THE QUEEN v. PETER LI KWOK SUI AND ANOTHER

------------------

JUDGMENT

------------------

Roberts, C.J. -

1. On 13th June, 1984,Petgr LI Kwok-sui (D.1) and Paul CHAN Shiu-hung (D.2) were found guilty by a jury on 7 counts of conspiracy to defraud (Counts 1, 2, 3, 4, 5, 7 and 8). Peter LI Kwok-sui was also found guilty on 2 counts of accepting an advantage as an agent (Counts 12 and 14). Both seek leave to appeal against conviction.

Burdeii of proof

2. The first main ground of appeal was that the judge wrongly directed the jury, as to standard of proof which must be satisfied by the Crown, in the following passage in his summing-up -

In subsequent passages Baber, J. repeated his direction that, before bringing in a conviction, the jury had to be "satisfied of guilt beyond all reasonable doubt".

3. In R. v. YEUNG Kuen-chi and CHEUK Tung-shing⁽¹⁾ , a division of this Court, quashed a conviction because the summing-up contained this passage -

"There is a burden on the Crown to prove every element of these charges and the Crown must do so beyond a reasonable doubt. He (sic) cannot do so to a hundred per cent certainty. That is not possible, but you must, in other words, be sure."

4. We do not seek to distinguish the words "with mathematical certainty" (whatever that phrase may mean) from "a hundred per cent certainty". Thus, on the authority of Yeng⁽¹⁾ , we must conclude that the phrase ought not to have been used.

5. However, this is not an end of the matter. As the court in Yeng⁽¹⁾ said, the summing-up has to be read as a whole, in order to ascertain whether the jury could be left in any doubt as to the standard of proof which the Crown must attain.

6. In Walters v. R.⁽²⁾ , which was not cited to the court in Yeng⁽¹⁾, Lord Diplock quoted a passage from the judgment of Lord Goddard, L.C.J. in an earlier case -

"It is not the particular formula that matters: it is the effect of the summing-up. If the jury are made to understand that they have to be satisfied and must not return a verdict against a defendant unless they feel sure, and that the onus is all the time on the prosecution and not on the defence, then whether the judge uses one form of language or another is neither here nor there."

7. Lord Diplock went on -

"By the time he sums up the judge at the trial has had an opportunity of observing the jurors. In their Lordship s view it is best left to his discretion to choose the most appropriate set of words in which to make that jury understand that they must not return a verdict against a defendant unless they are sure of his guilt; - Their Lordships would deprecate any attempt to lay down some precise formula or to draw fine distinctions between one set of words and another. It is the effect of the summing-up as a whole that matters."

Looking at the overall effect of this summing-up, we are satisfied that the jury must fully have understood that the standard of proof required of them was that they should not return a verdict against either defendant on any count unless they were sure of guilt. For this reason, we feel able to distinguish this case from Yeng⁽¹⁾, which does not go as far as to decide that it is necessary to upset a conviction, because unfortunate expressions were at some stage used, so long as the thrust of the direction as a whole is correct and unambiguous.

8. Although, therefore, it could be said that the passage in the summing-up which referred to "mathematical certainty" taken by itself, might have misled the jury, we are satisfied that the direction, read as a whole, left the jury in no doubt as to the correct standard of proof.

Duplicity of conspiracy counts

9. The first count charged the defendants with conspiracy to defraud, contrary to Common Law. The particulars (paraphrased) alleged that they, between 1st April and 3rd December, 1982, conspired together to defraud Lloyds Bank International Limited, by dishonestly -

10. The second count charged the defendants with conspiring to defraud the bank between 15th August and 3rd December, 1982 by dishonestly -

11. Counts 3, 4, 5, 6, 7 and 8 were in similar form, each relating to a different conspiracy and with an additional and different conspirator joined in each count (two in the case of count 8).

12. It was suggested that there was duplicity as between count 1 and counts 2 to 8 inclusive, in the sense that counts 2 to 8 all occurred at different dates between 1st April, 1982 and 3rd December, 1982, the same period as that comprehended in count 1.

13. In our opinion this is not a case of duplicity since, in ascertaining whether a conspiracy count is bad for this reason, it is unnecessary to look beyond the particulars of the count itself.

14. Thus each count should be examined separately, to see whether, by its terms, more than one offence has been charged.

15. The Crown adopted, as a precedent for the drafting of its charges, the formula suggested in R. v. Landy⁽³⁾ . In this precedent were set out, in considerable detail, the various methods by which the defendants- intended to carry out their agreement to defraud the bank.

16. Difficult though the language is to follow in places, we are satisfied that each count, taken by itself, charged only one offence, namely an agreement to defraud the bank, by one or more of the methods specified in the particulars.

17. As an alternative, it was submitted that only one conspiracy (count 1) should have been charged, since all the other conspiracy counts were no more than instances of overt acts giving effect to the main conspiracy charged in count 1.

18. It is, however, well established that the charging of a broad general conspiracy does not preclude the existence of sub-conspiracies, which can properly be charged as separate offences - see R. v. Coughlan and Young⁽⁴⁾ , in which it was held that a broad general agreement, to cause explosions in the United Kingdom, did not preclude the existence of sub-conspiracies to cause explosions in particular places.

19. We thus see nothing wrong in charging the two defendants with a broad general conspiracy to defraud the bank, by various means, as the opportunity arose, and to combine this in the same indictment with other counts alleging later agreements to defraud the bank in respect of separate and identified transactions.

Late Direction

20. The very detailed particulars given in the conspiracy counts, however, seem to have caused difficulties for the jury, which returned, after deliberating for five and three quarter hours, with this question -

"Is it necessary for all the points outlined in a count to be proven before the accused are judged guilty or does one of the points being proven, then suffice for a conviction?"

To which the judge replied -

To which the Foreman answered -

"Yes, it does, thank you."

The jury returned with its verdict about fifteen minutes' later.

21. When summing-up the judge had properly told the jury -

"The crime of conspiracy is an unusual one. Most crimes are concerned with the doing of something- something physical: killing somebody, burning a house down, robbing somebody, things like that. But the crime of conspiracy is complete once there is a concluded agreement to do something unlawful or to do something lawful but in an unlawful way. The essence of the crime is that there is an agreement, a meeting of minds. It is not necessary that what is conspired to do is ever done."

He used words to like effect later, and his directions as to the law on conspiracy were not challenged.

22. Counsel for the appellants argued, nevertheless, that the jury were misled by this final direction, in which they were told that they could convict if any one of the methods set out in each count of conspiracy was established beyond reasonable doubt. It was urged that such a direction was calculated to cause confusion, between the means by which a fraudulent agreement was put into execution and the necessary finding of a fraudulent agreement itself.

23. However, the judge, in the same part of his direction, reminded them that the important feature of each count was the "agreement, a meeting of minds to practise fraud". It must have been clear to the jury that they were being told that they must be satisfied of this in any event, but that they could only convict if any of the methods alleged was also proved to their satisfaction. We do not think that the judge's postscript amounted to a misdirection.

Insufficient direction

24. The trial lasted for 43 days, of which 17 were occupied by a voir dire and 3 by the speeches of counsel. At the conclusion of this, the judge, after an interval of 6 days, summed up in less than 2 hours. "It might well take a week if I try even to summarize it," he said. "I do not intend to do that in any detail."

25. Unfortunately, in relation to counts 1, 7, 8, 12 and 14, we think that he went too far in his understandable wish to keep his summing-up within the span of concentration to be expected of the jury. Apart from his general direction as to the law of conspiracy, his only remarks on counts 1, 12 and 14 were as follows -

26. The direction on count 7 consisted of seven lines; that on count 8 of six - as follows -

27. The judge's understandable concern for brevity resulted, in our opinion, in a failure to discharge the obligation which rested upon him, to outline the relevant issues of law and fact in relation to these counts in the manner described by Lord Hailsham, L.C. in R. v. Lawrence⁽⁶⁾

"A direction to the jury - should- include a succinct but accurate summary of the issues of fact as to which a decision is required, a correct but concise summary of the evidence and arguments on both sides, and a correct statement of the inferences which the jury are entitled to draw from their particular conclusions about the primary facts."

28. In the light of the scant references to them in the summing-up, we cannot feel sure that the jury properly applied their minds to the issues arising in relation to counts 1, 7, 8, 12 and 14; and therefore quash the convictions on them.

Insufficient direction on accomplices

29. There were several complaints with regard to the judge's direction on the evidence of accomplices -

30. The judge's main direction as to accomplices read as follows -

31. With regard to (a), the judge did not give such a direction. As Lord Hailsham, L.C. observed in D.P.P. v. Kilbourne⁽⁷⁾ -

"Corroboration is only required or afforded if the witness requiring corroboration or giving it is otherwise credible."

32. Lord Hailsham reinforced this view of the law in D.P.P. v. Boardman⁽⁸⁾ -

"What I said (in Kilbourne) was that unless a witness' evidence was intrinsically credible he could neither? afford corroboration nor be thought to require it.  In such cases, the witness' evidence is rejected before the question of corroboration arises."

33. The judge did not put it to the jury in this manner, which is the better way. But he did tell them that they were entitled to say "we believe what this accomplice has told us sufficiently to convict". This was followed by a reference to the danger of convicting on such evidence alone.

34. Taking the passage as a whole, we consider that the jury would have realized that the question of corroboration only arose if they thought the accomplice's evidence to be credible.

35. So far as (b) is concerned, we consider that the direction was sufficient, since the judge told the jury that "there certainly is danger in convicting on the uncorroborated evidence of a self-confessed accomplice," although the words "self-confessed" are better avoided, since many accomplices cannot be so classified.

36. As to (c), no such warning was given, but we do not think that this was necessary in view of the judge's general warning that the case against each accused must be treated separately, as must the evidence on each of the charges.

38. It is not disputed that there was no guidance to the jury as to whether or not there was evidence capable of amounting to corroboration of the evidence of any of the accomplices, on whose testimony the convictions on all counts largely depended.

39. It is well established that, where evidence requires corroboration, it is not sufficient merely to warn the jury of the danger of conviction on the basis of accomplice evidence unless it is confirmed. The jury must also be directed as to whether there is, or is not, evidence capable of confirming the "suspect" evidence.

40. If there is evidence which is so capable, the jury must be directed that the confirming evidence must itself be credible evidence. Just as a witness who is not credible cannot be corroborated, so evidence which is not itself credible cannot serve as corroboration.

41. It is a matter of some difficulty to distinguish what evidence may or may not amount to corroboration. The practice has therefore developed, to a point at which it has become a rule of law, that the judge should indicate to the jury the nature of evidence which could be used for the purpose of corroboration.

42. It is unfortunate that the judge made no attempt to direct the jury as to this. Having told them that there is "danger in convicting on the uncorroborated evidence of a self -confessed accomplice", he does not refer again to the question of corroboration, save that he warned them that one accomplice cannot corroborate another and told them that corroboration of the evidence of the persons offering bribes under the corruption counts is unnecessary.

43. The jury were thus left without assistance as to whether or not there was any evidence which was capable of corroborating that of the various accomplices.

44. The jury acquitted the defendants of count 6, the evidence on which depended on documents and on inferences to be drawn from them, since the accomplice named in the count was dead at the time of the trial.

45. This suggests that the jury were not prepared to convict on the documentary evidence alone, since that available on count 6 was of a similar strength, and showed a similar system, to that adduced on counts 2, 3, 4, 5, 7 and 8, of which the defendants were convicted.

46. The convictions on the other conspiracy counts are thus likely to have been based mainly on an acceptance of the evidence of accomplice. But because they had no direction as to what could amount to corroboration, it is impossible to do more than speculate as to what evidence, if any, the jury took into account as corroborative of the accomplices. They may well have regarded as corroboration matters which could not properly be so regarded and thus have decided that it was safe to convict on the basis that the accomplice was corroborated when he was not.

47. The Crown argued that the documentary evidence provided substantial corroboration on all charges. Even if this is so, we cannot feel that the jury  if properly directed, would necessarily have come to the same conclusion, a doubt strengthened by their rejection of the documentary evidence on count 6. We are therefore not able to apply the proviso in relation to the remaining conspiracy counts and must quash the convictions on counts 2, 3, 4 and 5.

Conclusion

48. The judge has our sympathy in his laudable attempt to summarize the evidence, though we have noted that in some respects his summing-up was deficient. His difficulty, and that of the jury, in making sense of complicated commercial and banking transactions, was aggravated both by the number of conspiracy counts; and by the addition of bribery counts as well. It is another illustration of Moult's Law, which holds that the more charges are joined in one indictment, the greater are the chances of acquittal on all of them.

49. We are satisfied, however, that there was a substantial case against both defendants on the conspiracy charges and that the interests of justice require that we should order, as we do, that the defendants be retried on a new indictment on counts 1, 2, 3, 4, 5, 7 and 8.

(1)    Crim. App. No. 266/84 (unreported)

(2)    [1969] 2 A.C. 26 (P.C.) at p.30

(3)    [1981] 72 C.A.R. 237.

(4)    [1976] 63 Cr.App.R. 33.

(6)    [1982] A.C. 510 at p.519

(7)    [1973] 57 Cr.App.R. 381 H.L. at p.402

(8)    [1975] 60 Cr.App.R. 165 at p.183