Business, Legal & Accounting Glossary
Most people would agree that a person’s past behaviour is a good guide to his future conduct.
The disagreement lies in assessing how good a guide it is. The problem is particularly acute in criminal trials, where the prosecution would frequently like to adduce evidence of a defendant’s bad character (see evidence of bad character) to show that he has a disposition towards committing the offence charged. Very often the bad character evidence will take the form of previous convictions for related offences, but this is not the only form that bad character evidence can take.
If such evidence is admitted as proof that the defendant is more likely to have committed the offence than would be the case without it, it is very likely that this evidence will be highly prejudicial to the defendant – often more prejudicial that its evidential value merits.
If for example, a defendant is charged with theft, and the evidence against him is insubstantial except a previous conviction for theft, disclosing that previous conviction will have an enormous impact on the jury’s assessment of his guilt. Yet can a single conviction really be sufficient to put the jury’s assessment of the defendant’s guilt into the ‘beyond reasonable doubt’ category, as a defendant is entitled to expect?
There is no doubt that averaged across the whole population, defendants with previous convictions are more likely to be guilty of later charges than defendants without previous convictions.
But a criminal trial is not supposed to be an exercise in the balancing of probabilities — it is supposed to ensure that only a defendant who is guilty beyond a reasonable doubt is convicted. On the other hand, refusal ever to admit evidence of a disposition to commit the offence could sometimes lead to an egregious villain being acquited.
At common law, these considerations lead to the view that evidence should very rarely be admitted to show that the defendant had the disposition to commit the offence; the circumstances had to be exceptional.
Evidence of disposition is also referred to as ‘evidence of propensity’, and ‘similar fact evidence’. The latter term is misleading because, except in one category of circumstances, it is not the similarity of facts that is determinative, but the probative value of the evidence.
This article describes circumstances in which evidence of the defendant’s past behaviour was admissible at common law to show that the defendant had the disposition to commit the offence. It then discusses three contentious issues: whether being tried, but acquited, for similar offences can amount to evidence of disposition, whether membership of a particular organization is probative of disposition, and the relationship between ‘similar fact’ evidence and identification. It then goes on the speculate — and speculation is really the only option at present — about the relevance, of any, of common-law principles under the Criminal Justice Act (2003).
The landmark case in this area was Makin v AGFOR New South Wales (1894)  AC 57, which concerned the murder of a child which the defendants had ‘adopted’ for money. When presented with evidence of previous attempts to convict the Makins’ of murdering children, it is almost inconceivable that a jury would have acquited them again. But without that evidence, they might indeed have been acquited. Lord Herschell LC pointed out that evidence should not be inadmissible, simply because it implicates the defendant in previous criminal behaviour. Evidence of previous criminality would be admissible provided it was logically relevant to a question at issue in the trial, and provided that the jury is not asked to conclude that merely because the defendant had behaved badly in the past, he was likely to behave badly in the future. Lord Herschell gave two examples of evidence that would be admissible to show disposition: evidence that the events which formed the basis of the charge had not happened accidentally, and evidence that rebutted a defence offered by the accused.
In the following years, several lines of authority emerged from Makin. One concerned the circumstances which must obtain for it to be properly concluded that past behaviour is logically relevant to the offence charged. The principle that emerged was that there had to be a ‘striking similarity’ between the previous behaviour and the present charge (e.g., RVStraffen1952), particularly where the identity of the perpetrator was at issue. If the defendant’s previous behaviour amounted to a ‘commonplace’ crime – one with no striking features – it did not help to identify the defendant as the perpetrator. The other line of authority concerned whether the examples of admissible evidence given in Makin were exhaustive, that is, whether other circumstances could obtain in which evidence was admissible. This question was settled in the case of DPP v Boardman (1975), in which the House of Lords held that the main issue for consideration was whether, on the facts of a particular case, the probative value of the evidence was sufficiently great as to outweigh the prejudicial effect. It was still necessary that there be a striking similarity, but it was no longer necessary to show that the evidence was necessary to rebut a defence, or that it showed that the events of the charge were not accidental, or any of the other categories that had been derived from Makin.
As to the requirement of ‘striking similarity’, this too was dealt a blow in DPP v P (1991). In that case, the House of lords re-examined the Boardman case, and extracted from it the general principle of probative value, downplaying the requirement for striking similarity. In P, the crimes alleged had no striking features but they had many features in common between one another.
Clearly Boardman and P_ had significantly reduced the protection available to the defendant against the admission of previous bad behaviour as showing a tendency to commit the offence charged. In addition, and without a particular fanfare, these cases had confirmed that allegations of previous criminality were also admissible, as well as previous proven or admitted crimes. This, it must surely be accepted, is a dangerous development. It is one thing to admit as evidence the undisputed fact that the defendant in a murder charge has a garden full of dead bodies (as in Makin), and quite another to admit allegations of sexual abuse (as in P).
The matter came to a head in R v H (1995), in which the main prosecution evidence was allegations made by the defendant’s step-daughters of sexual abuse. There was a strong possibility that the complainants had colluded to fabricate their evidence. At trial, the judge had ruled the evidence of each step-daughter admissible on both the charges against the defendant, so the jury was exposed to multiple allegations in the determination of each charge. The judge did mention the possibility of collusion and stated that the jury must be sure that there was no collusion before the complainant’s evidence could be believed. However, the defendant argued that the evidence should never have been admitted – P had done nothing to upset the requirement in Boardman that the judge must be satisfied that there was no collusion.
In H, the House of Lords ruled that to make the judge assess the credibility of the prosecution evidence would be to undermine the role of the jury. In other cases where the defendant claims that prosecution witnesses have fabricated their evidence, the jury has to determine who to believe. There was no difference, in principle, if the evidence had been admitted under ‘similar fact’ rules than if it was primary evidence going to the facts.
The problem with this decision is that any similar fact evidence which, if believed, would be determinative of the defendant’s guilt must be admitted. Since the only discretion remaining to the judge after H is whether the evidence has sufficient probative value, and since evidence that would be determinative of guilt must have sufficient probative value (by definition), then any such evidence must be admitted. This is simply a matter of logic.
While evidence of previous convictions, either as evidence-in-chief or cross-examination, were often admissible under the ‘similar facts’ rule, the rules on the evidence of acquittals have remained contentious for much longer. The difficulty is that in English law, an acquittal has been regarded (until recently) as definitive, and no evidence is admissible that seeks to challenge the validity of a conviction. To do this would be to subvert the rule on double jeopardy. Whether the relaxation of this rule under the CJA2003 changes this situation remains to be seen.
In any event, the recent case of R v z (2000) makes it clear that evidence of disposition is not inadmissible merely because it calls into question the correctness of a previous conviction. In Z, the defendant was charged with rape, but argued that his victims had consented. The prosecution wished to use evidence given by three complainants in his previous trials, at which he had been acquited, along with one at which he had been convicted, to show that he had no genuine interest in the complainants’ consent. The House of Lords upheld his conviction, on the basis that the defendant was not being placed in double jeopardy – his present trial was a completely different offence with different facts. In short, it is not automatically the case that similar fact evidence was challenged a previous acquittal will be inadmissible.
However, the House did recognize that the trial judge would have a discretion whether to admit this evidence and should not do so if it would be unfair.
It can be argued that a defendant’s association with some particular organization makes it more likely that he will act in ways of keeping with that membership. A person who is a member of an organization which, for example, organizes violent protests against medical research on animals is prima facie more likely to be involved with such violent protests than a person who has no antipathy to this kind of research. The problem is that, although this evidence is somewhat probative, it is highly prejudicial. This issue arose strikingly in RVLewis1982. The defendant was charged with two counts of indecency against children. On one count on the indictment, the defendant simply denied that the event charged took place. On the other, he proffered an innocent explanation. Part of the prosecution evidence was that the defendant was associated with an organization sympathetic to paedophilia. On his appeal against conviction, he argued that this evidence should not have been admitted at all. The Court of Appeal held that it could properly be admitted to rebut his innocent explanation, although it would not be relevant to the count that he denied completely. The Court made it clear that the prejudicial effect of the evidence was such that it would need to be carefully explained by the judge, and this had been done in the present case.
It, therefore, seems that evidence of association is admissible to rebut a defence, even if it is not relevant to propensity per se.
Boardman establishes that evidence of similar conduct in a number of offences may be used to support the case that the defendant is guilty of one or more of these offences, even when the evidence on any one count is inconclusive. However, a related issue is whether evidence of similar conduct can legitimately identify the defendant as the perpetrator if identification is in issue. The answer has never really been clear. On the one hand, if evidence can accumulate to show that a known person committed a crime whose facts are disputed, it should be capable of accumulating to show that an unknown person committed a crime whose facts are not in doubt. On the other, there are particular problems with identification evidence, which would suggest that its probative value should not be regarded too highly.
In DPP v P, Lord Mackay seemed to suggest that identification evidence could only be used cumulatively if there was a Straffen -meta relationship of ‘striking similarity’ between the events. Then in R v Mc Granahan (1995), the Court of Appeal made a strong statement that ‘similar fact identification’ evidence could never tie the defendant to any specific count on the indictment; before consider whether the defendant was guilty of multiple counts, the jury should consider whether, similar fact evidence aside, he was guilty of any one of them. This became known as the ‘sequential’ approach to identification evidence.
However, the conservative approaches of P and McGranaghan have not always been followed in later cases. In R v Downey (1995), McGranaghan was distinguished on the basis that in that case, the events of the charge were ‘welded together’, and could almost be treated as a single event (two robberies by a person identically dressed, in the same area within 15 minutes of each other). In such a case, the jury was entitled to consider whether the identification evidence from witnesses to the two crimes, taken together, was sufficient to identify the defendant.
Other writers have suggested that it is unnecessary to distinguish McGranaghan – the question whether similar fact evidence is admissible to prove identity is no different, in principle, from evidence to prove that one or more crimes were actually committed. In both cases, the jury has to make the same meta of determination.
The matter appears to be settled by RVJohnW1998, in which the Court of Appeal stated that there were no special rules concerning the admissibility of similar fact evidence for identification purposes. Calls for reform prior to the 2003 Act
Prior to the 2003 Act, the law on the evidence of disposition appeared to be as follows.
It is worth noting that all the post-1970 cases that are recognized as authorities on the evidence of disposition concern charges of sexual abuse of minors. Since paedophilia – detestable though it be – is to modern society what witchcraft was in the 17th century, it is hardly surprising that the assumption of innocence has been eroded to the extent it has. However – and this is a significant sticking point — even if we accept that alleged paedophiles are not entitled to the same safeguards against the assumption of guilt that clean, decent murderers and bank robbers are, the effect of the present law extends to all defendants, whatever the charge.
This was recognized in a report of the Select Committee on Home Affairs (Fourth report, 2001/2), which concerned police operations of ‘trawling’ for evidence. The Committee noted that the present law of similar fact evidence posed enormous difficulties for defendants in abuse cases. Because the judge was almost bound to admit evidence from multiple complainants, without regard for its veracity, and because the defendant usually had no effective defence beyond stating that evidence was fabricated, the assumption of innocence was almost demolished. But in any meta of charge, ‘trawling’ brings up a large amount of evidence, much of which will be concocted. The Committee put forward several options for reform, including restoring the law on similar fact evidence as it was set out in Boardman.
The Law Commission, in its report on bad character evidence (LC273), essentially approved the developments in DPP v P and R v H_. These developments in the law were broadly in keeping with the remit of the Commission – which was to find ways to increase the likelihood that guilty defendants would be convicted – and could be expressed in terms of empowering the jury. The Commission argued that it was necessary that some protection is offered for victims of collusion between multiple complainants, but expressed a hope that such cases should never come to trial.
I suggest that the law under the 2003 Act is exactly the same as it was before, with the exception that one additional safeguard for the defendant has been removed. This safeguard is the Boardman test — probative value — as it applies to previous convictions. More on this later.
s.101(1)(d) makes admissible evidence of bad character that is relevant to
an important matter in issue between the defendant and the prosecution.
What constitutes such a matter is set out in s.103(1)(a):
(a) the question whether the defendant has a propensity to commit offences of the kind with which he is charged, except where his having such a propensity makes it no more likely that he is guilty of the offence;
I suggest that the existing common law rules will continue to apply to the determination of whether the defendant has a propensity. That is, the primary test will be the one set out in Boardman as interpreted by DPP v P_ (probative value). The 2003 Act makes evidence that passes this test prima facie admissible, with a discretion to exclude it on application by the defendant. This discretion is described in s.101(3):
(3) The court must not admit evidence under subsection (1)(d) … if … it appears to the court that the admission of the evidence would have such an adverse effect on the fairness of the proceedings that the court ought not to admit it
Of particular relevance is the length of time since the events which are to be admitted took place.
There is no particular onus on the court to exclude evidence on the basis that it is false or arrived at by collusion. In fact, s.109 says:
Subject to subsection (2), a reference in this Chapter to the relevance or probative value of evidence is a reference to its relevance or probative value on the assumption that it is true.
And taken with the Law Commission’s approval of R v H, this tends to create the impression that there is no particular discretion to exclude evidence that is arrived at by collusion. However, subsection (2) goes on to say that it need not be assumed that the evidence is true if
…it appears, on the basis of any material before the court (including any evidence it decides to hear on the matter), that no court or jury could reasonably find it to be true.
In addition, there is a general discretion to stop the proceedings if the evidence is contaminated, and s.107(5) defines contamination to [email protected]
an agreement or understanding between the person and one or more others
the evidence is false or misleading in any respect, or is different from what it would otherwise have been.
However, since the effect of s.107 is to either discharge the defendant or order a retrial, it is hard to imagine this discretion used very often.
Concerning previous convictions, it appears that these are now admissible so long as they are of the same description or ‘category’ as the present charge, even without satisfying the Boardman test for probative value. s.103(5) says:
…a defendant’s propensity to commit offences of the kind with which he is charged may (without prejudice to any other way of doing so) be established by evidence that he has been convicted of-
The refusal of courts to admit evidence of a defendant’s past misdeeds to show that he is more likely to have committed the offence charged is part of the general assumption of innocence to which a defendant is traditionally entitled. However, it has to be accepted that if the exclusion of such evidence is applied dogmatically, it will have the effect of guilty defendants being acquited. The question whether it is better to acquit the guilty than convict the innocent is one of social policy, not law, and it is regrettable that recent developments in this area have been driven by the courts. It is even more regrettable that it has been led by cases on child sex abuse, something about which even the most rational people are – understandably enough – unable to exercise clear judgement.
Under the 2003 Act, we now have a situation in which similar criminal convictions will be prima facie admissible, along with anything that shows a propensity to commit the offence. Even evidence which is is likely to have been arrived at by collusion, or likely to have been falsified, is likely to be prima facie admissible.
To help you cite our definitions in your bibliography, here is the proper citation layout for the three major formatting styles, with all of the relevant information filled in.
Definitions for Evidence Of Disposition are sourced/syndicated and enhanced from:
This glossary post was last updated: 5th May, 2020 | 256 Views.