England and Wales Court of Appeal (Criminal Division) Decisions

IN THE COURT OF APPEAL
CRIMINAL DIVISION

		Royal Courts of Justice Strand London WC2A 2LL
		Friday 26th June 1998

REASONS FOR JUDGMENT

BUXTON LJ:

This appeal raises a short issue on the admissibility of evidence of the good character of prosecution witnesses. The question is not directly covered by authority: partly, we think, because the step taken in this case was distinctly unusual, and outside the experience of any member of this court. We have already determined that the appeal must be allowed, and we now give our reasons.

Mr Hamilton was convicted of unlawful wounding, the case for the prosecution being that while in a bar he had been gratuitously abusive to the victim, one Myers, and then lunged at him with a broken glass, stabbing him in the head. Myers, and his girl-friend Clarke, gave evidence to that effect. Mr Hamilton, on advice, did not give evidence. His defence consisted of a strong attack on the evidence of both Myers and Clarke, alleging that it was Myers who had been the aggressor. As part of that attack counsel for Mr Hamilton suggested to Myers that (i) Myers had been in the bar for the purpose of buying cocaine; (ii) it was Myers who had been the aggressor; (iii) both Myers and Clarke were lying. The Assistant Recorder further considered that an allegation had been made that Myers and Clarke had left the bar during their encounter with Mr Hamilton for the purpose of obtaining weapons, but Mr Baker who appeared for Mr Hamilton both at the trial and before us assures us that he was not in receipt of instructions to that effect, and that therefore the matter was simply raised with the witnesses without a specific allegation being made.

Mr Hamilton originally appealed on three grounds. Two of them, that the judge should have stopped the case after the prosecution evidence because of inconsistencies between the evidence of Myers and of Clarke; and that he should have directed the jury about self-defence on the basis of one passing use of that phrase by Mr Hamilton in the course of a police interview the burden of which was that he had no real recollection of the events of the evening; were plainly unfounded. The third complaint presents more difficulty.

At the end of the evidence prosecution counsel applied for leave to elicit before the jury the fact that neither Myers nor Clarke had any criminal convictions. That application seems originally to have been based on a submission that, by his attack on Myers and Clarke, counsel for Mr Hamilton had triggered the provisions of section 1(f)(ii) of the Criminal Evidence Act 1898. That was no doubt the case, were Mr Hamilton to give evidence; but the point is plainly irrelevant to any issue about the permitted content of the original prosecution case. The judge however allowed the evidence to be given, basing himself partly on what he thought to be practice and partly on statements in works of authority about the rehabilitation of a witness. Further, in his summing-up he reminded the jury that neither Myers nor Clarke had previous convictions, though without giving any direction as to the evidential value of that fact.

We are quite clear that the judge was wrong in the grounds on which he so ruled. The authorities to which he referred, and in particular those expounded in Phipson on Evidence (14th edition), para 12-38, go no further than to say that where evidence is (unusually) called to show that a witness has a general reputation for untruthfulness, such evidence may be rebutted by contrary evidence of reputation. Evidence to rebut particular allegations of misconduct may not be admitted (Wood [1951] 2 All ER 112), and rightly so. Either such allegations go simply to credibility, in which case the witness's answer is final; or they go to an issue, in which case they are adjudicated on the basis of the jury's assessment of the evidence in the case, without reference to collateral matters. That can be illustrated from the matters complained of in the present case. A suggestion that prosecution witnesses are lying may be, as the judge pointed out, an allegation of perjury; but for the purpose of the case in which the allegation is made it simply puts in issue that facts of that case, which the jury decide according to the evidence. Similarly the allegation that it was Myers who was the aggressor. By contrast, if the judge considered that the cocaine allegation was irrelevant and introduced solely to cast doubt on the character of Myers it was for him to direct the jury accordingly in the light of Myers's denial.

These considerations apply a fortiori to the calling by the prosecution of evidence of character to bolster the testimony of the prosecution wetness who were attacked. Such evidence has no probative value in relation to any issue in the case, and is rightly excluded on grounds of collaterality. And so far as issues of the witnesses' character are concerned, those are concluded, without recourse to any such evidence, by the finality of the witness's answer to specific allegations.

Counsel for the prosecution however pointed out that it is general practice to ask witnesses at the start of their evidence, for instance, their age and state of employment, no doubt with a view, he said, to underpinning their credit in the eyes of the jury. What had happened at this trial was only an extension of that practice. We cannot agree. The practice is noted by Professor Cross (Cross on Evidence (8th edition), p312), but is not thought by the learned author to undermine the law as set out above. The difference is not simply one of degree. Lack of convictions is not part of the general background of a witness, but goes and goes only to his credibility. Unlike the fact that a man is a plumber or chartered accountant, once a witness's lack of convictions is introduced a judge is likely to think, as the judge in this case thought, that he should remind the jury of it. Where the witness's credit is attacked the prosecution is not likely to argue that the nature of his employment helps the jury: not so, as this case demonstrates, his lack of convictions.

That, however, was not the main ground on which the prosecution resisted this appeal. Rather, counsel relied strongly on an unreported case in this court that had not been put before the judge below, O'Connor (29 October 1996). O'Connor was convicted of an affray involving, to put it shortly, serious assaults on police officers when they intervened to restore order after a brawl in a public house. His case was that the police evidence had been fabricated, in order to conceal the excessive and unreasonable violence that had been used when arresting him as a result of the original brawl (as to the appellant's participation in which the jury were not able to agree, and were discharged from giving a verdict). Allegations of perjury, conspiracy and collusion were duly put to each of the six police officers who gave evidence about the arrest. In re-examination prosecuting counsel was permitted by the judge to ask them whether they had any convictions recorded against them; and whether they had any disciplinary finding against them; the answers in both cases being in the negative. The appeal concerned the admissibility of that evidence.

The court dealt with the matter in this way:

The first question that we have to consider is whether evidence of disciplinary proceedings, which have established breaches of police regulations by a police officer witness, is admissible in a criminal trial. The answer is that such evidence is relevant and admissible, and that is established by the case of Edwards [1991] 1 WLR 207. The defendant who cross-examines a police officer about disciplinary proceedings will be bound by his answers because the questioning goes to the credit of the witness. As such evidence is relevant and admissible, it must follow that in appropriate circumstances evidence of an absence of a disciplinary finding will be relevant and admissible, and the judge has a discretionary power whether or not to allow such evidence to be adduced.

The same reasoning, it is to be assumed, was held by the court in O'Connor to apply to the officers' absence of criminal convictions.

We approach O'Connor with some caution, not least because the court in that case does not appear to have had the benefit of being shown the statements of principle to which we have referred earlier in this judgment. In particular, we would wish to reserve for further consideration the proposition that, because a witness can be attacked on grounds of bad character, evidence of his good character is by the same token admissible as part of the case of whichever side calls him. That proposition, if generally valid as an issue of admissibility, would seem to run contrary to the principles that we have sought to explain above.

In the present case, however, there is no need to pursue these difficulties. The judge did not think that he was applying the O'Connor approach, because the case was not ventilated before him. It therefore follows that, even if he should have applied O'Connor, he did not exercise the discretionary power nonetheless to exclude such evidence that is an important part of the O'Connor rule. Rather, he simply asked himself whether admission of the evidence was "fair". It thus falls to this court, even on the assumption that we should follow O'Connor, to exercise that discretion anew.

The court in O'Connor emphasised that the police officers in that case were being accused of conduct that is amongst the most serious of which a police officer could be guilty, including the infliction of serious violence on the appellant, and then a planned conspiracy of perjury to conceal that fact. Those factors, and the extreme nature of the case, weighed very heavily with the court in thinking that the judge's ruling fell within the proper scope of his discretion. The present case is very different. Apart, possibly, from the allegation about cocaine, the attack on Myers and Clarke was no more than the robust pursuit of a legitimate defence case. In particular, what the judge characterised, no doubt technically correctly, as allegations of perjury were couched in terms of suggesting that that witnesses were making their evidence up as they went along: clearly a far cry from the pre-planned conspiracy that was alleged in O'Connor. If evidence of good character were held to be properly before the jury in this case it is difficult to see how any defence that challenged the veracity of the prosecution account of events would not lead to the same conclusion. That is plainly not what the court in O'Connor had in mind.

For those reasons, therefore, we hold that the evidence of lack of previous convictions on the part of Myers and of Clarke should not have been admitted. In a case that depended so largely on their evidence, and where the defence consisted almost entirely of an attack on the veracity of that evidence, it is impossible to say what effect that evidence may have had upon the jury. We cannot say that this conviction was safe. We would normally consider a retrial in such a case, but Mr Hamilton has already served the equivalent of a sixteen month sentence. Were he to be tried again, and be convicted, he could only be sentenced to no more than the two years imposed at the first trial. We therefore do not consider that public funds would be properly employed in a further trial in this matter.