Judicial interventions during a trial – the Dayaram appeal


In HKSAR v. Dayaram [2017] HKCA 27, the Court of Appeal considered whether the trial judge’s numerous interventions during witness testimony made the trial unfair.

Nanik and Mahesh Dayaram, father and son, were convicted of 9 counts of conspiracy to defraud.  On the prosecution case, companies under their control purported to make loan applications to HSBC and other financial institutions.  They used fake documents and falsely represented that sham companies were genuine suppliers or purchasers of goods.  The net loss to the banks was in the region of $200-250 million.

They sought leave to appeal against their convictions on several grounds.  The sole ground upon which leave was granted, and the focus of this article, concerned interventions by the trial judge during witness testimony.  The appellants would have had good reason to be optimistic about chances of success: by the time of the appeal hearing, the Court of Appeal had quashed convictions due to excessive interventions by the same trial judge in a different case: see HKSAR v. Lai Oi Yan [2016] 3 HKLRD 273.

The legal test, as confirmed by the courts of Hong Kong and other jurisdictions, is whether looking at matters complained of in the context of the trial as a whole, and taking all the Judges’ interventions together, a fair-minded and informed observer may come to the view that there was a real possibility that the Court was biased and that the defendant did not have a fair trial: see, for example: HKSAR v. Sin Wing Yi [2008] 3HKLRD 352, Porter v. Magill [2002] 2 AC 257.

In particular, there are three situations which give rise to convictions being quashed: see R v. Hamilton [1967] Crim LR 486, and The Queen v. Lam Wai Hang [1997] HKLRD 562 at paragraph 53:

  1. Where the judge, by his intervention, was in effect inviting the jury to disbelieve the evidence for the defence.
  2. Where the interventions made it really impossible for counsel to do his or her duty to present the defence properly.
  3. Where the interventions prevented the accused from doing himself justice and telling the story in his own way.

Here, the Court of Appeal examined the context in which interventions were made, and applied the above principles.  Although there was force in some of the appellants’ arguments, ultimately the trial was fair.

For example, the judge had put to Mahesh Dayaram that the effect of his testimony was that an important prosecution witness had fabricated his evidence (whereas the defence case was that the witness was unreliable, due to fading memory over the passage of time).  The Court of Appeal accepted that there was “force” in the complaint about this intervention, and was satisfied the judge’s intervention was “unwarranted”.  However, after considering the details of how the defence case was put, it was satisfied that the latter was not undermined by the intervention.

A second example is that, the judge misrepresented forensic evidence when putting questions to the same defendant.   The judge referred to a computer server “having been emptied”, whereas there was no evidence to suggest that it had ever contained data; it was simply blank.   However, as the Court of Appeal pointed out, no objection was taken at trial, and defence counsel said in his closing speech that the defendants did not know how the server “became empty”. In other words, it seems defence counsel carried through the error and dealt with it.  In the circumstances, the intervention did not make the trial unfair.

Another complaint was that the judge stopped counsel from asking questions about difficulties in locating documents, which were in the possession of a third party.  However, this questioning was apparently in direct contradiction to a pre-trial ruling by the judge, and relevant matters had not been put to prosecution witnesses. In the circumstances, unsurprisingly, the Court of Appeal held that the judge was entitled to stop counsel from pursuing this line of questioning.

In summary, the Court of Appeal found that the judge’s interventions did not impede the defendants’ ability to put forward their defence properly before the jury.  The appeal was dismissed.

Perhaps the temptation to intervene is particularly great in complex fraud cases.   Judges may hold a perception – rightly or wrongly – that the jury need help to understand unfamiliar concepts and complicated evidence.   The jury might not get to grips with certain keys points without the judge’s interventions, ostensibly for the purposes of “clarification”.   Rather than leaving it to experienced advocates, the judge descends into the arena.   However, as  Dayaram illustrates, the Court of Appeal will examine trial transcripts carefully to understand the context in which interventions are made.   In this case, an examination of the surrounding context revealed that the test for quashing the convictions was not met.


Julia Faure Walker