HKSAR v. AU YEUNG CHI HO FRANKY

CACC231/2009

IN THE HIGH COURT OF THE

HONG KONG SPECIAL ADMINISTRATIVE REGION

COURT OF APPEAL

CRIMINAL APPEAL NO. 231 OF 2009

(ON APPEAL FROM DCCC 784 OF 2008)

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

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

REASONS FOR JUDGMENT

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

Hon Lunn J (giving the Reasons for Judgment of the Court) :

1.  On 12 November 2010 we dismissed the application made on behalf of the applicant for leave to appeal against conviction after trial on two charges by Deputy Judge Dufton in the District Court on 24 June 2009 and said that we would give our reasons for so doing in due course. This we do now.

THE CHARGES

2.  By charge 1, it was alleged that between June 2004 and June 2006 the applicant had conspired with Li May Tak (“Li”) to defraud Gammon Construction Limited (“Gammon”), contrary to Common Law.  The ‘Particulars of Offence’, alleged that the applicant had conspired with Li to defraud Gammon :

“... by dishonestly providing false measurements of elongation of stressed tendons during tendon stressing process, thereby causing Gammon its employees or agents to accept the said measurements as accurate and not to take any actions (including taking any rectification measures) in respect of the tendons or any structure relating thereto.”

3.  By charge 3, it was alleged that between March and August 2006 the applicant had committed fraud, contrary to section 16A of the Theft Ordinance, Cap. 210.  The ‘Particulars of Offence’ alleged that the applicant had committed the offence by deceitfully :

“... providing false measurements of elongation of stressed tendons through Tsoi Kuk Yau during the tendon stressing process, and with intent to defraud … Gammon, its employees or agents, to do an act and make an omission, namely accepting the said measurements as accurate and not taking any actions (including taking any rectification measures) in respect of the tendons or any structure related thereto, resulting in prejudice or a substantial risk of prejudice to any person other than Au Yeung Chi-ho, Franky.”

THE TRIAL

4.  The applicant, Li and Tsoi Kuk Yau (“Tsoi”) were all employees at the material time of Alga (Far East) Limited (“Alga”).  Alga was a sub-contractor of Gammon working on a contract awarded in 2003 by the Highways Department of the Government of the HKSAR to build an 800-metre long dual lane viaduct connecting Tai Lam Kok with Siu Lam.  Alga’s contract with Gammon, made in 2005, was to supply and install pre-stressed steel wire tendons and to conduct post-tensioning work.  After the installation of the tendons, they were subjected to stressing tests to specific tensions to ensure that they could take the load to which the viaduct would be subjected.  The applicant was a Senior Project Engineer, Li the site foreman and Tsoi a workman.

5.  The stressing tests were conducted in accordance with the specifications set out in the ‘Method Statement’, which governed the sub-contract between Gammon and Alga.  Clause 5.12 required that the average measured extension of the pre-stressing tendons on the deck to be be within 5% of the calculated extension and the measured extension of individual pre-stressing tendons to be within 10% of the calculated extension.  Failure to produce the specified pre-stressing force and calculated extension within a tolerance of +/– 5% required the cessation of all stressing operations and the making of a report to the Engineer, Ove Arup and Partners Hong Kong Limited (“Ove Arup”).  In consequence, Ove Arup would issue a non-conformance report to Gammon and an investigation would be required.  If necessary, Alga would be required to do the work again.  If the test results were within the tolerance range Alga would issue a Tensioning Certificate which was accepted in turn by Ove Arup and Gammon.

6.  The stressing process required that one end of the tendon was fixed at an anchorage point and a jack used at the other end to pull the tendon at fixed intervals of pressure until the required tension was reached.  Then, the tendon was fixed to a second anchorage point, securing it in place.  During the tensioning of the tendons measurements were taken of the actual elongation of the tendon, resulting from the pressure applied to it by the jack.  The employee of Alga who took the measurements would call them out and they were noted down.  Representatives of Gammon and Ove Arup were present to witness the process.  Design information in respect of the viaduct provided by Gammon enabled Alga to calculate the predicted elongation of the tendons in the stressing process.  Li and Tsoi were two of the employees of Alga, subordinates of the applicant, whose job it was from time to time to use a steel ruler to measure the elongated tendons in the stressing process.

The prosecution case

7.  The prosecution case was that the applicant provided Li (charge 1) and Tsoi (charge 3), each of whom gave evidence under an immunity from prosecution, with figures to call out in the measurement process, which figures it was agreed between the applicant and Li and the applicant and Tsoi respectively were to be called out if the actual measurement produced results that were significantly different from the figures provided by the applicant.  Pursuant to that agreement on many occasions Li called out the figures provided by the applicant rather than the actual measurement he obtained of the elongation caused by the stressing of the tendon.  Tsoi was not required to perform the task of measuring the elongation of the tendons in the stressing tests on more than 10 to 20 days.  Tsoi called out the figures given to him by the applicant rather than the actual measurements that he obtained of the elongation even though most of the actual results matched the figures given to him by the applicant.  However, on one occasion the difference between the figures supplied to him by the applicant, which he called out, and the actual measurement was in the order of 10 mm.  As a result, the measurements called out by Li and Tsoi were all within the tolerance range and Alga thereby enabled to issue Tensioning Certificates, which were accepted by Ove Arup and Gammon.

The defence case

8.  The applicant testified and called four witnesses in the defence case, all of whom were engineers who worked on the project, three of them for Gammon and one for Ove Arup.  The applicant denied that he had asked any employee of Alga to read out figures that he had given them in the process of measuring the tendons under stressing tests.  However, he said that he had provided Alga workers engaged in the measurement of tendons in stressing tests with the calculated projected elongation of the tendons and had done so for purposes of safety only.

REASONS FOR VERDICT

9.  The judge noted that Li and Tsoi gave evidence under an immunity from prosecution.  He said that he approached their evidence with “caution” having regard, in particular, to the fact that Li had an apparent motive to fabricate evidence against the applicant.

10.  The judge rejected the applicant’s testimony and the assertions that he had made in his out of court video recorded interview by officers of ICAC, that he had only given Li and Tsoi one figure in relation to the tendon stress testing, namely the calculated projected final elongation and that he did not tell them to read out that figure as the measurement result.

Charge 1

11.  In accepting the evidence of Li that the applicant had given him three figures to read out as the measurement obtained in the stress testing if necessary so do, namely if the difference between the actual measurement varied significantly from the figures supplied, the judge had regard to the adverse motive that Li might have towards the applicant.  There was no dispute that the applicant had been instrumental in having Li dismissed from the employment of Alga and that fact was known to Li.  Further, his dismissal came during the time that the tendons were being stress tested.  Nevertheless, the judge found that the obvious danger that a disgruntled employee might disclose the applicant’s misconduct, if indeed he was misconducting himself, did not cause him to doubt Li’s evidence.

12.  In the result, the judge said that he was sure that (paragraph 195) :

“... the only inference to draw is that the defendant gave figures to PW2 (Li) to read out in lieu of the actual measurements so as to ensure the readings fell within the tolerance range specified in the Method Statement and entered into an agreement with PW2 (Li) to provide false measurements ... between August 2005 and June 2006.”

Furthermore, the judge found that, having regard to the test in R v Ghosh [1982] 3 WLR 110, in so conducting themselves the applicant and Li “both acted dishonestly”.

Charge 3

13.  In accepting the evidence of Tsoi, the judge noted that, in contrast to the position of Li, there was no apparent motive for Tsoi to lie and fabricate evidence against the applicant.  He found it to be “inherently improbable” that, if Tsoi always read out the actual measurements, that he would make up his account that the applicant had provided him with four figures to read out if necessary and that he had done so when the difference between the figures provided by the applicant were significantly different from the actual measurement.

14.  In the result, the judge determined that he was sure that (paragraph 216) :

“... the only inference to draw is that the defendant by deceit provided false measurements to PW4 (Tsoi) to read out in lieu of the actual measurements so as to ensure the readings fell within the tolerance range as specified in the method statement. I am satisfied so I am sure in providing PW4 (Tsoi) with the false measurements the defendant did so with intent to defraud as particularised in the charge, knowing this might result in prejudice or substantial risk of prejudice to Gammon, in that the signing of the tensioning certificates would not be based on accurate figures thereby putting at risk the integrity and safety of the viaduct. Applying the test in R v Ghosh I find the defendant in so doing acted dishonestly.”

GROUNDS OF APPEAL AGAINST CONVICTION

Ground 1

15.  At the outset of the oral hearing, Mr McGowan abandoned Ground 1 of the Grounds of Appeal against Conviction, namely that the judge erred in convicting the applicant in circumstances where the prosecution had failed to provide sufficient particulars in the ‘Particulars of Offence’ or in the prosecution case.  He was well advised to do so.  There was no merit whatsoever in the argument.

Ground 2

16.  By Ground 2, it was submitted that the judge had erred in convicting the applicant in respect of both charges given that the prosecution had failed to prove essential ingredients of the offence, namely :

(i) dishonest intent (charge 1); intention to defraud (charge 3);

(ii) risk of economic loss to Gammon/HKSARG (charge 1); substantial risk of prejudice to Gammon/HKSARG (charge 3); and

(iii) that the applicant and Li shared a common intent to practise a fraud (charge 1).

As to (i) : dishonest intent

17.  Of the issue of the ingredient of dishonest intent in respect of charge 1, it was submitted that the judge had erred in not accepting Li’s acceptance, in cross-examination, that he did not feel that he was being dishonest in reading out the figures that the applicant had provided to him as though it was the actual measurement.  It was submitted that the judge erred in the reliance that he had placed on the applicant’s earlier testimony that, “… making up figures was not an honourable thing to do”.  The judge failed to have regard to contrary evidence of Li that he had done nothing “improper” or “wrong”.

As to (ii) : risk of economic loss

18.  Mr McGowan submitted that, having regard to the evidence of Mr Stephen Chau (PW12), a senior engineer with the Highways Department, that loading tests conducted in February and March 2007 confirmed that the viaduct met the required test load and therefore met the design intent, there was no economic loss to Gammon/HKSARG.  Even if there was some risk of economic loss, it might well have been trivial.

As to (iii) : agreement to defraud (charge 1); intention to defraud (charge 3)

19.  It was submitted by Mr McGowan there was no evidence that there was an agreement between the applicant and Li to defraud Gammon.  It was Li’s evidence (charge 1) that he had read out the figures given to him by the applicant because he was his superior and they were the result of the applicant’s calculations and “most likely ... correct”.

20.  Furthermore, it was submitted that the judge erred in rejecting the applicant’s testimony that he gave the figure of the calculated projected elongation of the tendon to Li and Tsoi for reasons that included ‘safety’.  Similarly, the judge erred in rejecting the evidence of Mr Yeung Wo Hei (PW8) and Mr Leung King Fai (PW11), a former Senior Project Manager of Alga, that was supportive of the applicant’s evidence in this respect.

Ground 3

21.  By Ground 3, it was contended that the judge had erred, “… in failing to give a ‘lack of corroboration’ direction to himself”.  Such a direction was required in the circumstances of this case.  Further, the directions the judge gave himself were inadequate in all the circumstances, in particular since charges 1 and 3 were dependent entirely on the evidence of Li and Tsoi respectively and there was no supporting evidence.

22.  Moreover, Li’s knowledge of the fact that the applicant was instrumental in his dismissal from Alga gave him a motive to make a false allegation against the applicant, so that the judge erred in determining that it was “inherently improbable” that Li made up his account of the misconduct of the applicant.  In all circumstances, it was “unsafe and unsatisfactory” for the judge to rely on Li’s testimony.

23.  In respect of Tsoi, the judge erred in determining that he was possessed of no apparent motive to lie and fabricate evidence against the applicant, given that he had been arrested by ICAC and that it was in his interest to testify in accordance with his previous accounts to them of events, in order to comply with the terms of his immunity from prosecution.

A CONSIDERATION OF THE SUBMISSIONS

GROUND 2

24.  The criticism that is made, that the judge was in error in determining that in reading out the figures given to him by the applicant, Li realised what he was doing was wrong (paragraphs 181,182 and 195) and that he and the applicant acted dishonestly, mirrors the submissions that were made to the judge at trial, namely that in light of the evidence the judge could not determine that Li was dishonest.

25.  Reliance was placed on passages of Li’s testimony in which he had asserted that he “trusted” the applicant, believed the figures to be “correct”, that he did “nothing improper”, it was not “wrong” to do that and that he had agreed with the assertion, made in cross-examination, that he did not feel that he was being “dishonest” (see paragraphs 180 and 182).  In rejecting those submissions, the judge said (paragraph 181) :

“Whilst PW2 (Li) thought the figures given by the defendant were correct his evidence was clear that he realised what he was doing was wrong because the defendant’s figures were different to the actual measurements. In re-examination when answering questions about whether anyone was present when the defendant gave the figures PW2 said making up figures was not an honourable thing to do.”

26.  No doubt, the judge was there referring to the evidence of Li, in which he explained why it was wrong to do as he did (page 818D-E) :

“... because those figures were different from the actual measurements I got, so I knew they were wrong.”

Earlier, in his evidence-in-chief Li had said that if he did not read out the figures supplied to him by the applicant but read out the actual measurements the tendons would be categorised as “substandard” (see page 813K-T).  In cross-examination, Li had said of the circumstances in which the applicant had given him the figures (863D-H), “initially, there was no one else present ... initially it was hidden away from other people’s knowledge.”  He explained that describing it as “hidden away” he referred to the figures and went on to say :

“... hidden away from people involved in the tensioning work because, after all, it’s not an honourable thing to do in our trade, we call that making up figure.”

27.  Clearly, the judge had the advantage of hearing the oral evidence and was entitled to accept or reject all or part of the evidence of a particular witness.  Notwithstanding Li’s vacillation in testimony as to his dishonesty, in all the circumstances it would have been astonishing if the judge had made a determination other than that Li was dishonest in doing what he did.

Agreement : intent to defraud

28.  Of the criticism made of the judge’s finding that in giving Li figures to read out in lieu of the actual measurements, so as to ensure the measurements fell within the tolerance range the applicant, “... entered into an agreement with PW2 to provide false measurements as particularised in the charge, save and except...”, it is to be noted that the judge referred to Li’s evidence that in reading out measurements based on the figures given to him by the applicant he had not done so on his own initiative but, “... upon the explanation by Franky (the applicant) to us about how to do it.”  (paragraph 191; transcript page 814S).

29.  Further, the judge noted that Li had testified that he had reported to the applicant differences between the figures given to him by the applicant and the actual measurements but had received no response.  In context, it is to be remembered that this was a course of conduct over a period of time.  The evidence was that Li had participated in measurements approximately 50 times, and that was, by reference to the Tensioning Certificates, in the period “between August 2005 and June 2006” (see paragraph 195).  Also, that on about half of those occasions he had used the figures given to him by the applicant when calling out the measurements.

30.  Clearly, the judge was entitled to draw the inferences that the applicant had given the figures to Li “... to be read out instead of the actual measurements” (paragraph 191), “... so as to ensure the readings fell within the tolerance range specified in the Method Statement” (paragraph 195).  Similarly, he was entitled to reach the conclusion that he reached from those findings of fact, namely that the applicant and Li had entered into an agreement “... to provide false measurements as particularised in the charge...” save as to the dates stipulated in the ‘Particulars of Offence’.

Risk of economic loss/substantial risk of prejudice

31.  There is no merit whatsoever in the submission that the judge erred in determining, in respect of charge 1, that the prejudice to Gammon was the “risk of economic loss” and, in respect of charge 3, that there was a “substantial risk of prejudice” to Gammon.  At an early stage of his reasons for verdict (paragraph 10) the judge made those determinations, noting that the nub of the prejudice was “... that the integrity and safety of the viaduct might be compromised”.  The judge found that the prejudice to Gammon arising from the conduct subject of charge 3 was of a similar nature (see paragraph 216).  As the judge noted, in respect of charge 1 (paragraph 195) the object of the conspiracy was to have the tendon stress test measurements accepted, “so as to ensure the readings fell within tolerance range as specified in the Method Statement.”  In consequence of the measurements reported, Tensioning Certificates were issued and the danger avoided that the actual measurements might be outside the tolerance range, so that Gammon could avail themselves of their rights to have the work suspended and investigations and, if necessary, rectification put in place by Alga.

The applicant’s case : the giving of one figure only in respect of elongation to Li and Tsoi—reasons of safety

32.  In the context of the judge’s rejection of the applicant’s evidence, that he gave one figure only in respect of elongation of the tendon to Li and Tsoi and that he did so, inter alia, for reasons of safety, complaint is made of the judge’s rejection of what it is said was the supportive evidence, in particular of Mr Leung King Fai (PW11).  It is to be noted, that the judge found that on the occasions that the applicant had given figures to Li he had provided him with three figures, not one (see paragraph 174) and Tsoi four figures (see paragraph 205).  Further, the judge noted that in the video recorded interviews conducted by officers of the ICAC the applicant had not said that he had given figures to Li or other workmen “for reasons of safety” (see paragraph 164).

33.  Of the submission that the evidence of the applicant, that the provision of an elongation measurement by the applicant to Li and Tsoi was justified for ‘reasons of safety’, was “corroborated” inter alia by the evidence of Yeung Wo Hei (PW8), the Resident Engineer of Ove Arup, it is to be noted that the judge had rejected the submission (see paragraph 165).  The judge noted that his agreement that for ‘safety reasons’ workers could be provided with the ‘elongated foreseen’ measurement, namely the calculated projected total elongation measurement, was qualified by his evidence that there was no need to provide a worker with the various calculated projected incremental elongations since it was the worker’s job to take the measurements, nothing else : “… it was not for the worker to decide that the measurements are up to standard or not.  It would be something to be decided basing on the actual measurement.” (see transcript, page 978T-979A).  There was no issue of danger during the incremental stages at which pressure was increased in the testing process, “... because the maximum pressure would not have been reached yet.” (see transcript, 977R-978R).

34.  The judge rejected the submission, repeated to this court, that the evidence of Mr Leung King Fai (PW11) a Senior Project Manager of Alga also supported the evidence of the applicant of the provision of an elongation measurement to the foreman Li for ‘safety reasons’, in particular that it was a ‘good practice’ to do so (see paragraph 170).  The judge noted that Mr Leung said that he was not aware of such a practice at Alga and that he himself did not provide measurement figures to Li.  He did not know Li and, since the applicant was present on site, he thought communication of figures would be between two of them (see paragraph 169; transcript 1001-2).  It is to be noted that Mr Leung had joined Alga in September 2005, had been present at tendon stress testing and had signed many Tensioning Certificates (see transcript 986-7).  Finally, it is to be noted that the judge rejected Mr Leung’s evidence that the provision of the measurements was justified for reasons of safety specifically on the basis that he had accepted the evidence of Mr Yeung Wo Hei that it was not necessary to provide the measurements for safety reasons.

GROUND 3 : an inadequate warning in respect of Li and Tsoi

35.  The judge acknowledged the importance of the evidence of Li and Tsoi twice in his reasons for verdict, noting that the prosecution case in respect of charges 1 and 3 “rested very much”, and “relies on” Li and Tsoi respectively and that they were giving evidence under an immunity from prosecution (see paragraphs 14, 32 and 173).  Clearly, he approached them as accomplices in the offences alleged in charges 1 and 3 (see paragraph 176).  He said that, in consequence, he approached their evidence with “caution, in particular Li May Tak (PW2) and ... who are said to have apparent motives to fabricate evidence against the defendant.” (see paragraph 76).

36.  Of his approach to considerations of the reliability and credibility of witnesses, in particular Li and Tsoi, the judge noted that he had had the opportunity to hear them testifying for about two and one days respectively.  Nevertheless, citing the judgment of this court in R v Ng Suen Chau [1995] 2 HKCFLR 64, he acknowledged that “inherent improbabilities provides a better yardstick than reliance upon demeanour of witnesses” (see paragraph 75).  That approach resonates with the nature of the fact-finding process of a tribunal described in the judgments of Bokhary PJ and Litton NPJ in the Court of Final Appeal in HKSAR v Egan & Ors [2010] 5 HKC 180 at 194 F-G, paragraph 2, and 272 C-F, paragraph 304.

37.  In advance of considering in detail the testimony of Li and Tsoi the judge made a general finding in respect of the witnesses called by the prosecution that “unless stated otherwise” they were “honest and reliable witnesses trying their best to recollect events which occurred in 2005 and 2006.” (see paragraph 79).

The judge’s analysis : credibility and reliability of Li

38.  The judge subjected the testimony of Li to a detailed and critical analysis, encompassing no less than eight pages of his reasons for verdict.  In doing so, he addressed the multiple arguments raised in the 34 pages of written submissions advanced on behalf of the applicant.  Clearly, the judge was alive to the issue of a motive in Li, arising from the applicant’s role in Li’s dismissal from Alga during 2006 whilst the tendon stressing tests were ongoing, to fabricate the allegations he made against the applicant.  Li denied that he blamed the applicant for losing his job.  He accepted that he had suffered a loss in salary in returning to his former job as a casual labourer, but added “ When Alga wanted me to become a foreman I told them that I did not think I was qualified for it.” (see transcript 829-830).  At no stage in a lengthy cross-examination of Li was it suggested to him that he had fabricated allegations against the applicant out of a sense of grievance arising from the applicant’s role in his dismissal from Alga.  In making his determination that, if Li was calling out the actual measurements he obtained in the stress tests, it was inherently improbable that Li had fabricated an account of events in which he implicated himself the judge said that he had considered and rejected the issue of motive in Li (see paragraphs 175-179).

39.  The judge went on to consider but reject the submission that the system of ‘spot checks’, on the measurements obtained in the tendon stress tests, implemented by Gammon and Ove Arup, fundamentally undermined Li’s testimony (see paragraphs 183-6).  He accepted Li’s evidence that ‘spot checks’ of the actual measurements obtained, as opposed to witnesses being present at the process, were rare.  Also, he noted the discrepancy between Li’s evidence and his out-of-court statements to ICAC as to the number of occasions on which he had called out the figures supplied to him by the applicant rather than the actual measurements he obtained as being 50% as opposed to 60-70% respectively.  He accepted Li’s evidence, that he was simply giving an approximate estimation (see paragraph 192).

The judge’s analysis : credibility and reliability of Tsoi

40.  The judge’s finding in respect of Tsoi that “… there is no apparent motive for him to fabricate evidence against the defendant” is challenged.  It is submitted to this court that Tsoi had “… a clear motive to fabricate against the applicant in his own interest having been arrested by ICAC and to comply with the terms of his immunity...”.  No such specific allegation was made to Tsoi in the course of his testimony.  Whilst no evidence was led at trial that Tsoi had been arrested there is no dispute that was the case.  However, as noted earlier it is clear that the judge was alive to the fact that Tsoi was giving evidence under an immunity from prosecution.  The immunity was conditional on giving full and true evidence at trial.  There is no dispute that in practice that is understood by all to mean conditional on giving evidence consistent in broad terms with out-of-court statements made to the authorities.

41.  The judge’s analysis of the testimony of Tsoi encompassed four pages of his reasons for verdict.  In common with his approach to the testimony of Li, he addressed the submissions raised on behalf of the applicant.  The judge found it “impossible to accept” and “inherently improbable” that, if Tsoi had read out the actual measurements he had obtained in the tendon stress tests, that he would fabricate an account in which he himself was implicated (see paragraph 206).

42.  Of the issue of the evidence related to ‘spot checks’, and the submission that they undermined the account of Tsoi, the judge noted that it was Tsoi’s testimony that no such ‘spot checks’ occurred when he took measurements on the ‘Deck’ and that when he was working in the ‘Box’ it occurred “... not very often” and, in any event, people conducting the ‘spot checks’ did not stand near him.  Moreover, Tsoi had testified that he was not “worried” by the fact of ‘spot checks’, because most of the actual measurements were accurate, namely in accordance with the figures given to him by the applicant.  The judge noted that Tsoi stressed that to be the case.  Further, no one was present on the occasion that the actual measurement was 10 mm different from the measurement given to him by the applicant (see paragraphs 207-211).  Finally, the judge noted that although at one stage Tsoi had accepted that the issue of being given a measurement for ‘safety reasons’ had been raised with the applicant, nevertheless he remained adamant that he was given four figures not one measurement.

THE LAW

43.  The requirement that a judge must warn the jury or himself that it is dangerous to convict on the uncorroborated evidence of a witness was abrogated in respect of an accomplice by section 60 of the Criminal Procedure Ordinance, Cap. 221.  The guidance given by Lord Taylor LCJ in the judgment of the Court of Appeal of England and Wales in R v Makanjuola [1995] 2 Cr. App. R. 469 at 473 of the approach to be taken, in light of an identical legislative change in the United Kingdom, was adopted by this Court in R v Chu Ip Pui [1997] HKLRD 549 and approved by the Court of Final Appeal in HKSAR v Leung Chi Keung (2004) 7 HKCFAR 526 at paragraph 33 and re-affirmed by the Court of Final Appeal in Wong Chi King v HKSAR (2009) 12 HKCFAR 481 at paragraphs 13-15.  In Makanjuola, Lord Taylor said (page 473) :

“It is a matter for the judge’s discretion what, if any warning he considers appropriate in respect of such a witness, as indeed in respect of any other witness in whatever type of case. Whether he chooses to give a warning and on what terms will depend on the circumstances of the case, the issues raised and the content and quality of the witness’s evidence.”

44.  Of the issue of the ambit of the judge’s discretion, Chan NPJ said in his judgment in Wong Chi King, with which judgment the other judges concurred, (paragraph 16) :

“The trial judge has charge of the trial. He has heard the evidence of the witness and is well aware of the issues involved, the relevance and importance of the evidence to those issues and other relevant circumstances. He is thus in a position to assess whether there is any need to give a caution warning and if so, in what terms. An appellate court will usually be slow to interfere with the exercise of his discretion unless the failure to give a caution warning or to use the appropriate terms when giving a warning, as, adopting the words of Lord Sumner in Ibrahim v R [1914] AC 599, 615, ‘deprived the accused of the substance of a fair trial’.”

45.  It is clear that not only did the judge state that he was approaching the evidence of Li and Tsoi with caution, mindful of the fact that they were accomplices giving evidence under an immunity from prosecution and that he was alive to an additional possible motive in Li to fabricate evidence against the applicant, but also by his careful and detailed analysis of the evidence he demonstrated that he implemented that warning.  The nature of the warning that he gave himself lay within the discretion available to him as the trial judge.  There is no merit in this proposed ground of appeal.