XN000022-1996-07-10 — Page 39

- 17 -

Mr Ho's amendment also would inadequately protect the integrity of investigations. This is not simply the view of the Administration. The Privy Council, in the recent Ming Pao case, made the following comments on the second limb of section 30 subsection (1), which prohibits disclosures to persons other than the suspect.

"Lord Lester argued that the restrictions in the second limb were disproportionate in that they criminalised disclosures even when no prejudice was caused or likely to be caused to an ICAC investigation and even if the accused believed that there would be no prejudice. The difficulty about this argument is that in many cases it will be impossible to know whether disclosure has prejudiced an investigation or not, for example, a suspect might destroy incriminating documents of which the investigator was not and never would be aware but which he would have discovered had there been no prior disclosure. For the same reason the suggestion that the desired aim could have been achieved by qualifying the second limb subsection with some such words as 'likely to prejudice the investigation' fails because of the difficulty of establishing when a disclosure satisfied the test. If the restriction is to be effective it cannot draw distinctions between prejudicing and non-prejudicing disclosures nor have regard to the state of mind of the discloser."

The "likely to prejudice" test would fail to achieve the desired aim of protecting the integrity of ICAC investigations because of the difficulty of establishing when a disclosure satisfied the test. Let me give an example. Let us assume that a government official is under investigation but does not know this. Someone in the same department as the suspect learns of the investigation and tells a colleague. The colleague tells someone else, and so on. Would any of these disclosures be likely to prejudice the investigation and, if so, which one? The fact that the suspect may or may not eventually learn of the investigation does not answer the question. The suspect may overhear a conversation that was not, of itself, likely to prejudice the investigation. If he then destroys all evidence of his corruption the disclosure would still not have been an offence. Even if the suspect is informed of the investigation by a colleague, that does not necessarily mean that the disclosure was likely to prejudice the investigation. Under the "likely to prejudice" test, it is not clear whether it would be an offence for a person to disclose to a suspect the fact that he was under investigation. It is unacceptable from a legal policy point of view that a criminal offence should be subject to such uncertainty.

These examples show that the "likely to prejudice" test does not adequately protect ICAC investigations. I am aware that a similar test appears in certain other pieces of legislation. But that proves nothing. Two of the precedents for the "likely to prejudice" test relate to investigations into drug trafficking and terrorism. Investigations into such offences are of a completely different nature to investigations into corruption. Again I quote from the Privy Council decision in the Ming Pao case [at p.9] -