Jenkins v R [2004] HCA 57; 79 ALJR 252; 211 ALR 116

High Court of Australia|2004-11-17|Before: Coldrey J, Gleeson CJ, Heydon JJ

P F Tehan QC with L C Carter for the appellant (instructed by Worcester & Co)

J D McArdle QC with J B Saunders for the respondent (instructed by Solicitor for Public Prosecutions (Victoria))

Notice: This copy of the Court's Reasons for Judgment is subject to formal revision prior to publication in the Commonwealth Law Reports.

Criminal law - Direction to jury - Accomplice warning - Prosecution witness pleaded guilty to criminal offences related to charges against appellant - Testimony of witness substantially undisputed - Defence did not seek to attack credit of witness - Defence case relied in part on evidence of witness - No warning sought at trial as to reliability of witness - Whether trial judge obliged to warn jury that it would be dangerous to convict on uncorroborated evidence of an accomplice - Whether trial judge obliged to warn jury that evidence was potentially unreliable and must be subjected to careful scrutiny.

GLEESON CJ, GUMMOW, HAYNE, CALLINAN AND HEYDON JJ. Following a lengthy trial in the Supreme Court of Victoria before Coldrey J and a jury, the appellant was convicted of five offences of obtaining a financial advantage by deception, and five offences of dishonestly furnishing false information for the purpose of obtaining a loan. He was sentenced to a term of imprisonment. An appeal to the Court of Appeal of Victoria resulted in the quashing of the convictions for dishonestly furnishing false information, but the convictions for obtaining a financial advantage by deception were upheld[1]. The appellant argues in this Court that those convictions ought also to have been quashed, on the ground that the trial judge failed to give the jury an accomplice warning in relation to the evidence of a witness called by the prosecution. It is common ground that the Court of Appeal dealt with that issue, adversely to the appellant, on the basis of a factual misunderstanding. Nevertheless, the respondent contends that the appeal should be dismissed for the reason that, in the circumstances of the case, no such warning was necessary.

The offences with which this Court is concerned occurred in the period between May 1988 and August 1989. The appellant was a property developer. He, and companies with which he was associated, obtained financial accommodation in the form of credit or guarantee facilities from the Order of the Sons of Temperance National Division Friendly Society (OST). It is convenient to describe those facilities compendiously as loans. Under the Friendly Societies Act 1986 (Vic), OST was limited in the amount it could advance on the security of a mortgage over commercial land by a "loan valuation ratio" related to the value of the land. The essential allegation against the appellant was that the loans from OST were arranged on the basis of false or misleading valuation reports. OST, the prosecution alleged, was provided with inflated valuations.
All of the loans in question were approved by Paul Robinson, the Investment Director of OST. They were arranged on behalf of the appellant by a mortgage broker, Keith Bulfin of McKinley Wilson & Co Ltd. In each case, the loan application was supported by a valuation made by an accredited valuer, Tibor Verebes.
Although this was not revealed in evidence, Robinson, Bulfin and Verebes had all pleaded guilty to criminal offences relating to OST's lending to the appellant. Robinson was discharged without conviction. Bulfin was convicted and sentenced. So also was Verebes.
The Court of Appeal mistakenly thought that the convictions of Verebes related to properties different from those with which the appellant was concerned, and that Verebes and the appellant could not have been regarded as accomplices. That was the basis on which the Court of Appeal disposed of the complaint about the absence of an accomplice warning. The confusion probably arose because one of the charges to which Verebes pleaded guilty was a "rolled-up" count of furnishing false information, in the form of valuation reports, to OST. In fact, the information in question related to the very properties the subject of five of the counts against the appellant. In this Court, the respondent conceded that, on the prosecution case at trial, Verebes and the appellant were accomplices.
Although the evidence was detailed, and in some respects complicated, the case against the appellant may be reduced to quite simple terms. It is not so easy to formulate the defence.
We are presently concerned with counts 2, 5 and 13, 7, and 14 in the indictment. The following facts were admitted.
Count 2 concerned land and a building at Glen Crag ("Glen Crag"). On 22 April 1988, the appellant, on behalf of himself or a nominee, contracted to purchase the property for $2.41 million. On 29 April 1988, Verebes produced a valuation report stating that the fair market value of the property was $6.67 million. This valuation was sent to OST. In early May 1988, Toptown Pty Ltd, the appellant's nominee, applied for, and obtained, a loan from OST of $4.2 million, secured by a mortgage over Glen Crag.
Counts 5 and 13 concerned a property at 50-60 High Street, Southport ("High Street"). On 30 June 1988, Jenkins Development Corporation Pty Ltd, a company controlled by the appellant, agreed to purchase High Street for $1.45 million. On 29 July 1988, Verebes produced a valuation report stating that the current fair market value of High Street, assuming the conversion and redesign of the building, was $8.17 million and that the "as is" value was $6.5 million. This valuation was sent to OST. In August 1988, Jenkins Development Corporation Pty Ltd borrowed $3.8 million from OST, secured by a mortgage over High Street.
On 11 May 1989, Verebes revalued High Street. The new valuation stated that the current fair market value of High Street was $8.186 million and that on completion of refurbishment it would be $9.5 million. This valuation was sent to OST. On 12 May 1989, OST advanced a further $2.61 million which was secured on Glen Crag, High Street, and two other properties.
Count 7 related to the Ashmore Commercial Centre ("Ashmore"). On 18 October 1988, Jenkins Development Corporation Pty Ltd agreed to purchase Ashmore for $4 million. On 22 November 1988, Verebes produced a valuation report stating that the current fair market value of Ashmore was $7 million. This valuation report was sent to OST. On 15 December 1988, Jenkins Development Corporation Pty Ltd borrowed $4.5 million from OST, secured by a mortgage over Ashmore.
Count 14 related to the Dreamworld Theme Park ("Dreamworld") and a group of properties described as the Great Adventures Cairns Portfolio ("Great Adventures"). In 1989, the appellant entered into negotiations to purchase Great Adventures and Dreamworld. They were ultimately purchased for $150 million. On 10 April 1989, Verebes produced a valuation report stating that the fair market value of Dreamworld was $186.5 million. In its final version, this report allocated $179.5 million to land and improvements and $7 million to business and goodwill. On 6 April 1989, Verebes produced a valuation report relating to Great Adventures, stating that its current fair market value was $53.6 million. Part of the Great Adventures Portfolio was the Cairns Ferry Terminal site, which Verebes valued at $12 million. On 4 August 1989, Verebes revalued the site at $36.5 million, increasing the total valuation of Great Adventures to about $78 million. These valuations were sent to OST and became the basis of a guarantee facility of $96 million and a credit facility of $54 million.
The prosecution called independent expert evidence to establish the falsity of all the valuations. It was alleged that the valuation reports were false and misleading in material particulars, that the appellant knew that fact, and that the appellant intended the reports to go to OST in support of his loan applications. The first two of those allegations gave rise to the principal issues relevant to this appeal. The appellant did not give evidence. Bulfin was not called by either side.

The financial needs of the appellant, the acquisitions of land, the borrowings from OST, and the valuations were formally admitted at the trial. The main role of Verebes as a prosecution witness was to explain matters of detail in relation to the valuation reports, and to give evidence of his communications with Bulfin, the appellant and OST. O'Bryan AJA said in the Court of Appeal: "Without his evidence it would have been difficult to admit in evidence the valuation reports, or to explain how they came about".
In relation to each count, the prosecution led expert evidence to prove that the properties were worth much less than the valuations of Verebes showed. The defence case did not accept that evidence. In considering the defence approach to the evidence of Verebes, it is important to bear in mind that the jury did not know, and the defence did not wish it to appear, that Verebes, by his pleas of guilty, had already admitted the falsity of the valuations. Plainly, cross-examining counsel did not want that to emerge. Furthermore, the evidence of Verebes was in many respects supportive of the defence case. There was, therefore, no attack in cross-examination on the credit of Verebes. Rather, the defence case was to challenge the inferences which the prosecution sought to draw from his evidence, and to use his evidence to justify competing inferences. In a few instances, where the evidence of Verebes was difficult to reconcile with a view of the facts for which the defence contended, the approach of cross-examining counsel was, in effect, to put his instructions to Verebes, elicit the response of Verebes, and leave the matter there. It is possible that, at that stage, no decision had been taken as to whether the appellant would give evidence. This did not amount to an attack on the credit of Verebes, with all the dangers that could involve. Not the least of those dangers was the possibility that it could in some way lead to a revelation of Verebes' conviction and sentence. Counsel's approach to the cross-examination of Verebes was one of understandable caution.
In relation to count 2, concerning Glen Crag, the prosecution case was based mainly on the difference between the price for which the property was purchased on 22 April 1988 ($2.41 million) and the amount at which it was valued on 29 April 1988 ($6.67 million). However, Verebes gave evidence about the original inclusion, and later removal at the request of Bulfin, of an "as is" figure, as distinct from a figure assuming complete refurbishment and change of use. He did not say that the appellant spoke to him about that change. He also gave evidence about some rental figures, assuming refurbishment, that he said had been supplied to him by the appellant. That the appellant had supplied these figures was not disputed. In closing address, counsel for the appellant relied strongly on the evidence of Verebes that his instructions, including the instruction to remove the "as is" valuation, came from Bulfin. In relation to this count, the defence case drew support from Verebes.
In relation to counts 5 and 13, once again the crux of the prosecution case was that the appellant could not possibly have believed the High Street property was worth $6.5 million when he was purchasing it for $1.45 million. Again, however, Verebes gave evidence about some matters of detail concerning a reference in his valuation report to a proposed lease by Porsche of part of the premises. He said he obtained this information from the appellant. He could not remember whether he was told that the lease proposal was speculative or definite, but said he would have been unlikely to include reference to it if it had been merely speculative. In cross-examination he accepted that the appellant may have said only that he was hoping to get Porsche as a tenant. Defence counsel relied on that evidence in his final address.
As to count 7, again the prosecution case was that the appellant must have known that a building bought for $4 million would not have a fair market value of $7 million. Rental figures had been supplied by the appellant to Verebes. Verebes said that these figures were given to him as current rental income. The defence case was that the appellant had supplied the figures to Verebes in good faith. In cross-examination Verebes accepted that a valuer ought to check such figures.
As to count 14, the prosecution case about Dreamworld relied on an undisputed fact that Verebes had originally allocated $15 million to goodwill and $171.5 million to land and improvements. This was later altered to $7 million for goodwill and $179.5 million for land and improvements. The evidence showed that OST did not take account of goodwill for the purpose of its loan valuation ratio. Verebes said that he had a conversation with the appellant about the original valuation and that the appellant asked him to speak to Bulfin. It was Bulfin who suggested the adjustment. This evidence was relied upon by defence counsel in address. It was said to show that it was Bulfin, not the appellant, who was directing the valuation process.
In relation to Great Adventures, the ferry terminal site was the focus of attention. Verebes said the appellant drew to his notice some newspaper publicity about the value of a comparable site . This was not challenged in cross-examination, although in address counsel disputed the inference to be drawn from it.
In final address, defence counsel argued that any valuation errors were attributable to Bulfin and Verebes, but not the appellant. He claimed support for this from some of the propositions agreed to by Verebes in cross-examination. In particular, in cross-examination, counsel put to Verebes, and Verebes agreed, that Bulfin was a very important source of work to him, and that Bulfin regularly put at least subtle pressure on him to increase valuations. By the time of final address, the pressure had been magnified considerably.
Because the appellant did not give evidence, and Bulfin was not called as a witness, the defence contention that it was Bulfin, rather than the appellant, who was responsible for any inflation of valuations, and who was the dominant influence behind the valuation process, depended largely upon the evidence of Verebes. Objectively, if highly inflated valuations were being made in respect of properties the appellant had recently bought, or was negotiating to buy, and if those valuations were submitted to OST in support of applications for loans to the appellant and his companies, it was hardly likely that this would have escaped the notice of the appellant. Objectively, it was also improbable that OST would have been forcing upon an unwilling property developer loans in excess of the real value of the mortgaged properties. It is one thing to say that Bulfin was a dishonest and domineering person. It is another thing to say that he was acting as he did independently of the appellant, and solely in his own personal interest. The person with the principal financial interest in all these transactions was the appellant. The defence case depended for such plausibility as it had largely upon the acceptance by Verebes of suggestions put to him in cross-examination about the influence upon him of Bulfin. Without the evidence of Verebes, the jury might well have thought that the defence attempt to isolate the appellant from Bulfin was incredible. Furthermore, since isolating the appellant from Bulfin was a key part of the defence strategy, and since the line taken in cross-examination of Verebes was that he was dominated by Bulfin, the last thing defence counsel would have wanted the jury to hear from the trial judge was that they should or could approach the evidence of Verebes upon the basis that he was an accomplice of the appellant. The defence case was that Verebes was an accomplice of Bulfin, and the appellant was an innocent dupe.
The defence suggestion (accompanied by a denial that the valuations were excessive), that if the valuations were inflated this was purely the result of the influence of Bulfin on Verebes, was not founded upon any attack on the credibility of Verebes. Because it was not supported by any evidence from the appellant himself, it depended upon the evidence of Verebes. Verebes was not challenged in cross-examination by the proposition that he was either consciously or unconsciously seeking to minimise his own responsibility and exaggerate the role of the appellant. On the contrary, he was encouraged in his evidence to emphasise the influence of Bulfin and, to a considerable extent, he responded favourably to that encouragement.
Trial counsel did not ask the judge to give the jury a warning about the evidence of Verebes. The tactical reason for that is clear.

The common law imposes on trial judges a duty to warn of the danger of convicting on evidence which is potentially unreliable[2]. Because of the variety of the circumstances that might exist, in general the law has endeavoured to avoid inflexible rules, and has left it to judges to sum up in the manner best suited to the facts of the particular case[3]. Judicial experience, however, was regarded as identifying certain, limited, kinds of case which required a departure from that general approach, and which called for a particular kind of warning based upon categorisation of a witness rather than upon some particular feature of the circumstances. Relating unreliability to classes of person, rather than to the circumstances of cases, involved stereotyping of a kind which is now out of favour. Rules developed which obliged a trial judge to warn of the danger of convicting upon the uncorroborated evidence of an accomplice, the victim of a sexual offence, and a child. These were regarded as "cases where the evidence suffers from some intrinsic lack of reliability going beyond the mere credibility of a witness"[4]. In 1988, the view was expressed in this Court that those categories ought to be regarded as closed, and that, in general, the interests of justice would benefit from allowing trial judges the freedom to tailor a summing up to the exigencies of the case[5]. In 1982, the Supreme Court of Canada changed the common law in that country to do away with the rigidity of the previous rules[6].
We are here concerned with the first category, accomplices. As to the second and third categories, the subject has now generally been overtaken by legislation. Even in relation to accomplices, in most Australian jurisdictions the common law has been altered substantially by statute[7]. The Commonwealth Evidence Act and the New South Wales Evidence Act, for example, both provide that it is not necessary that a judge warn the jury that it is dangerous to act on uncorroborated evidence, and substitute a more flexible requirement in the case of evidence that may be unreliable, including evidence given by a witness "who might reasonably be supposed to have been criminally concerned in the events giving rise to the proceeding"[8]. It may be noted in passing that the statutory description just quoted was intended to deal with a long-standing question in the common law as to exactly who should be regarded as an accomplice for the purpose of the rule requiring a warning[9]. The concession made by the respondent in this Court relieves us of any need to become involved in that debate. The argument has been conducted on the assumption that Verebes was an accomplice of the appellant.
Another question was whether the common law rule requiring an accomplice warning was a rule of law, or merely a rule of practice. The generally held opinion was that it was once a rule of practice which later hardened into a rule of law[10]. The issue was decided by the House of Lords in England in 1954 in the case of Davies v Director of Public Prosecutions[11], where Lord Simonds LC stated three propositions:

"First proposition: In a criminal trial where a person who is an accomplice gives evidence on behalf of the prosecution, it is the duty of the judge to warn the jury that, although they may convict upon his evidence, it is dangerous to do so unless it is corroborated.

Second proposition: This rule, although a rule of practice, now has the force of a rule of law.

Third proposition: Where the judge fails to warn the jury in accordance with this rule, the conviction will be quashed, even if in fact there be ample corroboration of the evidence of the accomplice, unless the appellate court can apply [the proviso]."

These propositions have generally been understood as an accurate summary of the common law in Australia. But they do not exhaust the subject, and they are not to be treated as if they did.

The fundamental principle is that the trial judge must give the jury proper and sufficient instructions to enable them to decide the real issue or issues in the case[12]. The circumstances of cases are infinitely various. To speak of a "person who is an accomplice giv[ing] evidence on behalf of the prosecution" encompasses many different kinds of circumstances. It includes both the trial where the prosecution's case depends upon the jury accepting the evidence given by an accomplice, and the trial where the prosecution leads no evidence-in-chief from the accomplice but simply makes that person available for cross-examination. It is necessary to take account of this variety.
Where a warning in the terms proposed in Davies is given, the trial judge will have to explain to the jury what is meant by corroboration (not necessarily using that precise term), and direct their attention to evidence that is capable of being regarded as corroborating the accomplice. For this reason, among others, an accomplice warning is not always welcomed by defence counsel. It frequently means that the prosecution argument will be reinforced by a recitation by the trial judge of all the evidence that is capable of corroborating the accomplice. That, no doubt, is one reason why the Commonwealth and New South Wales Evidence Acts, in their provisions relating to judicial warnings about unreliable evidence, attach significance to whether there is a request for such a warning[13].
In the practical application of the rule, it must be observed that what is involved is a warning about the danger of "convicting upon the evidence" of the accomplice, unless it is corroborated. This is not a mere incantation. It must be related to a forensic contest, and its operation in a particular case must be explained to the jury by reference to the evidence and the issues. Those issues will be determined partly by the conduct of trial counsel, and might not emerge in final form until the closing addresses. The rule exists for a reason. That reason is related to the potential unreliability of accomplices, an unreliability thought to be so well known in the experience of courts that judges are required, not merely to point it out to jurors, but to tell them that it would be dangerous to convict upon the evidence of an accomplice unless it is corroborated. The principal source of unreliability, although it may be compounded by the circumstances of a particular case, is what is regarded as the natural tendency of an accomplice to minimise the accomplice's role in a criminal episode, and to exaggerate the role of others, including the accused. Accomplices are regarded by the law as a notoriously unreliable class of witness, having a special lack of objectivity. The warning to the jury is for the protection of the accused. The theory is that fairness of the trial process requires it. It is a warning that is to be related to evidence upon which the jury may convict the accused. The reference to danger is to be accompanied by a reference to a need to look for corroboration. The hypothesis is that the evidence in question is in contest, and that it inculpates the accused. Why would a jury be directed to consider whether there is corroboration of evidence that is not in dispute? Why would a jury be directed to consider whether there is corroboration of evidence upon which the accused seeks to rely? What is the danger involved in acting on evidence which is accepted by both parties?
Following Davies, it was regarded in England as self-evident that the law only requires a direction on corroboration in relation to evidence of an accomplice that is adverse to an accused[14]. Nor is a direction required in relation to undisputed evidence. It often happens in a criminal trial that a witness who is technically an accomplice is called to give evidence of some fact which is not formally admitted but which, once proved, is not challenged by the defence. Judges are not obliged to warn of a supposed danger of accepting such evidence. It is an aspect of the adversarial system of criminal justice that an accused person may put the prosecution to proof of facts yet, once some evidence of those facts is given, it will emerge that they are not then disputed. If such evidence is given by an accomplice, but is unchallenged, then there may be no occasion for an accomplice warning. In some cases, an accomplice might give evidence, some of which is in contest and some of which is not. The direction would then need to be related to the disputed evidence. Characterising evidence as favourable or unfavourable to an accused may not always be easy. On the other hand, there may be cases where the prosecution, in performance of its duty of fairness, calls an accomplice whose evidence is wholly favourable to, and accepted by, an accused. It would be absurd, and contrary to the rationale of the rule, to require the trial judge, in such a case to give an accomplice warning, sending the jury off on a search for corroboration of evidence on which the accused relies.
Although the common law rule about accomplice warnings is a rule of law, and although (subject to the proviso) in the ordinary case the requirement for a warning does not depend upon a request being made by trial counsel, the rule is not so mechanical as to call for a warning in any case in which an accomplice gives any evidence which may be relied upon to establish the prosecution case. The application of the rule must be related to its purpose, and will require a consideration of the issues as they have emerged from the way in which the case has been conducted.
If there is an issue which the jury might have to resolve in order to reach a verdict of guilty, and an accomplice's evidence relates to that issue, an accomplice warning must be given if the acceptance of that evidence is or could be a step taken by the jury in reasoning to a finding of guilt. Ordinarily it would be expected that the use to which the accomplice's evidence may be put will be apparent from the examination or cross-examination of the accomplice, or at least from what is said in closing addresses to the jury. That course of evidence, or the addresses, will reveal whether accepting the accomplice's evidence could be a step which the jury would take along the path to a guilty verdict. If the evidence of an accomplice is not controverted, there will be no issue to which the accomplice's evidence relates and which the jury will, or may, resolve in reasoning to a verdict of guilty. In that latter case no accomplice warning will be necessary. It will not be necessary because there is no issue for the jury to decide to which the instruction could relate.
The present was a case in which the primary facts established by the testimony of the accomplice were substantially undisputed. Counsel for the appellant elicited from him, in cross-examination, additional facts which formed a basis of the ultimate defence case, and without which the defence case, being unsupported by any evidence of the appellant, would have been hopeless. It is not possible to point to any particular part of the evidence of Verebes concerning which the jury reasonably might have been told they ought to look for corroboration. Each side, in argument, sought to draw competing inferences from the evidence of Verebes. Each side emphasised some aspects of his evidence rather than others. On the prosecution case, virtually the whole of the remainder of the evidence, including the admitted facts, supported the inferences available from the evidence of Verebes and, if there had been a specific challenge to any significant part of his evidence, would have supported his version of events. However, the primary facts that emerged from the evidence of Verebes, as has been said, were largely undisputed. In the circumstances, an accomplice warning would have been inappropriate, and was not required.
In the alternative, it was submitted on behalf of the appellant that, even if the case did not call for a warning that it would be dangerous to convict on the evidence of Verebes unless it was corroborated, at least there should have been a warning of the potential unreliability of the evidence and of the need to subject it to careful scrutiny. The answer to that submission, once again, is related to the way in which the trial was conducted, the issues as they emerged, and the use that both the prosecution and the appellant sought to make of the evidence of Verebes. If there had been a substantial contest as to the primary facts which Verebes was called to prove, then there may have been force in the argument. That, however, is not the way in which the trial was conducted. A clear indication of that is the failure of defence counsel to seek any warning of the kind now suggested. Counsel could have sought a warning which did not identify Verebes as someone who might be an accomplice, or involve the trial judge in reciting the evidence that could be regarded as corroborative. Counsel's failure to do so was consistent with the tactical approach taken by the defence towards the evidence of Verebes. As the issues emerged at the trial, the judge was not obliged to give a warning.

No error has been shown. It is not necessary to consider the proviso. The appeal should be dismissed.

Pollitt v The Queen (1992) 174 CLR 558 at 599 per Dawson and Gaudron JJ. ↑

Carr v The Queen (1988) 165 CLR 314 at 318-319 per Wilson and Dawson JJ. ↑

Carr v The Queen (1988) 165 CLR 314 at 319 per Wilson and Dawson JJ. ↑

Carr v The Queen (1988) 165 CLR 314 at 319 per Wilson and Dawson JJ. ↑

eg Evidence Act 1995 (Cth), s 164, Evidence Act 1995 (NSW), s 164, Evidence Act 1906 (WA), s 50 (inserted in 1988), Criminal Code Act 1899 (Qld), s 632 (inserted in 1997). See also Criminal Justice and Public Order Act 1994 (UK), s 32. ↑

Evidence Act 1995 (Cth), s 165(1)(d), Evidence Act 1995 (NSW), s 165(1)(d). ↑

Tripodi v The Queen (1961) 104 CLR 1 at 9 per Dixon CJ, Fullagar and Windeyer JJ. The history of the rules about corroboration warnings was discussed by Jacobs J in Kelleher v The Queen (1974) 131 CLR 534 at 564-569 and Ormiston J in R v Rosemeyer [1985] VR 945 at 960-971. ↑

Evidence Act 1995 (Cth), s 165; Evidence Act 1995 (NSW), s 165. ↑

R v Royce-Bentley [1974] 1 WLR 535; [1974] 2 All ER 347. ↑

As to count 7, again the prosecution case was that the appellant must have known that a building bought for $4 million would not have a fair market value of $7 million. Rental figures had been supplied by the appellant to Verebes. Verebes said that these figures were given to him as current rental income. The defence case was that the appellant had supplied the figures to Verebes in good faith. In cross-examination Verebes accepted that a valuer ought to check such figures.

As to count 14, the prosecution case about Dreamworld relied on an undisputed fact that Verebes had originally allocated $15 million to goodwill and $171.5 million to land and improvements. This was later altered to $7 million for goodwill and $179.5 million for land and improvements. The evidence showed that OST did not take account of goodwill for the purpose of its loan valuation ratio. Verebes said that he had a conversation with the appellant about the original valuation and that the appellant asked him to speak to Bulfin. It was Bulfin who suggested the adjustment. This evidence was relied upon by defence counsel in address. It was said to show that it was Bulfin, not the appellant, who was directing the valuation process.

In relation to Great Adventures, the ferry terminal site was the focus of attention. Verebes said the appellant drew to his notice some newspaper publicity about the value of a comparable site . This was not challenged in cross-examination, although in address counsel disputed the inference to be drawn from it.

In final address, defence counsel argued that any valuation errors were attributable to Bulfin and Verebes, but not the appellant. He claimed support for this from some of the propositions agreed to by Verebes in cross-examination. In particular, in cross-examination, counsel put to Verebes, and Verebes agreed, that Bulfin was a very important source of work to him, and that Bulfin regularly put at least subtle pressure on him to increase valuations. By the time of final address, the pressure had been magnified considerably.

Because the appellant did not give evidence, and Bulfin was not called as a witness, the defence contention that it was Bulfin, rather than the appellant, who was responsible for any inflation of valuations, and who was the dominant influence behind the valuation process, depended largely upon the evidence of Verebes. Objectively, if highly inflated valuations were being made in respect of properties the appellant had recently bought, or was negotiating to buy, and if those valuations were submitted to OST in support of applications for loans to the appellant and his companies, it was hardly likely that this would have escaped the notice of the appellant. Objectively, it was also improbable that OST would have been forcing upon an unwilling property developer loans in excess of the real value of the mortgaged properties. It is one thing to say that Bulfin was a dishonest and domineering person. It is another thing to say that he was acting as he did independently of the appellant, and solely in his own personal interest. The person with the principal financial interest in all these transactions was the appellant. The defence case depended for such plausibility as it had largely upon the acceptance by Verebes of suggestions put to him in cross-examination about the influence upon him of Bulfin. Without the evidence of Verebes, the jury might well have thought that the defence attempt to isolate the appellant from Bulfin was incredible. Furthermore, since isolating the appellant from Bulfin was a key part of the defence strategy, and since the line taken in cross-examination of Verebes was that he was dominated by Bulfin, the last thing defence counsel would have wanted the jury to hear from the trial judge was that they should or could approach the evidence of Verebes upon the basis that he was an accomplice of the appellant. The defence case was that Verebes was an accomplice of Bulfin, and the appellant was an innocent dupe.

The defence suggestion (accompanied by a denial that the valuations were excessive), that if the valuations were inflated this was purely the result of the influence of Bulfin on Verebes, was not founded upon any attack on the credibility of Verebes. Because it was not supported by any evidence from the appellant himself, it depended upon the evidence of Verebes. Verebes was not challenged in cross-examination by the proposition that he was either consciously or unconsciously seeking to minimise his own responsibility and exaggerate the role of the appellant. On the contrary, he was encouraged in his evidence to emphasise the influence of Bulfin and, to a considerable extent, he responded favourably to that encouragement.

Trial counsel did not ask the judge to give the jury a warning about the evidence of Verebes. The tactical reason for that is clear.

The common law imposes on trial judges a duty to warn of the danger of convicting on evidence which is potentially unreliable[2]. Because of the variety of the circumstances that might exist, in general the law has endeavoured to avoid inflexible rules, and has left it to judges to sum up in the manner best suited to the facts of the particular case[3]. Judicial experience, however, was regarded as identifying certain, limited, kinds of case which required a departure from that general approach, and which called for a particular kind of warning based upon categorisation of a witness rather than upon some particular feature of the circumstances. Relating unreliability to classes of person, rather than to the circumstances of cases, involved stereotyping of a kind which is now out of favour. Rules developed which obliged a trial judge to warn of the danger of convicting upon the uncorroborated evidence of an accomplice, the victim of a sexual offence, and a child. These were regarded as "cases where the evidence suffers from some intrinsic lack of reliability going beyond the mere credibility of a witness"[4]. In 1988, the view was expressed in this Court that those categories ought to be regarded as closed, and that, in general, the interests of justice would benefit from allowing trial judges the freedom to tailor a summing up to the exigencies of the case[5]. In 1982, the Supreme Court of Canada changed the common law in that country to do away with the rigidity of the previous rules[6].

We are here concerned with the first category, accomplices. As to the second and third categories, the subject has now generally been overtaken by legislation. Even in relation to accomplices, in most Australian jurisdictions the common law has been altered substantially by statute[7]. The Commonwealth Evidence Act and the New South Wales Evidence Act, for example, both provide that it is not necessary that a judge warn the jury that it is dangerous to act on uncorroborated evidence, and substitute a more flexible requirement in the case of evidence that may be unreliable, including evidence given by a witness "who might reasonably be supposed to have been criminally concerned in the events giving rise to the proceeding"[8]. It may be noted in passing that the statutory description just quoted was intended to deal with a long-standing question in the common law as to exactly who should be regarded as an accomplice for the purpose of the rule requiring a warning[9]. The concession made by the respondent in this Court relieves us of any need to become involved in that debate. The argument has been conducted on the assumption that Verebes was an accomplice of the appellant.

Another question was whether the common law rule requiring an accomplice warning was a rule of law, or merely a rule of practice. The generally held opinion was that it was once a rule of practice which later hardened into a rule of law[10]. The issue was decided by the House of Lords in England in 1954 in the case of Davies v Director of Public Prosecutions[11], where Lord Simonds LC stated three propositions:

"First proposition: In a criminal trial where a person who is an accomplice gives evidence on behalf of the prosecution, it is the duty of the judge to warn the jury that, although they may convict upon his evidence, it is dangerous to do so unless it is corroborated. > > Second proposition: This rule, although a rule of practice, now has the force of a rule of law. > > Third proposition: Where the judge fails to warn the jury in accordance with this rule, the conviction will be quashed, even if in fact there be ample corroboration of the evidence of the accomplice, unless the appellate court can apply [the proviso]."

These propositions have generally been understood as an accurate summary of the common law in Australia. But they do not exhaust the subject, and they are not to be treated as if they did.

The fundamental principle is that the trial judge must give the jury proper and sufficient instructions to enable them to decide the real issue or issues in the case[12]. The circumstances of cases are infinitely various. To speak of a "person who is an accomplice giv[ing] evidence on behalf of the prosecution" encompasses many different kinds of circumstances. It includes both the trial where the prosecution's case depends upon the jury accepting the evidence given by an accomplice, and the trial where the prosecution leads no evidence-in-chief from the accomplice but simply makes that person available for cross-examination. It is necessary to take account of this variety.

Where a warning in the terms proposed in Davies is given, the trial judge will have to explain to the jury what is meant by corroboration (not necessarily using that precise term), and direct their attention to evidence that is capable of being regarded as corroborating the accomplice. For this reason, among others, an accomplice warning is not always welcomed by defence counsel. It frequently means that the prosecution argument will be reinforced by a recitation by the trial judge of all the evidence that is capable of corroborating the accomplice. That, no doubt, is one reason why the Commonwealth and New South Wales Evidence Acts, in their provisions relating to judicial warnings about unreliable evidence, attach significance to whether there is a request for such a warning[13].

In the practical application of the rule, it must be observed that what is involved is a warning about the danger of "convicting upon the evidence" of the accomplice, unless it is corroborated. This is not a mere incantation. It must be related to a forensic contest, and its operation in a particular case must be explained to the jury by reference to the evidence and the issues. Those issues will be determined partly by the conduct of trial counsel, and might not emerge in final form until the closing addresses. The rule exists for a reason. That reason is related to the potential unreliability of accomplices, an unreliability thought to be so well known in the experience of courts that judges are required, not merely to point it out to jurors, but to tell them that it would be dangerous to convict upon the evidence of an accomplice unless it is corroborated. The principal source of unreliability, although it may be compounded by the circumstances of a particular case, is what is regarded as the natural tendency of an accomplice to minimise the accomplice's role in a criminal episode, and to exaggerate the role of others, including the accused. Accomplices are regarded by the law as a notoriously unreliable class of witness, having a special lack of objectivity. The warning to the jury is for the protection of the accused. The theory is that fairness of the trial process requires it. It is a warning that is to be related to evidence upon which the jury may convict the accused. The reference to danger is to be accompanied by a reference to a need to look for corroboration. The hypothesis is that the evidence in question is in contest, and that it inculpates the accused. Why would a jury be directed to consider whether there is corroboration of evidence that is not in dispute? Why would a jury be directed to consider whether there is corroboration of evidence upon which the accused seeks to rely? What is the danger involved in acting on evidence which is accepted by both parties?

Following Davies, it was regarded in England as self-evident that the law only requires a direction on corroboration in relation to evidence of an accomplice that is adverse to an accused[14]. Nor is a direction required in relation to undisputed evidence. It often happens in a criminal trial that a witness who is technically an accomplice is called to give evidence of some fact which is not formally admitted but which, once proved, is not challenged by the defence. Judges are not obliged to warn of a supposed danger of accepting such evidence. It is an aspect of the adversarial system of criminal justice that an accused person may put the prosecution to proof of facts yet, once some evidence of those facts is given, it will emerge that they are not then disputed. If such evidence is given by an accomplice, but is unchallenged, then there may be no occasion for an accomplice warning. In some cases, an accomplice might give evidence, some of which is in contest and some of which is not. The direction would then need to be related to the disputed evidence. Characterising evidence as favourable or unfavourable to an accused may not always be easy. On the other hand, there may be cases where the prosecution, in performance of its duty of fairness, calls an accomplice whose evidence is wholly favourable to, and accepted by, an accused. It would be absurd, and contrary to the rationale of the rule, to require the trial judge, in such a case to give an accomplice warning, sending the jury off on a search for corroboration of evidence on which the accused relies.

Although the common law rule about accomplice warnings is a rule of law, and although (subject to the proviso) in the ordinary case the requirement for a warning does not depend upon a request being made by trial counsel, the rule is not so mechanical as to call for a warning in any case in which an accomplice gives any evidence which may be relied upon to establish the prosecution case. The application of the rule must be related to its purpose, and will require a consideration of the issues as they have emerged from the way in which the case has been conducted.

If there is an issue which the jury might have to resolve in order to reach a verdict of guilty, and an accomplice's evidence relates to that issue, an accomplice warning must be given if the acceptance of that evidence is or could be a step taken by the jury in reasoning to a finding of guilt. Ordinarily it would be expected that the use to which the accomplice's evidence may be put will be apparent from the examination or cross-examination of the accomplice, or at least from what is said in closing addresses to the jury. That course of evidence, or the addresses, will reveal whether accepting the accomplice's evidence could be a step which the jury would take along the path to a guilty verdict. If the evidence of an accomplice is not controverted, there will be no issue to which the accomplice's evidence relates and which the jury will, or may, resolve in reasoning to a verdict of guilty. In that latter case no accomplice warning will be necessary. It will not be necessary because there is no issue for the jury to decide to which the instruction could relate.

The present was a case in which the primary facts established by the testimony of the accomplice were substantially undisputed. Counsel for the appellant elicited from him, in cross-examination, additional facts which formed a basis of the ultimate defence case, and without which the defence case, being unsupported by any evidence of the appellant, would have been hopeless. It is not possible to point to any particular part of the evidence of Verebes concerning which the jury reasonably might have been told they ought to look for corroboration. Each side, in argument, sought to draw competing inferences from the evidence of Verebes. Each side emphasised some aspects of his evidence rather than others. On the prosecution case, virtually the whole of the remainder of the evidence, including the admitted facts, supported the inferences available from the evidence of Verebes and, if there had been a specific challenge to any significant part of his evidence, would have supported his version of events. However, the primary facts that emerged from the evidence of Verebes, as has been said, were largely undisputed. In the circumstances, an accomplice warning would have been inappropriate, and was not required.

In the alternative, it was submitted on behalf of the appellant that, even if the case did not call for a warning that it would be dangerous to convict on the evidence of Verebes unless it was corroborated, at least there should have been a warning of the potential unreliability of the evidence and of the need to subject it to careful scrutiny. The answer to that submission, once again, is related to the way in which the trial was conducted, the issues as they emerged, and the use that both the prosecution and the appellant sought to make of the evidence of Verebes. If there had been a substantial contest as to the primary facts which Verebes was called to prove, then there may have been force in the argument. That, however, is not the way in which the trial was conducted. A clear indication of that is the failure of defence counsel to seek any warning of the kind now suggested. Counsel could have sought a warning which did not identify Verebes as someone who might be an accomplice, or involve the trial judge in reciting the evidence that could be regarded as corroborative. Counsel's failure to do so was consistent with the tactical approach taken by the defence towards the evidence of Verebes. As the issues emerged at the trial, the judge was not obliged to give a warning.

No error has been shown. It is not necessary to consider the proviso. The appeal should be dismissed.

[2] Pollitt v The Queen [1992] HCA 35; (1992) 174 CLR 558 at 599 per Dawson and Gaudron JJ.

[3] Carr v The Queen [1988] HCA 47; (1988) 165 CLR 314 at 318-319 per Wilson and Dawson JJ.

[4] Carr v The Queen [1988] HCA 47; (1988) 165 CLR 314 at 319 per Wilson and Dawson JJ.

[5] Carr v The Queen [1988] HCA 47; (1988) 165 CLR 314 at 319 per Wilson and Dawson JJ.

[6] Vetrovec v The Queen [1982] 1 SCR 810.

[7] eg Evidence Act 1995 (Cth), s 164, Evidence Act 1995 (NSW), s 164, Evidence Act 1906 (WA), s 50 (inserted in 1988), Criminal Code Act 1899 (Qld), s 632 (inserted in 1997). See also Criminal Justice and Public Order Act 1994 (UK), s 32.

[9] See, for example, McNee v Kay [1953] VicLawRp 71; [1953] VLR 520.

[10] Tripodi v The Queen [1961] HCA 22; (1961) 104 CLR 1 at 9 per Dixon CJ, Fullagar and Windeyer JJ. The history of the rules about corroboration warnings was discussed by Jacobs J in Kelleher v The Queen [1974] HCA 48; (1974) 131 CLR 534 at 564-569 and Ormiston J in R v Rosemeyer [1985] VicRp 90; [1985] VR 945 at 960-971.

[12] Alford v Magee [1952] HCA 3; (1952) 85 CLR 437 at 466.

[14] R v Royce-Bentley [1974] 1 WLR 535; [1974] 2 All ER 347.

Under the Friendly Societies Act 1986(Vic)

1988), Criminal Code Act 1899(Qld)

Justice and Public Order Act 1994

At a glance

Court

Decision date

Before

Catchwords

Source

Catchwords

Judgment (56 paragraphs)

Parties

Legislation Cited (8)

Cases Cited (6)