danger of wrongful convictions; past miscarriages of justice would be compounded; defendants would be unlikely to admit the conduct if the consequence was the admission of similar previous convictions; and difficult in practice to determine in a particular case whether the conduct is admitted or denied.

11.7      Only four respondents specifically addressed this option and, of those, only one gave it any support. That respondent was concerned that evidence of bad character which tends to disprove an innocent explanation should be admissible. We reject this option. Our recommended exception would make evidence which tends to disprove an explanation given by the defendant admissible where its value in so doing outweighs the prejudice attaching to it, but without the attendant disadvantages of this option.

Option 3: adducing bad character evidence to show disposition^[6]

11.8      In the consultation paper we considered a reform proposed by the CLRC,^[7] which would allow evidence of "a disposition to commit the kind of offence with which [the defendant] is charged … if the disposition which the conduct tends to show is, in the circumstances of the case, of particular relevance to a matter in issue …".^[8]

11.9      We did not favour this option because: it would allow evidence of disposition to be given without that evidence being "highly relevant"; and it proposed (in clause 3(2)(c)) that disposition evidence be admitted where it "tends to confirm the correctness of an identification of the accused". We noted that identification evidence is recognised as giving cause for especial concern,^[9] and we were troubled by the prospect of it being supported by prejudicial evidence.

11.10      Only two respondents mentioned this option, and neither of them thought it was preferable to other options. We do not pursue this option any further.

Option 4: the Australian common law test, namely whether there is any reasonable explanation for the similar fact evidence other than that the defendant is guilty^[10]

11.11      The fourth option which we canvassed was the Australian common law test, as stated in Pfennig:^[11] the evidence is inadmissible unless there is no rational view of the evidence that is consistent with the innocence of the accused. The majority in Pfennig said that

for propensity or similar fact evidence to be admissible the objective improbability of its having some innocent explanation [must be] such that there is no reasonable^[12] view of it other than as supporting an inference that the accused is guilty of the offence charged.^[13]

This test is simple, and it minimises the risk of wrongful convictions: as Roderick Munday put it, "Practically speaking to admit similar fact evidence at all is to determine that, barring a miracle, the defendant will be convicted".^[14]

11.12      We identified three disadvantages of this option. We thought the test stricter than necessary, and likely to lead to the exclusion of evidence which was probative and of little prejudicial effect. We noted that, because it requires the judge to assess the strength of the evidence, it would require the judge to apply the same test to the evidence as the jury would have to apply, if it were admitted. Thirdly, we feared that adoption of this test would lead to an increase in the need for voir dires.

11.13      Only five respondents referred specifically to the Australian common law test. Three of them preferred it. We take the point made by one respondent who said that the Australian common law test is "clear, understandable, and appears likely to ensure that justice is done to the defendant", but, given the breadth of evidence to which this test is to apply, we do not think it is the right one. The test in Pfennig is not necessarily appropriate for bad character evidence of a different kind from that in issue in Pfennig: bad character evidence is not always either so probative, nor so prejudicial. We are still of the view expressed in the consultation paper: we think wrongful acquittals would result from setting the test so high, and we do not favour the consequences for the judge's role and the trial process which would follow from it. We note also that this option has no particular advantage in the magistrates' courts where the justices decide on the admissibility of evidence and on the question of guilt. We therefore believe that this option should be rejected.

Option 5: the scheme of the Australian Evidence Act 1995 (Cth)^[15]

11.14      The Evidence Act 1995 (Cth) provides that all relevant evidence is admissible unless excluded by the Act, but has two exclusionary rules that are relevant for present purposes: one is confined to "tendency" and "coincidence" evidence, and the other extends to any evidence adduced by the prosecution.

11.15      Under the "tendency" and "coincidence" rules, evidence may not be adduced to show

(i) that a person has a tendency to act in a particular way, or

(ii) that a person did a particular act (or had a particular state of mind) because two or more events have occurred which are unlikely to have occurred by coincidence,

unless the evidence has significant probative value.^[16] In addition, tendency evidence and coincidence evidence adduced by the prosecution about a defendant may not be used against the defendant unless its probative value substantially outweighs any prejudicial effect that it may have on the defendant.^[17]

11.16      Finally, any evidence adduced by the prosecution (not just tendency and coincidence evidence) must have probative value which outweighs the danger of prejudice.^[18]

11.17      The effect is that prejudicial evidence adduced by the prosecution is admissible only if the following conditions are met:

(i) the evidence's probative value must outweigh the danger of prejudice;^[19] and

(ii) if the evidence is tendency or coincidence evidence, its probative value must in addition ?

substantially outweigh the danger of prejudice,^[20] and ?

be significant.^[21]

11.18      Four of the five respondents who commented on this option were in support of the Australian statutory scheme. We address the points they made in the following paragraphs.

11.19      In the consultation paper we set out three drawbacks to the Australian statutory scheme.^[22] First, we were unsure what it might mean for the probative value of evidence to outweigh the risk of prejudice substantially, or for evidence to have significant probative value as opposed to some probative value. Secondly, we thought the effect of the rules was that tendency and coincidence evidence would sometimes be inadmissible even if its probative value outweighed its prejudicial effect. Thirdly, we thought the Australian statutory scheme unnecessarily complicated.

Option 6

11.20      The preferred option in the consultation paper^[23] was to allow bad character evidence to be adduced if

(i) it is relevant to a specific fact in issue; and

(ii) on the assumption that the evidence is true, the degree to which it is relevant to that fact (in other words, its probative value) outweighs the risk that, if admitted, it might

(A) result in prejudice;

(B) mislead, confuse or distract the fact-finders; or

(C) cause undue waste of time.

11.21      Of those making specific comments on this proposal, 27 respondents were in favour and ten were against it. The responses from consultees on this option and on option 5 informed our views on the following issues.

OUR CONCLUSIONS

A test of enhanced relevance

11.22      The Australian statutory scheme requires the bad character evidence itself to have "significant" probative value, whereas our preferred option did not set a standard of enhanced relevance for the probative value of the evidence. As will have been seen from our recommendation for a test of enhanced relevance for all evidence of bad character which goes outside the central set of facts,^[24] we now accept that there is a meaningful difference between evidence having some probative value and having substantial probative value. We think evidence which the prosecution wishes to adduce as part of its case against the defendant should meet this test of enhanced relevance.

The relationship between probative value and prejudicial effect

11.23      The Australian statutory scheme requires the evidence in question substantially to outweigh the danger of prejudice. Our preferred option merely required the probative value to outweigh the risk that, if admitted, it might (inter alia) result in prejudice.

11.24      We were concerned that the result of the Australian statutory scheme "must be that tendency and coincidence evidence adduced by the prosecution is sometimes inadmissible even if its probative value outweighs the risk of prejudice. … [I]f the evidence's probative value outweighs the risk of prejudice, we see no reason why it should be excluded merely because it does not substantially outweigh that risk." ^[25]

11.25      Mr Justice Tim Smith, Commissioner in Charge of the ALRC Report and a respondent to the consultation paper, expressed grave concern that the preferred option failed to give sufficient weight to the risk of wrongful convictions. He thought that the test proposed was not sufficiently strict:

In paragraph 10.84 the paper refers to what I see as the major difficulty with the proposal and that is its potential to increase the risk of wrongful conviction. It is said that this is always a risk with this kind of evidence but the view of the Commission is that this option would reduce it "as far as is reasonably practicable". I would prefer to phrase the issue slightly differently – whether the option would reduce the risk of wrongful conviction to an acceptable level. …, something would be need to be added to the discretion to ensure that evidence would only be admitted in cases where it is clear that the probative value will outweigh its likely prejudicial effect.

Ultimately, the problem is that the rule, however it is formulated, will let into evidence before the tribunal of fact evidence which carries with it a risk of wrongful conviction because of its prejudicial effect. … If a balancing test is to be used, however, one which incorporates the requirement that the probative value of the evidence substantially outweigh any probative effect would seem to me to be the minimum requirement if the concern to minimise wrongful conviction is to be adequately addressed.

11.26      The point made by Mr Justice Tim Smith is that the judge may underestimate the prejudicial effect of the evidence and, if he or she does so, then the risk of wrongful conviction is greatly increased. If, on the other hand, the evidence is only admissible if there is a significant margin between its probative value and its prejudicial effect then a minor under-estimation will not result in the admission of evidence which is in fact more prejudicial than probative.

11.27      The counter-argument is that, if there has to be a significant margin, it is possible for evidence which is more probative than prejudicial to be excluded. In the consultation paper we gave an example of a case where the higher standard would lead to the exclusion of evidence which we thought should not be excluded:

the fact-finders might already know (perhaps because the fact has been admitted as "background" evidence, or because the defendant is notorious) that the defendant has a long history of serious crime. The additional prejudice likely to result from the revelation of one more minor offence is very small, and would be outweighed by a comparatively small degree of probative value. In such a case we think that the evidence ought to be admissible, even if the probative value of the evidence is not "significant". Similarly, if the evidence's probative value outweighs the risk of prejudice, we see no reason why it should be excluded merely because it does not substantially outweigh that risk.^[26]

11.28      Because we are now proposing a test of enhanced relevance for all bad character evidence,^[27] this example is not so convincing: if the prejudice attaching to the evidence is indeed slight, then given that the character evidence must be of substantial probative value, it is likely that the probative value would substantially outweigh the prejudice.

11.29      The question is what degree of risk of wrongful conviction is acceptable. We acknowledge that there is some force the point made by Professor McEwan^[28] that people do not agree on when one fact tends to prove or disprove another, let alone about the degree to which one fact tends to prove or disprove another. Although there might be consensus about the cases at each end of the scale, so that most courts, if not all, would allow bad character evidence in one situation and would disallow it in another, there is a body of cases in the middle of the spectrum where it is likely that courts will differ on whether the probative value of the evidence in question outweighs the prejudice attaching to it. If the test is that the probative value must substantially outweigh the prejudice, then a court will exclude evidence which it might have let in after hesitation.

11.30      Our view is that such an approach runs the risk of focusing on a question which, though it might be useful to ask, is not the real issue to be addressed. The purpose of a trial is to do justice. We believe that there is no better way to express the test than in terms of this central purpose. Thus the test is best expressed in qualitative rather than quantitative terms, that is, what the interests of justice require. This test will apply in circumstances in which the prosecution is seeking to adduce evidence about the defendant. Just as the burden of proof falls on the prosecution in criminal proceedings, so this test should place the onus on the prosecution to show that the interests of justice require that the evidence be admitted.

The importance of the matter in issue

11.31      As we say in relation to non-defendants, the value of evidence is made up of its relevance to a matter in issue and the importance of that matter in issue in the context of the case as a whole. The court must therefore have regard to the significance of the issue on which the evidence in question is said to be probative in the context of the case as a whole.^[29]

The defendant's propensity to be untruthful

11.32      In the majority of contested cases, whether the defendant is telling the truth or whether the defence is true is itself an issue in the case. It is possible that evidence of bad character could have some value in indicating whether the defendant is generally untruthful. We take the view, however, that the defendant's general propensity to be untruthful is not a matter which it would be fair to allow the prosecution to assert as part of its case against the defendant. Where the defendant simply denies the truth of some or all of the prosecution's evidence in relation to the offence charged, and makes no attempt to attack anyone else's credibility, we think it virtually inconceivable that evidence of the defendant's general untruthfulness could ever have sufficient probative value to outweigh the risk of prejudice. Rather than relying on courts to find that such evidence in fact fails to meet the requirements of the incriminatory exception in any case where it is tendered, we have decided that such evidence should not even be eligible for admission under this exception.

11.33      We believe that evidence of a defendant's propensity to be untruthful should be admissible in certain limited circumstances, but only if certain further conditions are satisfied in addition to those required by the incriminatory exception. For simplicity we treat these further conditions as representing a separate "credibility" exception, which we explain in Part XII below. Thus, where the evidence has probative value only in showing that the defendant has a propensity to be untruthful, leave may not be given unless the requirements of the credibility exception are satisfied.

The credibility of a defence

11.34      Sometimes the prosecution will seek to argue that the defendant's explanation is so similar to one advanced by the defendant on a previous occasion that it is unlikely to be true. For example, in Reid^[30] the defendant, charged with robbing a mini-cab driver at knife-point, claimed that he had entered the cab only after the robbery had taken place – which was almost identical to a defence which he had earlier raised, unsuccessfully, to a similar charge of robbery. The similarity of the defence would, in that case, still have been probative even if it had previously been successful.

11.35      In that case the similar defences were unusual, and their similarity was therefore probative in itself, but if a defence is a common one, then repetition of it is not very significant. There is usually a limited range of defences available to most criminal charges, and it has therefore been argued that a court should not place too much weight on the fact that the accused has used a similar general defence (such as self-defence) in the past.^[31]

11.36      In the consultation paper, we expressed the provisional view that where a defendant has previously put forward substantially the same, unusual defence, that fact may be relevant to his or her credibility.^[32] This view was largely supported on consultation.^[33] Evidence of other misconduct might have substantial probative value on the credibility of the defence that is being advanced, as held by Gage J:

Similarities of defences which have been rejected by juries on previous occasions, for example false alibis or the defence that the incriminating substance has been planted and whether or not the accused pleaded guilty or was disbelieved having given evidence on oath, may be a legitimate matter for questions. These matters do not show a disposition to commit the offence in question; but they are clearly relevant to credibility.^[34]

11.37      There is, however, a distinction between the credibility of the defendant and that of the defence. Where the defendant is putting forward a defence which would be persuasive if she had not run it before, the fact that she has run it before goes to whether her defence in these proceedings is truthful, not to whether she is generally disposed to be untruthful. The point can be illustrated by supposing that Reid had been acquitted of the first robbery. On that hypothesis, his previous reliance on a defence which had on that occasion been successful would not in itself suggest that he had a propensity to lie; and in this respect it would make no difference that he was now relying on the same defence again. What the similarity between the two defences would suggest, however, is that on the second occasion the defence was unlikely to be true. Under our recommendations, therefore, the fact that Reid had run the same defence before would be eligible for admission under the incriminatory exception, if it satisfied the requirements of that exception. The only evidence that is ineligible for admission under the incriminatory exception, and admissible instead under the credibility exception, is evidence going to the defendant's propensity to be untruthful – not evidence which goes directly to the central issue of whether the defendant's version of events is in fact true or false.

The additional value of the evidence

11.38      Evidence might have substantial probative value in relation to a central issue in a case if taken on its own, but its value might be considerably reduced if there has already been, or it is clear that there will be, other, less prejudicial, evidence to prove the fact. We therefore think that the court should also have regard to what the evidence in question adds to the other evidence in the case.^[35]

11.39      We can see no reason for any artificial limit to be included prescribing that evidence of bad character going to a matter in issue must be adduced as part of the prosecution's case in chief. It may be that issues arise later in the case, possibly as a result of evidence elicited or adduced by the defendant, which alter the probative value of evidence of the defendant's bad character. The essence of the test should be the relationship between the probative value of the bad character evidence and the risk of prejudice.

Statutory guidance

11.40      We have said at paragraph 5.3 above that the test to be applied by the court should be capable of being applied predictably and consistently but be sufficiently flexible to cater for the infinite variety of factual situations which will occur. We believe this is best achieved by a test which leaves a court some room for individual judgment, while guiding that judgment on the matters to be borne in mind.

Provisional proposals on statutory guidelines

11.41      As part of our preferred option in the consultation paper we proposed that the court be required to take certain factors into accounts, and we proposed that they should include

(i) in the case of the evidence's probative value,

(A) the extent (if any) to which the evidence tends to suggest that the defendant has a propensity to act in the manner alleged;

(B) any similarities between the facts revealed by the evidence and those now alleged;

(C) the extent to which any such similarities may reasonably be attributed to coincidence; and

(D) any dissimilarities between the facts revealed by the evidence and those now alleged; and

(ii) in the case of the evidence's likely prejudicial effect,

(A) the risk of the fact-finders attaching undue significance to the evidence in question in determining whether the defendant is guilty as charged; and

(B) the risk of their convicting the defendant on the basis of his or her conduct on some other occasion or occasions, rather than because they are satisfied that he or she is guilty as charged.^[36]

11.42      One of our main criticisms of the test in DPP v P was, and still is, that it does not give sufficient guidance to courts on how to apply the test that it prescribes. We therefore proposed an approach based on structured guidelines. Three respondents supported it on the basis that it will promote greater consistency as between different judges and will ensure that all the various factors are taken into account. Three others were apprehensive about the proposal because of the perceived risk that adopting a "structured discretion" would generate a large number of appeals based on the precise meaning of the words used. Three respondents did not favour the guideline approach and argued that it is essential that, in the context of similar fact evidence, maximum flexibility is retained. For example, Ian Kennedy J said that

the variety of points which call for consideration in similar fact cases … is so wide that no structure could hope to highlight all of them, and the longer the list the more easy it is to argue that Parliament can not have given much weight to the unlisted point.

11.43      The argument for the guidelines is one of accessibility: that it is desirable for all parties and the court to know what features of evidence are pertinent to the question of admissibility. The danger is said to be that if it is spelt out exactly what the court ought to have been taking account of, then the defence have ammunition to appeal if the court failed to mention relevant factors. Of course, if the defence are right and the court not only failed to mention them but failed to take them into account, there can be no objection to the defence appealing against the ruling. The real fear is of encouraging unmeritorious appeals. One might argue that this should be met by training magistrates and judges to give reasons for their decisions.

11.44      Guidelines are not a panacea: reaching the decision about whether it is fair to admit potentially prejudicial evidence is an intrinsically difficult task, and will always require the careful exercise of judgment. But we do not accept that guidelines would be useless or, worse, a positive nuisance to those performing this balancing exercise. We note that the NZLC has recommended the following draft clause:

Propensity evidence offered by prosecution about defendants

45(3) When assessing the probative value of propensity evidence, the judge may consider, among other matters, the following:

(a) the frequency with which the acts or omissions which are the subject of the evidence have occurred;

(b) the connection in time between the acts or omissions which are the subject of the evidence and the acts or omissions which constitute the offence for which the defendant is being tried;

(c) the extent of the similarity between the acts or omissions which are the subject of the evidence and the acts or omissions which constitute the offence for which the defendant is being tried;

(d) the number of persons making allegations against the defendant that are the same as or similar to that which is the subject of the offence for which the defendant is being tried and whether those allegations may be the result of collusion or suggestibility;

(e) the extent to which the acts or omissions which are the subject of the evidence and the acts or omissions which constitute the offence for which the defendant is being tried are unusual.^[37]

11.45      We are still of the view that structured guidelines on how to apply the test are conducive to comprehensible and consistent application of the requisite test, and we recommend that the legislation set out factors for the court to take into account.^[38]

THE RECOMMENDATION: EVIDENCE OF SUBSTANTIAL PROBATIVE VALUE

11.46      We recommend that leave may be given to the prosecution to adduce evidence of the bad character of a defendant if

(1) the evidence has substantial probative value in relation to a matter in issue (other than whether the defendant has a propensity to be untruthful) which is itself of substantial importance in the context of the case as a whole, and

(2) the interests of justice require it to be admissible, even taking account of its potentially prejudicial effect.^[39]

11.47      We now address three special cases where evidence of bad character may be relevant to the case against the defendant.

THE THEFT ACT 1968, SECTION 27(3)

11.48      Section 27(3) of the Theft Act 1968 permits the prosecution to adduce in chief evidence of criminal disposition, in the form of evidence of prior possession of stolen goods (section 27(3)(a)) or previous convictions (section 27(3)(b)) in order to prove that the defendant knew or believed the goods to be stolen.^[40]

The defects of the current law and the proposals in the consultation paper

11.49      We have set out the problems in this subsection at paragraphs 4.17 – 4.23 above.

11.50      In the consultation paper, we considered the possibility of proposing that evidence of previous convictions should be admissible in handling cases where the accused admits the conduct alleged, but denies criminal knowledge. This option was proposed by the CLRC in its Evidence Report,^[41] and was adopted by the Royal Commission.^[42] It has one striking advantage over the existing provision in that it enables the prosecution to rely only on previous convictions, whereas paragraph (a) permits evidence of the mere fact that the defendant has had stolen goods in his or her possession, albeit without knowledge or belief that they were stolen.

11.51      However, we suspected that if this proposal were implemented, its effect could be easily avoided by disputing some other ingredient of the offence. For example, if it were disputed that the goods in question were stolen, it would not be possible to invoke the proposed rule as there would be an outstanding issue in the case other than knowledge. We had little doubt that this would seriously undermine the usefulness of this option, and we rejected it in the consultation paper.^[43]

11.52      We provisionally proposed that section 27(3) be repealed, and that handling cases be dealt with under the same rules as other cases.^[44] There are many other instances where the mental element of an offence may be difficult to prove, and previous misconduct might be of assistance – though this may be less of a problem now that inferences can be drawn from a defendant's failure to give evidence^[45] – and we see no reason to single out one such offence for special treatment. Evidence of previous misconduct is not necessarily more probative, or less prejudicial, in the case of handling than in any other case. We did not think that any special rule for handling cases was needed or justified.

The response on consultation

11.53      Thirty-three respondents addressed this proposal explicitly. Of those, 27 respondents were in favour of the provisional conclusion and three were against. Two respondents favoured the second option, and one thought the provision should remain as it is on the basis that it "does little, if any, harm and sometimes may do some good". One of the three who were against the proposal conceded that if our other proposals were to be implemented, section 27(3) would become otiose. One respondent thought the provision should be amended, so that such evidence could only be used in rebuttal.

11.54      Our view, that the provision is unnecessary under the current law, and would continue to be so following our recommendations for incriminatory evidence, was fortified by the views of respondents. The most common argument given in support of our proposal was that the section is very rarely used. The Criminal Bar Association commented that "if the prosecution does not have a sufficient case without resorting to this provision, then the general thinking is that the prosecution ought to fail".

11.55      As no strong arguments have been put forward which cause us to change our view from that in the consultation paper, we recommend that section 27(3) of the Theft Act 1968 should be repealed.^[46]

THE OFFICIAL SECRETS ACT 1911, SECTION 1(2)

11.56      Section 1(2) of the Official Secrets Act 1911 allows the accused's purpose (an element of the offence) to be proved by reference to his or her "known character as proved".^[47] In the consultation paper we invited views as to whether this provision should be repealed or amended.^[48] Eighteen respondents expressed a view on this proposal, eight thinking it should be repealed, and eight thinking it should not be changed.^[49]

11.57      The main argument in favour of repeal was that the provision is unfair to the accused. Some respondents suggested that evidence introduced under this provision should be required to meet a specific standard of probative value before it is admitted. Those who opposed change pointed to the unusual nature of the offence and the circumstances in which it arises, argued that "the difficulty of proving this offence requires this special provision", or referred to the fact that the Attorney General's consent is required as an important safeguard against its abuse.

11.58      We have taken account of the views of experienced respondents that this is an offence with unique difficulties for the prosecution, given that the subject matter may inhibit the calling of certain types of evidence. The character of the accused is not an element of the offence charged. It is merely one route by which one element, namely purpose, may be proved.

11.59      It is not for us, dealing as we are with matters of evidence, to address the substantive question whether, as a matter of policy, the purpose of a person charged under the Official Secrets Act should be capable of being proved by reference solely to character. It is, however, within our remit to consider how, and by reference to what mechanisms and tests, such evidence may be admitted given that the explicit terms of the section require the person's character to be "proved". There is nothing in section 1(2) which is inimical to the application of the principles we have identified as appropriate to determining when evidence of bad character may be admitted.

11.60      The statutory provisions in the draft Bill (which set limits on how certain matters of character may be proved) will apply where the prosecution seeks to prove a person's known character under section 1(2). An application to admit such evidence would be made under clause 8 and the evidence would only be admitted if it were of substantial probative value, and the interests of justice required it to be admissible, taking account of the risk of prejudice.

11.61      We recommend that the admissibility of evidence of the defendant's bad character pursuant to section 1(2) of the Official Secrets Act 1911 should be determined by the above rules.

THE EFFECT OF AN EARLIER ACQUITTAL

11.62      The law as it stood at the time of the consultation paper had the effect that, no matter how probative evidence of previous alleged offences might be in relation to a new prosecution, if the defendant had previously been acquitted of them, no evidence could be adduced of them.^[50] We did not explore this issue in the consultation paper, but simply proposed no change. We did, however, explore this issue in depth in our double jeopardy consultation paper, and proposed that the law should be changed.^[51] This is no longer held to be the law in consequence of a decision of the House of Lords last year in a case where evidence of previous acquittals was admitted under the "similar fact" rubric.^[52] The result of that decision is to effect the change which we had thought was required: there is now no special rule of inadmissibility for evidence of previous offences of which the defendant has been acquitted.^[53] This effect, which would apply to all the exceptions we recommend but would be particularly applicable to the incriminatory exception, is preserved by the terms of the draft Bill appended to this Report. The defence would still be able to apply to have the evidence excluded under section 78(1) of PACE.

Ý
Ü   Þ

Note 1   Paras 10.20 – 10.22 of the consultation paper.    [Back]

Note 2   Paras 10.23 – 10.36 of the consultation paper.    [Back]

Note 3   Contrary to the Theft Act 1968, s 9(1)(a).    [Back]

Note 4   See Cokar [1960] 2 QB 207, n 47 in Part IV above.    [Back]

Note 5   Sang [1980] AC 402.    [Back]

Note 6   Paras 10.37 – 10.46 of the consultation paper.    [Back]

Note 7   Although it was presented by the Committee as making no change to the then existing law, it would in fact have extended the circumstances in which previous misconduct was admissible. See para 10.38 of the consultation paper.    [Back]

Note 8   Clause 3(2) of the Bill accompanying the CLRC Evidence Report.    [Back]

Note 9   Hence the guidelines for trials involving disputed identification laid down in Turnbull [1977] QB 224.    [Back]

Note 10   Paras 10.47 – 10.59 of the consultation paper.    [Back]

Note 11   (1995) 127 ALR 99. This authority has been superseded in those jurisdictions where the Evidence Act 1995 (Cth) and the Evidence Act 1995 (New South Wales) apply.    [Back]

Note 12   The majority regarded “reasonable” and “rational” as synonymous in this context: ibid, at p 114.    [Back]

Note 13   Pfennig (1995) 127 ALR 99, 113, per Mason CJ, Deane and Dawson JJ, summarising the effect of Hoch (1988) 165 CLR 292, 294. Pfennig has since been applied in, eg, Robertson (1997) 91 A Crim R 388: the defendant was charged with administering a stupefying drug or thing and indecently assaulting the complainant. Accounts from others who claimed he had drugged their drinks and possibly acted indecently towards them were permitted as evidence against R on the basis that, once concoction was excluded, there was no rational explanation other than that they supported an inference of guilt.    [Back]

Note 14   R Munday, “Similar Fact Evidence and the Risk of Contaminated Testimony” [1995] CLJ 522, 524.    [Back]

Note 15   Paras 10.60 – 10.72 of the consultation paper.    [Back]

Note 16   Evidence Act 1995 (Cth), ss 97(1)(b), 98(1)(b).    [Back]

Note 17   Ibid, s 101(2).    [Back]

Note 18   Ibid, s 137.    [Back]

Note 19   Ibid.    [Back]

Note 20   Ibid, s 101(2).    [Back]

Note 21   Ibid, ss 97(1)(b), 98(1)(b).    [Back]

Note 22   See paras 10.64 – 10.70.    [Back]

Note 23   Provisional proposal 16, set out at paras 10.73 – 10.85.    [Back]

Note 24   See para 7.7 above and the discussion of “substantial” at paras 7.8 – 7.17 above.    [Back]

Note 25   Para 10.65 (emphasis in original). The omitted section of the para addressed (and rejected) the possible requirement that the evidence have “significant” probative value.    [Back]

Note 26   Para 10.65 (emphasis in original).    [Back]

Note 27   See para 7.7 above.    [Back]

Note 28   See para 7.16 above.    [Back]

Note 29   See cl 8(2)(b) and 8(3)(b)(iii) of the draft Bill.    [Back]

Note 30   [1989] Crim LR 719.    [Back]

Note 31   R Munday, “The Paradox of Cross-Examination to Credit – Simply Too Close for Comfort” [1994] CLJ 303, 315-316.    [Back]

Note 32   Para 6.73.    [Back]

Note 33   Sixteen respondents addressed this provisional view. Thirteen agreed, (although two added caveats), and one substantially endorsed it. One respondent thought that we had understated the case, but one would state the proposition more narrowly, and would agree “only were the description ‘unusual’ to be narrowly interpreted so that it meant, in effect, ‘peculiar’.”    [Back]

Note 34   In a judgment prepared by Stuart-Smith LJ, in McLeod [1994] 1 WLR 1500, 1512H– 1513A.    [Back]

Note 35   See cl 8(3)(b)(i) and (ii) of the draft Bill.    [Back]

Note 36   Paras 10.79 and 10.80 and provisional proposal 17 of the consultation paper.    [Back]

Note 37   NZLC, Evidence: Reform of the Law (1999) Report 55.    [Back]

Note 38   See para 7.19 above.    [Back]

Note 39   This recommendation is given effect by cl 8 of the draft Bill.    [Back]

Note 40   The case law on this provision is described at paras 2.27 – 2.30 above     [Back]

Note 41   Para 92.     [Back]

Note 42   Report of the Royal Commission, ch 8 para 31, and Recommendation 192.    [Back]

Note 43   Provisional proposal 43(2). Paras 14.11 – 14.12.    [Back]

Note 44   Provisional proposal 44. Para 14.13.    [Back]

Note 45   Criminal Justice and Public Order Act 1994, s 35. See para 4.40 above.    [Back]

Note 46   See cl 20(3)(b) of the draft Bill.    [Back]

Note 47   See para 2.31 above.    [Back]

Note 48   Provisional proposal 42, para 14.3.    [Back]

Note 49   One advised amendment, rather than repeal, and one collective response was divided on the matter.    [Back]

Note 50   The rule in Sambasivam v Public Prosecutor, Federation of Malaya [1950] AC 458.    [Back]

Note 51   See Double Jeopardy (1999) Consultation Paper No 156, Part VIII and provisional proposal 24.    [Back]

Note 52   Z [2000] 2 AC 483.    [Back]

Note 53   We therefore did not make any recommendation for abolition of the Sambasivam rule in Double Jeopardy and Prosecution Appeals (2001) Law Com No 267 paras 2.22 – 2.28.    [Back]