By Stephanie Carvin and Thomas Juneau

May 21, 2026

One of the most enduring questions confronting democratic governance was posed by the Roman poet Juvenal: Quis custodiet ipsos custodes? That is, who watches the watchers?

In the context of national security and intelligence, the question relates to who oversees charged with overseeing intelligence agencies entrusted with extraordinary powers to investigate threats to Canada and ensure those powers are not abused.

Our research — involving interviews with 25 members of the national security and intelligence community and members of all of Canada’s review bodies — suggests that the 2017-2019 reforms to the country’s review architecture have, in many cases, improved the governance and operations of the country’s security agencies.

Yet, a system designed to promote transparency and reassure Canadians that national security institutions operate lawfully and effectively has become overly adversarial, risking the effectiveness of the broader accountability framework.

Some of these tensions are inherent to any relationship between reviewer and reviewee; when it functions optimally, the relationship is not meant to be harmonious but inherently tense.

Other points of conflict, however, appear more cultural or “attitudinal” in nature, especially persistently confrontational approaches, particularly by the National Security and Intelligence Review Agency (NSIRA), resistance within parts of the national security and intelligence community itself, and differing views on the role of trust in the review process.

Canada is once again moving to expand the authorities of its national security and intelligence institutions through Bill C-22.

This piece of legislation, proposed by the government in March 2026, would expand the power of law enforcement agencies and the Canadian Security Intelligence Service (CSIS) to obtain digital information from telecommunications and electronic service providers while also creating a new regulatory framework compelling providers to maintain technical capabilities that facilitate lawful access.

Striking the proper balance between an evolving threat environment and technological landscape on the one hand, and the privacy of Canadians on the other, is therefore urgent.

The reforms of 2017-19, including the creation of NSIRA, should, in theory, reassure Canadians that there are appropriate safeguards built into the system. Yet, as our research shows, the performance of the system overall has been mixed. As Canadians continues to debate Bill C-22, it is essential to understand what has gone right, what has gone wrong, and what can be improved with NSIRA, given the essential role it plays.

How Did We Get Here?

In response to evolving global security threats and the corresponding expansion of state powers, the Trudeau government significantly restructured Canada’s national security accountability framework through the passage of Bill C-59 in 2019.

Historically, Canada’s oversight and review mechanisms had been weak compared to those of many of its allies. In less than a decade, however, Canada developed one of the most comprehensive national security accountability regimes in the democratic world.

Complementing the creation of the National Security and Intelligence Committee of Parliamentarians (NSICOP) in 2017 — Canada’s first parliamentary body composed of members of both the House of Commons and Senate with a mandate to review national security and intelligence activities — Bill C-59 established two additional institutions.

NSIRA was created as an independent body empowered to review all national security and intelligence activities across government, while the Intelligence Commissioner was established to provide quasi-judicial oversight of certain intelligence and national security operations before they are carried out.

In practice, the system evolved so that NSICOP has largely focused on efficacy review (how well do agencies and departments perform), while NSIRA has focused on compliance review (are agencies respecting the law?).

The Challenges

There have been many successes brought about by NSIRA’s reviews. Our research has documented how benefits include strengthened governance, compliance, and professionalism across the intelligence community by encouraging agencies to improve internal processes, documentation, and accountability mechanisms.

Our interviewees often described NSIRA recommendations as driving meaningful institutional reform and providing momentum for policy and governance changes, even when agencies disagreed with aspects of its analysis.

Yet, our research identified increasingly entrenched cultural and attitudinal tensions between NSIRA and the agencies it oversees.

Interviewees consistently described NSIRA’s approach as overly adversarial, legalistic, and focused on exposing failures rather than fostering accountability and constructive solutions. Disputes over access to information, aggressive timelines, and highly contentious reviews have contributed to a deterioration of trust between reviewer and reviewees.

In several cases, agencies viewed NSIRA as more interested in asserting its authority and maximizing its mandate than building the collaborative relationships necessary for effective accountability. Interviewees regularly described unnecessarily aggressive – involving shouting and bullying by NSIRA personnel – processes which damaged the overall ecosystem’s ability to improve.

At the same time, it was clear from our research that resistance to enhanced scrutiny within parts of the national security and intelligence community also contributed to these tensions, particularly in organizations historically unused to external review such as the Department of National Defence and the Canadian Armed Forces as well as Global Affairs Canada.

Beyond specific aspects of contentious individual reviews, the deeper divide appears to concern the role of trust itself. Officials within departments and agencies argued to us that trust is essential to meaningful accountability, open communication, and effective cooperation.

By contrast, some within NSIRA viewed trust as secondary to rigorous legal scrutiny and expressed deep skepticism toward the agencies they review. The result has become a relationship increasingly characterized by suspicion and defensiveness, putting at risk the effectiveness and long-term sustainability of Canada’s otherwise robust national security accountability framework.

Improving the System

In calling for changes to the system, we should first be clear in what we are not calling for. We are not challenging the importance of review; in fact, we strongly believe in the necessity, for both Canada’s democratic health and its national security, of a properly functioning system of intelligence review and oversight.

Nor are we calling for a reduction of the powers of NSIRA or any other body reviewing the activities of national security and intelligence agencies. These powers should be kept in place.

Rather, our goal is to shed light on an aspect of national security policy in Canada that remains poorly understood: how to make intelligence review work better.

Possible reforms could include mechanisms to resolve disputes (for example, on timelines and access to information), both upstream and downstream, between NSIRA and the agencies it reviews. Other mechanisms would seek to improve communication between the two sides. Departments being reviewed should also receive more resources to help them manage what is, inevitably, a time-consuming burden.

The question of who reviews the reviewers, finally, is a much more complex one than it seems. In theory, we should want any publicly funded body, especially one entrusted with such sensitive responsibilities, to be held to the highest standards of accountability.

In practice, however, this has consistently been a vexing question in other democracies. One possibility would be for the chair of NSIRA to appear, for example, before a parliamentary committee once a year to defend the agency’s performance. This would attract some democratic scrutiny on NSIRA, hopefully steer it to justify and explain its actions, and right the ship if necessary.

More broadly, the government should ensure that it remains loyal to the principles of national security transparency that are essential to the effectiveness and sustainability of the system.

NSIRA, for its part, should engage in a critical self-examination — in the absence of any mechanism to review the reviewer — and ponder whether what is widely viewed as its overly aggressive approach to review is the best path to greater accountability for Canada’s national security and intelligence community.

Stephanie Carvin is an associate professor with the Norman Paterson School of International Affairs at Carleton University; Thomas Juneau is a professor with the Graduate School of Public and International Affairs at the University of Ottawa.