Research Brief: Negative Partisanship and the "Silence Pattern" in Canada and the US
Research Brief: Negative Partisanship and the "Silence Pattern" in Canada and the US
1. Bill C-22 (Canada, 2026): What It Actually Contains
Bill C-22, titled the Lawful Access Act, 2026, was introduced on March 12, 2026 by Public Safety Minister Gary Anandasangaree under the Carney Liberal government. It passed second reading in the House of Commons on April 20, 2026 and was referred to the Standing Committee on Public Safety and National Security (SECU), where hearings began May 7, 2026. The bill is a slimmed-down reintroduction of Parts 14 and 15 of Bill C-2, the "Strong Borders Act," which was tabled in June 2025 and abandoned after intense civil-liberties pushback.
The bill has two main parts:
Part 1 — Criminal Code / CSIS Act amendments. It creates new lawful-access tools, including a "confirmation of service demand" (a warrantless yes/no question to telecom providers about whether someone is a customer), production orders for subscriber information, exigent-circumstance disclosures, computer-data examination warrants, tracking-data requests, and a formalized framework for Canadian police to seek court orders aimed at foreign platforms such as Google, Meta and OpenAI. The threshold for subscriber-information production orders is "reasonable grounds to suspect" — lower than the conventional "reasonable grounds to believe." Some categories that Bill C-2 would have allowed police to compel without a warrant now require judicial authorization, and the demand power is narrowed to telecommunications providers rather than "any person who provides services to the public."
Part 2 — Supporting Authorized Access to Information Act (SAAIA). This is the more contested half. It would empower the Minister of Public Safety (with approval from the Intelligence Commissioner) to compel "electronic service providers" — a category broad enough to include telecoms, ISPs, messaging apps, cloud providers, social-media platforms, AI services and "smart" devices — to:
- Develop, install and maintain technical capabilities for extracting and organizing information that law enforcement and CSIS are authorized to obtain;
- Retain "metadata" (including transmission data and location/contact data) for up to one year on every user, regardless of suspicion;
- Refrain from disclosing the existence of such orders publicly;
- Be subject to monetary penalties for non-compliance.
The bill says these mandates cannot impose a "systemic vulnerability," but critics — including the EFF, the Canadian Civil Liberties Association, OpenMedia, Citizen Lab's Ron Deibert, the Centre for Free Expression, Michael Geist, Apple, Meta and Signal — argue that the definitions of "systemic vulnerability," "encryption" and "metadata" are loose enough that cabinet can reinterpret them by regulation, effectively creating a back-door mandate for encrypted services. Signal and the VPN provider Windscribe have publicly threatened to leave Canada rather than comply. Apple has stated it "will never" build a back door. In May 2026, the chairs of the U.S. House Judiciary and Foreign Affairs committees (Reps. Jim Jordan and Brian Mast) sent a letter to Minister Anandasangaree warning that the bill would "drastically expand Canada's surveillance and data-access powers" and create cross-border risks for Americans. More than 25 rights groups and 15 privacy scholars signed an open letter on April 21, 2026 calling the bill "the most expansive invasion of Canadian privacy rights in modern history."
The government's defense has been consistent but narrow. Anandasangaree calls the bill "encryption-neutral," says it is "not about surveillance of Canadians going on about their daily lives," and frames it as catching Canada up to its Five Eyes peers. Spokesperson Simon Lafortune has said critics are "misunderstanding how Bill C-22 would function in practice." Justice Minister Sean Fraser argues the redraft (informed by former NSIRA chair Murray Rankin and Prof. Leah West) will survive Charter scrutiny.
2. Tobi Lütke's Criticism — In Context
Lütke, Shopify's co-founder and CEO (Canada's largest tech company by market cap behind RBC), has a long, well-documented pattern of publicly criticizing Canadian tech and digital-economy legislation on X. Verifiable instances include:
- Bill C-11 (Online Streaming Act, 2023): Lütke amplified Glenn Greenwald's characterization of it as "one of the world's most repressive online censorship schemes" and asked "why would one want to do such a thing?"
- Bill C-18 (Online News Act, 2023): Lütke joined online critics of the bill, which Meta responded to by blocking news links on Facebook and Instagram in Canada.
- Voluntary code of conduct on generative AI (2023): Lütke called it a "field test for regulation" and wrote "Another case of EFRAID. I won't support it. We don't need more referees in Canada. We need more builders."
- CRTC online-streamer registration rules (2023): He expressed "bafflement."
- Trump 25% tariffs / Trudeau retaliatory tariffs (Feb. 2025): "I love Canada and want it to thrive. I built Canada's biggest tech company here because I know it's a special place… Action has to be judged based on what it leads to, not how good it sounds or feels."
- Build Canada (2025): Lütke is a backer of the policy advocacy platform led by ex-Shopify executive Daniel Debow, which has been characterized as part of Canadian tech's "not-so-quiet shift to the right" away from the Liberals.
Is the underlying concern warranted? On the substantive merits, Lütke would have plenty of company. The bill has been opposed by:
- Apple, Meta and Signal (publicly, in committee testimony and corporate statements);
- The Canadian Chamber of Commerce;
- The Canadian Civil Liberties Association, BCCLA, OpenMedia, Amnesty International Canada, the Centre for Free Expression, the Canadian Constitution Foundation, Ligue des droits et libertés and the International Civil Liberties Monitoring Group;
- Citizen Lab director Ron Deibert, Canada Research Chair Teresa Scassa, and University of Ottawa Internet Law Chair Michael Geist (who has written more than a dozen detailed posts dissecting C-22's metadata-retention, systemic-vulnerability and Charter problems);
- The U.S. House Judiciary and Foreign Affairs committee chairs.
The technical concerns — that lawful-intercept architecture mandated by CALEA-style laws was exactly what the 2024 Salt Typhoon hack exploited at U.S. telecoms, that the EU Court of Justice has repeatedly struck down blanket metadata retention as disproportionate, and that "back doors only for the good guys" is not technically feasible — are mainstream views among cryptographers and privacy lawyers, not fringe positions.
3. The "Silence Pattern": Where Are the Liberal/Left Defenders?
Across coverage of Bill C-22, an unusual feature stands out. The discourse is structurally one-sided. Government ministers issue carefully worded talking points ("encryption-neutral," "you misunderstand the bill"), the Liberal caucus votes the bill through second reading, and then… very little. There is no prominent ecosystem of progressive commentators, columnists, podcasters, NDP-aligned activists or Liberal-leaning tech voices arguing affirmatively in favour of mandatory metadata retention, ministerial back-door orders or warrantless confirmation-of-service demands. The substantive defense of the bill in public commentary is conducted almost entirely by:
- Government ministers and their communications staff;
- Police chiefs (Ottawa Police Chief Eric Stubbs offered case-timeline arguments in committee, and one chief told the C-22 hearing that three years of metadata retention would be "ideal");
- CSIS-aligned national-security commentators.
Michael Geist has explicitly characterized this asymmetry, titling one of his analyses "Lawful Access Heads to Committee: The Opposition Found Its Voice, the Government Never Found Its Defence." That is the silence pattern in microcosm: critics — including civil-liberties groups that historically lean progressive, plus international privacy organizations, plus opposition MPs from across the spectrum — are doing all the public talking. Liberal-aligned commentariat is largely absent from the defense.
4. Five Concrete Examples of the Silence Pattern
Example 1: Bill C-22 / Lawful Access Act (Canada, 2026) — the anchor case
As described above, the defense of mandatory metadata retention and government-mandated capability orders against encrypted services is being carried almost entirely by ministers, police and security agencies. Civil-society opposition spans an unusual coalition (CCLA, BCCLA, OpenMedia, Amnesty Canada, Citizen Lab, the Canadian Chamber of Commerce, U.S. Republican committee chairs, Apple, Meta, Signal). Progressive media and Liberal-aligned commentators have generally not produced the kind of vocal, sustained defense that accompanied, say, the Online News Act when it targeted Meta and Google. Geist's framing — "the opposition found its voice, the government never found its defence" — captures it precisely.
Example 2: Bill C-63 / Online Harms Act (Canada, 2024) — split criticism, near-absent defense
The Trudeau government's Online Harms Act bundled child-safety platform regulation (Part 1) with hate-speech-related amendments to the Criminal Code and Human Rights Act including potential life sentences for "hate-motivated" offences and pre-crime peace bonds (Parts 2 and 3). The Centre for International Governance Innovation observed that "the Liberal government is surely ecstatic about the relatively muted response," noting that most actual public criticism focused on the criminal-law amendments rather than on the platform-regulation core, and that "there hasn't been much opposition because there's not much here to oppose" — but also that there was almost no enthusiastic public defense. The civil-libertarian critique came overwhelmingly from the BCCLA, OpenMedia, Amnesty, JCCF, the Canadian Constitution Foundation and conservative commentators; left-progressive defenders of the more draconian Part 2 measures were difficult to find. The bill died at prorogation in January 2025; advocacy groups have since urged the Carney government to split the bill rather than reintroduce it whole — implicit recognition that Parts 2 and 3 had no constituency willing to defend them in public.
Example 3: Bill C-11 (Online Streaming Act) and Bill C-18 (Online News Act) — Liberal defenders fell quiet after rollout
C-11 was sold as updating CanCon rules for streamers. After passage and the CRTC's 5% contribution requirement, public Liberal-aligned defenses became scarce; even Liberal-appointed Senator Paula Simons voted against the final version, calling some of the more apocalyptic framing "fear-mongering" but warning the bill was poorly drafted. C-18 produced the clearest "silence-after-rollout" pattern: the bill was supposed to force Meta and Google to pay Canadian publishers, but Meta simply blocked news on Facebook and Instagram in Canada — dealing what Canadian Dimension called a "death blow to our emerging online media." Liberal commentators who championed the bill before passage largely stopped defending it once Meta's response made the policy look like a strategic miscalculation. Geist has explicitly drawn the analogy: the government's dismissive response to Signal, Apple and U.S. Congress on C-22 "runs back the disastrous Online News Act playbook."
Example 4: U.S. — FISA Section 702 reauthorization under the Biden administration (2023–2024)
This is the cleanest U.S. analogue. In December 2023, a short-term Section 702 extension was slipped into the NDAA. In April 2024, the Reforming Intelligence and Securing America Act (RISAA) reauthorized Section 702 for two years without a warrant requirement for "backdoor searches" of Americans' communications. President Biden signed it. The House warrant-requirement amendment failed 212–212 — 128 Republicans and 84 Democrats voted yes; 126 Democrats voted no. The Biden White House issued a statement "strongly opposing" the warrant requirement.
The asymmetric engagement pattern was stark. Loud opposition came from the ACLU, Brennan Center, Demand Progress, EFF, the Project for Privacy and Surveillance Accountability and a coalition of more than 30 civil-society groups — most of them traditionally coded progressive. Loud opposition also came from libertarian Republicans (Mike Lee, Jim Jordan, the House Freedom Caucus). But the broad ecosystem of progressive commentators, opinion writers and Democratic-aligned podcasters who had been vocal about NSA surveillance abuses during the Snowden era and the Trump first-term Section 702 debates was substantially quieter when their own administration was pushing reauthorization. (GovFacts, summarizing the politics for the 2026 reauthorization fight, notes plainly: "In 2024, when Biden was president, many Democrats voted against the warrant requirement. Now that Trump is president, some are reconsidering" — Jamie Raskin, who voted against the warrant requirement in 2024, is now using the language of "assault on the FBI's internal guardrails." That is negative partisanship in real time.) Lawfare's piece "Mum's the Word on FISA Section 702 Reauthorization" captures the silence in its title.
Example 5: U.S. — NDAA national-security provisions and other "must-pass" Biden-era vehicles
The same pattern recurred with the use of the NDAA as a vehicle to slip in surveillance reauthorizations (December 2023) and the Biden administration's filing for a FISA Court certification that effectively converted a 4-month extension into a 16-month one — a maneuver the Brennan Center's Elizabeth Goitein flagged in real time. Civil libertarian opposition was loud; affirmative defense from Democratic-aligned media voices was sparse compared with how the same voices treated equivalent Trump-era surveillance debates.
5. The Underlying Concept: Negative Partisanship
The term "negative partisanship" is most closely associated with political scientist Alan Abramowitz of Emory University, particularly his work with Steven Webster:
- Abramowitz, A. I. & Webster, S. (2016), "The rise of negative partisanship and the nationalization of U.S. elections in the 21st century," Electoral Studies 41, 12–22.
- Abramowitz, A. I. & Webster, S. (2018), "Negative Partisanship: Why Americans Dislike Parties But Behave Like Rabid Partisans," Political Psychology 39(S1), 119–135.
The core empirical claim: since roughly the 1980s, Americans' "feeling-thermometer" ratings of their own party have stayed roughly flat (mid-60s to low 70s), but ratings of the opposing party have collapsed. Voters increasingly form political identities not by what they support but by what they oppose. Abramowitz's metaphor: politics has become "like a bitter sports rivalry, in which the parties hang together mainly out of sheer hatred of the other team, rather than a shared sense of purpose." Rachel Bitecofer built on this work to argue elections are now driven by turnout among negative partisans rather than persuasion of swing voters.
Canadian research is robust and confirms the pattern in a multi-party system:
- Caruana, McGregor & Stephenson (2015), "Negative Partisanship in a Multi-party System: The Case of Canada," Journal of Elections, Public Opinion and Parties 25(3), refining how to operationalize negative partisan ID (NPID) in Canada's multi-party context using five federal-election datasets.
- McGregor, Caruana & Stephenson, "The Power of the Dark Side: Negative Partisanship and Political Behaviour in Canada," using 2008 and 2011 Canadian Election Study data, finds NPID independently predicts turnout, vote choice and political activity.
- Anderson, McGregor & Stephenson (2022), "Us versus them: Do the rules of the game encourage negative partisanship?", European Journal of Political Research, using the Comparative Study of Electoral Systems to show electoral-system features condition negative partisanship.
- Cross-national work (cited in Iyengar & Krupenkin 2018 and others) finds negative partisanship undermines satisfaction with democracy.
The social-media engagement asymmetry — which is the mechanism behind the "silence pattern" — is also well-documented:
- Rathje, Van Bavel & van der Linden (2021), "Out-group animosity drives engagement on social media," PNAS 118(26): a study of 2,730,215 posts from U.S. congressional members and news media accounts. Posts referring to the political out-group received roughly twice as many shares as posts about the in-group; "out-group" language was the single strongest predictor of social-media virality, outperforming negative emotion and moral language.
- Brady, Wills, Burkart, Jost & Van Bavel (2019), "An ideological asymmetry in the diffusion of moralized content on social media among political leaders," Journal of Experimental Psychology: General 148, 1802.
- A pre-registered analysis of 1,195,844 tweets by 564 U.S. political elites (members of Congress, presidential and VP nominees 2000–2020, Trump Cabinet) published in Political Behavior (2023) found a revealing split: politicians themselves post more in-party-positive than out-party-negative tweets, but ordinary users engage (likes, shares) much more with the out-party-negative posts. In other words, the supply side rewards positive partisanship from elites, but the engagement side rewards negative partisanship from the audience — which is exactly the dynamic that produces the silence pattern. There is no organic audience demand for "I defend this controversial thing my party did"; there is enormous audience demand for "look at what the other team is doing."
- Conover, Gonçalves, Flammini & Menczer's "Partisan Asymmetries in Online Political Activity" (2012, examining 18,000+ politically active Twitter users around the 2010 midterms) found highly segregated, well-clustered partisan community structures — early empirical groundwork for the silos in which the silence pattern now operates.
The picture from this literature is consistent: dislike of the out-party is a stronger driver of online engagement and political behavior than affection for the in-party; partisans get rewarded — in attention, in social validation, in algorithmic amplification — for attacking the out-group, and not rewarded for defending controversial decisions by their own side. The rational response of a partisan with limited time and emotional energy is to be vocal about what the other team does and quiet about what their own team does. That is the silence pattern.
6. Summary: Anchors for the Essay
The pattern the essay describes is real and documented at three layers:
- Quantitative behavioral evidence (PNAS 2021, Political Behavior 2023, Brady et al. 2019): out-group hostility produces 2× the engagement of in-group affinity on social media; ordinary partisans engage far more with attacks on the out-party than with defenses of the in-party.
- Theoretical framework (Abramowitz & Webster 2016, 2018; Iyengar & Krupenkin 2018; in Canada, Caruana/McGregor/Stephenson 2015, Anderson et al. 2022): negative partisanship has replaced positive partisanship as the dominant axis of political identity in both the U.S. and Canada.
- Five concrete cases where supporters of the governing party went conspicuously quiet on controversial legislation that they would predictably have attacked had the other side passed it:
Two important honesty notes for the essay:
- The specific Lütke "death blow to Canadian tech viability" quote about C-22 could not be independently verified in indexed sources during this research. It is wholly consistent with Lütke's documented pattern of public statements on Canadian tech regulation (C-11, C-18, the AI voluntary code, CRTC rules) and with his role in Build Canada, but the essay should either source it to a primary post or attribute it carefully ("Lütke has reportedly called…" or quote his verifiable statements on adjacent bills).
- The silence pattern is not absolute. Some Liberal-aligned voices did defend C-11 (CIMA's Andrew Cash, NDP heritage critic Peter Julian) and child-safety advocates have defended Part 1 of C-63. The claim isn't that defenders don't exist; it's that the ratio of attention — vocal critics to vocal defenders — is dramatically lopsided in a way that doesn't occur when the opposing party is the one acting. That asymmetry is what the engagement research at PNAS, Political Behavior, and JEP:G actually measures, and it is the mechanism that makes negative partisanship not just an attitude but a structural feature of how political discourse now propagates.
24 facts · 16 assertions → Ron Deibert · Teresa Scassa · U.S. House Foreign Affairs · Liberal caucus · EU Court of Justice · Liberal government · BCCLA · OpenMedia. Every one is a verbatim span; nothing was paraphrased into the graph.
This is a signed piece; its findings carry their sources inline, in the text. The piece argues; the sources carry the proof.