Privileging Secrecy: The CSE Act’s New Class Privilege -- Part II

23 May 2019

By Leah West

Bill C-59: An Act respecting National Security Matters is currently awaiting third reading in the Senate. In the second of this two-part blog series, I highlight CSE’s new class privilege and some of the history behind the provision.  (Part 1 is here).

For ease of reference, I’ll remind readers that s 55 of the proposed CSE Act states:

It is prohibited, in a proceeding before a court, person or body with jurisdiction to compel the production of information, to disclose the identity of a person or entity that has assisted or is assisting the Establishment on a confidential basis, or any information from which the identity of such a person or entity could be inferred.

            The roots of this provision stem from a 2014 Supreme Court of Canada decision which found that the class privilege afforded police informants did not extend to CSIS human sources, and that it was not their place to create one.[1]  “If Parliament deems it desirable that CSIS human sources' identities and related information be privileged” wrote Chief Justice McLachlin for the majority “it can enact the appropriate protections”[2] 

The Harper Government took up the Court’s offer the following year by amending the CSIS Act  in bill C-44 to create a new statutory privilege for human sources defined under the Act as “an individual who, after having received a promise of confidentiality, has provided, provides or is likely to provide information to the Service.”

Section 18.1 of the CSIS Act now prohibits the disclosure of the identity of a CSIS human source or any information from which the identity of a human source could be inferred in a proceeding before a court, person or body with jurisdiction to compel the production of information. It’s clear that this privilege, like police informant privilege would not protect the identities of those who assist CSE. Section 55 seeks to remedy that fact.

That said, the drafters of the CSE Act didn’t simply copy and past the text of 18.1 into their new legislation, and for good reason. The invocation of s. 18.1 privilege by the government has had a bumpy start and resulted in a fair amount of judicial commentary in its early years. Much to my delight, the drafters of the CSE Act appear to have turned their mind to the lessons learned from this litigation when crafting s.55.

First, as I mentioned in Part 1, under the CSE Act there is no need to establish a “promise of confidentiality” when asserting the privilege, only that the assistance was provided to CSE on a “confidential basis.” While this may seem like a distinction without a difference, the Federal Court of Appeal found that s. 18.1 protects those who fulfilled the definition of a human source before the passage of C-44.[3] Given that for almost 30 years, the Service was under the impression that their sources were protected by informer privilege (which has no hard and fast requirement that a “promise” be exchanged),  it’s easy to see how trying to go back in time to find evidence of if/when the “promise” was originally received in order to prove an individual’s’ status as a source might be tricky.

Second, the question of whether a judge hearing a s. 38 CEA application or s. 87 IRPA application could test the assertion of s. 18.1 privilege claim was also a matter of heated debate. In two separate decisions, the Federal Court made clear that invoking s. 18.1 ousts the Court’s jurisdiction to consider human source privilege in s. 38 and s. 87 proceedings.[4] Most recently, the Court of Appeal also suggested that designated judges adopt a somewhat complicated multistep approach when deciding whether they should appoint an amicus curiae or special advocate to formally challenge the assertion of 18.1 privilege when the affected party chooses not to. Resort to this process would be discretionary and could arguably lead to unacceptable delays.

Section 55(4) of the CSE Act eliminates this entire issue. Instead of creating a whole new regime for challenging the assertion of CSE privilege (like the one found in s.18.1) the new s. 55 regime taps into existing processes:

Sections 38 to 38.‍16 of the Canada Evidence Act, or sections 83 and 87 of the Immigration and Refugee Protection Act, as the case may be, apply to a proceeding referred to in subsection (1), with any necessary modifications.

What those modifications are and how this will work in practice remains to be seen, but I am optimistic that working within a familiar and tested framework will ultimately lead to less litigation and a faster process.

Third, unlike s. 18.1 of the CSIS Act, the CSE regime gives the Federal Court the explicit authority to keep the mere fact that an application for the non-disclosure has been made confidential if disclosure of that fact would reveal the source’s identity. This is a powerful measure that may be necessary where, for example, there is a single person with the access or capacity to provide CSE with the assistance delivered. That said, such orders should only be sought and issued in the clearest of cases and arguably, never in a criminal case.

 Finally, the CSE Act is clear about who falls within the circle of privilege beyond the individual or entity and CSE.  In 2017, the Federal Court actually had to issue a judgment affirming that designated judges were not prohibited from seeing s.18.1 material in closed proceedings (remarkably it wasn’t the Government who made that argument).[5] Section 55 stipulates that the Attorney General, the judge or a court discharging their duties in relation to the assertion of the CSE privilege, the Intelligence Commissioner and the National Security and Intelligence Review Agency are all within the circle of privilege. What’s more, a person or entity who aided CSE may tell their lawyer in connection with a proceeding where that is a relevant fact. Presumably amicus curiae and special advocates would also be privy to the information if appointed in s. 38 and s. 87 proceedings.

Taking a step back, this new class privilege is a big deal. The fact that received zero attention from Parliament, the media, and national security nerds like those of us at Intrepid until now is further proof of just how massive Bill C-59 really is.  

[1] Canada (Citizenship and Immigration) v Harkat, 2014 SCC 37at para 86-87, [2014] 2 SCR. 33.

[2]  Ibid at para 87.

[3] Canada (Attorney General) v Almalki, 2016 FCA 195.

[4]  Canada (Attorney General) v. Almalki, 2016 FCA 195; Canadian Security Intelligence Service Act (Can.) (Re), 2018 FCA 161.

[5] [XXX], 2017 FC 136.