The Judicialization of Bulk Powers for Intelligence Agencies
Monday, February 26, 2018 at 11:48AM
craigforcese in CSE, CSIS, Chapter 11: Surveillance, bulk powers, c-59

Personal Speaking Notes (February 2018)

(posted publicly with permission)        

I have been asked to reflect on common trans-Atlantic intelligence dilemmas, and then a variation on our traditional trans-Atlantic search for solutions.  To that end, I’ll say a few words about both the UK Investigatory Powers Act and some of the proposed aspects of bill C-59. 

In some large measure, both the UK IPA (Investigatory Powers Act) and C-59 constitute what former CSIS director Jim Judd once called “the judicialization of intelligence”. Mr Judd raised concerns about this development.  Intelligence has traditionally operated in a manner obliquely governed by law, if at all. There is a disconnect between a covert intelligence function – and its requirements – and the more overt culture of law and lawyers and judges. Intelligence needs are fluid.  Law is rigid. Intelligence needs are immediate and exigent. Law can be laborious.

But law has inevitably encroached on intelligence. An academic colleague – Dennis Molinero – has uncovered a trove of documents from the 1950s.  At that time, these documents show, national security domestic intercept warrants were issued by Prime Minister Louis St Laurent as an exercise of discretionary power under something called the Emergency Powers Act. There was the vaguest of statutory imprimaturs, and certainly no independent judicial oversight in the form of preauthorization.

We abandoned that approach in 1974, and the original iteration of the what is now Part VI of the Criminal Code.  And in 1984, we built CSIS search and seizure around a judicial warrant process – and the next year, the Supreme Court decided Hunter v Southam. Since then, in cases like the Federal Court of Appeal’s decision in Atwal, through to Justice Crompton’s recent decision in the In the Matter of Islamist Terrorism case, the domestic intelligence search and seizure expectations have been placed on a constitutional footing largely indistinguishable from that of criminal law.

In the IPA, the UK has moved considerably closer to our model than had been the case before. Once the purview of ministers, executive warrantry is now supplemented by review by judicial commissioners.  The shorthand is: double-lock (executive approval of a warrant supplemented by judicial review, prior to execution).

But in Canada, we have yet to address two dilemmas also at issue in the IPA. Both fall in the realm of what in the UK context is called “bulk powers”.  And since in bill C-59 we moving in this area, and judicializing, it is on this topic I wish to focus a few remarks.

So first, let me define bulk powers: a bulk power is one that allows intelligence agencies access to a large quantity of data, most of which is not associated with existing targets of investigation. It is the mass access, in other words, to data from a population not itself suspected of threat-related activity. The commonplace example, since Snowden, is internet or telephony metadata for entire populations of communications users.  But bulk powers can also involve content, and not just the metadata surrounding that content.

Bulk powers are controversial – they are the heart of the post-Snowden preoccupations. They inevitably raise new questions around privacy, and in the Canadian context, Charter rights.  Not least: bulk powers are irreconcilable with the requirements of classic warrants. There is no specificity. By definition, bulk powers are not targeted; they are indiscriminate.

In the IPA context, the world of bulk powers can be divided into bulk interception; bulk equipment interference; bulk acquisition; and bulk personal datasets.  Of these, I want to focus on bulk interception and bulk personal datasets.

Bulk interception is what is sounds like: the collection of transiting communications passing through communications providers or otherwise through the ether. 

Canadian law permits bulk collection by the Communications Security Established, our signals intelligence service. It is subject to the caveat that acting under its foreign intelligence or cyber security mandate, CSE may not direct its activities at Canadians or persons in Canada. But in practice, bulk interception cannot be limited to foreigners, even if the objective is foreign intelligence. The way communications transit the internet and other communications systems creates a certainty that bulk intercept directed outside the country will intercept the communications of Canadians and persons in Canada.  This is known as incidental collection.

In Canada, we have struggled with this issue. Part of the answer is in Part VI Criminal Code. As you know, it outlaws unauthorized intercept of private communications. A private communication is one with at least one end in Canada. Since in bulk interception, at least some private communications would be captured in a manner meeting this definition of intercept in Part VI, CSE must be exempted from its reach.  And that is what the National Defence Act does, where CSE acquires a defence minister authorization in advance for at least the class of foreign intelligence or cybersecurity activities that might capture this private communication.

The constitutional issue is more fraught. Not least, the defence minister is not the independent judicial officer invoked as the gold standard under Hunter v Southam for Charter section 8.  The consequence has been the constitutional lawsuit brought against CSE by the BCCLA associations and now efforts at refinement in C-59.  And specifically, C-59 anticipates a quasi-judicial intelligence commissioner who will review the ministerial authorization before its execution. This past week, representatives of the CSE testifying before the Commons committee accepted the underlying constitutional expectation: They said under C-59, CSE will seek ministerial authorization (which in term triggers review by the intelligence commissioner) for any activity that would interfere with the reasonable expectation of privacy of a Canadian or a person in Canada, or contravene an Act of Parliament.

I am hoping that signals a willingness to amend the bill to say just that, on its face, but for our part my key point is this: C-59 clearly accepts the underlying premise: judicialization of bulk intelligence interception. In this respect, C-59 emulates the IPA.

But I wish to be clear, again: this is not a warrant. It will lack specificity. It will be issued for classes of activities, not specific activities or operations. It is review on reasonableness of a ministerial authorization, not the more hands-on warrant process. Does that meet Hunter’s standards?  I am inclined to suggest, yes, because the warrant cookie cutter cannot possibly apply to a form of bulk intercept in which intercept of s.8 rights-bearer communications is entirely incidental, and not targeted.

Before leaving CSE, I will say a word about another C-59 change.

We have also gone one step further than the IPA in giving CSE a specific offensive cyber mandate – called active cyber.  This could and almost certainly would implicate equipment interference, but interference untied to information acquisition and instead done “on or through the global information infrastructure to degrade, disrupt, influence, respond to or interfere with the capabilities, intentions or activities of a foreign individual, state, organization or terrorist group as they relate to international affairs, defence or security.” 

At present, there is considerable debate in Parliament about whether the intelligence commissioner should have advance oversight responsibilities in relation to this mandate.  Currently, he or she will not.  I am of two views on whether judicialization in this area would be wise or not.

Turning to domestic-facing bulk powers, I need to switch agencies and talk about CSIS.  And here we have drawn clear inspiration from the IPA in the area of bulk personal datasets.  The UK understanding of this expression is an apt descriptor of what is now also in play in Canada:

"A bulk personal dataset includes personal data relating to a number of individuals, and the nature of that set is such that the majority of individuals contained within it are not, and are unlikely to become, of interest to the intelligence services in the exercise of their statutory functions. Typically these datasets are very large, and of a size which means they cannot be processed manually."

Why have such things? The C-59 changes are a response, yes, to the Federal Court’s 2016 decision on what was known as ODAC.  But it also responds to a broader concern about the ambit of the Service’s threat investigation mandate. That mandate is anchored in s.12 of the CSIS Act. As interpreted by the courts, it permits the Service to collect, and analyse and retain information and intelligence respecting activities that may on reasonable grounds be suspected of constituting threats to the security of Canada, to the extent strictly necessary.  As Justice Noel and Justice Crampton concluded in both the ODAC case and the more recent In the Matter of Islamist Terrorism decision, this is a significant fetter on CSIS. It ties information collection, retention and analysis to a narrow band of threat investigations.  It also makes it difficult for CSIS to change the frequency of its fish radar and expand its reach to search other parts of the ocean for fish that have not already come to its attention.

A spy service fishing in more ocean is, in some eyes, the stuff of Big Brother and nightmares. On the other hand, an intelligence service that cannot have access to the ocean in performing its function is also likely unable to perform its functions very well.  And there is a lot of ocean out there in the digital era.  So how can we reconcile oceans full of data generated by innocents with the intelligence function of clearing the fog of uncertainty and revealing not just the known threats but also the unknown threats?

The solution in both the UK and Canadian context is to judicialize the fish detecting radar. And the model is again a double lock: ministerial approval for ingestion of datasets and judicial commissioner approval.

The result, in the Canadian context, is enormous complexity. Broadly speaking, there are a set of legislated rules in C-59 for the ingestion of datasets, and then a more demanding set of rules for the digestion. (I credit a Department of Justice lawyer for this ingestion/digestion analogy, which is quite apt).  So for Canadian datasets – datasets primarily comprising Canadian information – there is approval of classes of datasets that may be ingested by CSIS by both the minister and the quasi-judicial intelligence commissioner.  Once ingested, there is a limited vetting by CSIS.  And then any subsequent retention for actual use – that is digestion -- must be approved by the Federal Court, which is empowered to impose conditions on that subsequent use.  There is also a requirement that querying generally be done only where strictly necessary in performance of CSIS’s mandates.

I have included charts in the materials. (See also here).

Those charts show why some intelligence operators complain that C-59 is a gift to lawyers.  I suppose it is no surprise, then, that I think this is a clever regime.  Not least, it short circuits inevitable frontier s.8 issues; to wit, does s.8 attach to the big data analysis of information, the individual bits of which triggers no reasonable expectation of privacy. It seems almost certain that the jurisprudence will get there. C-59 heads this issue off at the pass by superimposing independent judicial authorization guiding and conditioning that big data analysis.

So, on that happy note, I shall end there.

Thank you.

Article originally appeared on (
See website for complete article licensing information.