Holding Back the Race Card: Black Candidates, Twitter, and the 2021 Canadian Election

Abstract

Politicians carefully construct a public persona that is authentic to who they are as individuals but also addresses voter expectations. Many Black candidates follow a deracialization strategy in which they downplay their racial identities to seek voter support while some follow a racial distinction strategy in which they highlight their racial identities but situate them within hegemonic national narratives. But questions remain about whether a candidate’s decision to use one strategy over another is shaped by national context, partisanship, political position, and riding competitiveness. This paper thus asks the question: How do Black candidates in Canadian elections deploy race in their campaign communications, and what factors might explain any differences in their strategies? To answer this question, we analyze how Black candidates used Twitter during the 2021 Canadian election. Our analysis reveals that Black candidates generally used a deracialization strategy when communicating on Twitter, opting to celebrate the many cultural groups in their riding rather than casting their appeal only to Black voters. They only highlighted their racial identities or racial issues when world or campaign events gave them the political cover to do so. But the degree to which Black candidates engaged in (de)racialized communications differed by party.

“If me speaking out and calling things out is going to make me a one-term MP, so be it. I accept it. But that doesn’t mean I’m going to stop.”

Celina Caesar-Chavannes, Liberal MP

“That’s the main difference between us. You think the world revolves around your skin colour. My goal is to bring better policies to all Canadians. That’s an MP’s job.”

Maxime Bernier, Conservative MP

In May 2018, Canadian news organizations reported that two members of Parliament had been locked in a months-long “Twitter brawl over race and identity politics” (Wherry 2018). One point of contention was government funding for minority communities. Centrist politician Celina Caesar-Chavannes challenged right-wing politician Maxime Bernier over his claim that Canada should strive for a “colour-blind society where everyone is treated the same.” ¹ She argued that such an approach actually contributes to racism and urged him to “check your privilege and be quiet.” ² Caesar-Chavannes later apologized for the “be quiet” comment but not for her views on racism. The heated online exchange between the country’s only Black female MP and a white male MP caused a public uproar, with both politicians receiving critique and praise for their comments.

This Twitter tiff is just one example of the challenges that Black politicians face in explicitly addressing race in politics. Reflecting on her time in federal office, Caesar-Chavannes (2021) remarked that her political battles with opposition parties were less challenging than the trials she experienced with her own party, whom she accused of leveraging her in a tokenistic fashion, failing to adequately support her at the height of the Bernier controversy, and prioritizing party loyalties above all. She levelled most of her criticism at her party leader, Liberal Prime Minister Justin Trudeau, whom she blasted for, among other things, failing to include any Black ministers in his first gender-balanced and ethnically diverse cabinet.

Caesar-Chavannes admitted that she struggled to find her voice as a Black female politician and to express her “authentic” self. She did not focus on her race or gender during the 2015 election campaign. Instead, her messaging highlighted her business acumen and personal success as an entrepreneur and advanced the local economy as her top priority (Pessian 2015). She was also a loyal Liberal brand ambassador, “‘selling the product’ by knocking on doors” (Caesar-Chavannes 2021: 143). She became more outspoken on identity issues after she got into office, including speaking in Parliament against Black hair shaming, a move that resulted in a profile in O, The Oprah Magazine. However, resistance to her strong Black agenda and other challenges proved too much. Caesar-Chavannes publicly stepped down from the Liberal caucus in February 2019 to sit as an independent and did not contest the October 2019 election.

This brief account of one politician’s rise and fall demonstrates the importance of understanding the role of racial identity in the political communication strategies of Black politicians, especially in party-centred systems. Black political representation in Canada has grown in recent decades as more Black individuals than ever before seek elected office at the national and subnational levels. But at 3.5 percent of the Canadian population, Black candidates cannot just rely on Black voters to get elected. They need to convince Canada’s multicultural electorate to support them. Further, Canada’s Black population itself is highly diverse from a demographic and socio-economic perspective, reflecting the country’s comparatively open immigration system. This makes it less likely that Black voters in Canada will share a sense of linked fate and group consciousness that might shape political attitudes, policy preferences, and vote choice. Finally, Canada has a strong party system where candidates and MPs are expected to act as local brand ambassadors for the party (Marland and Wagner 2020). Strict party discipline might leave little scope for Black candidates to personalize their campaign message around issues of race. It remains an open question whether an emphasis on one’s racial identity is a viable election strategy for Black candidates in Canada. We therefore seek to answer the following research question: How do Black candidates in Canadian elections deploy race in their campaign communications, and what factors might explain any differences in their strategies? To address this, our study analyzes how Black candidates used Twitter to address racial issues and discuss their racial identities over a 9-week federal election campaign period in August and September 2021.

The increase in Black candidacies in recent Canadian federal elections makes it pertinent to examine the types and extent of racial messages in campaign communication. While the message strategies of Black candidates in the United States have been extensively studied, much less is known about how different racial and institutional dynamics play out in local campaigns within other national contexts. Studies in the U.S. have shown that the situational context is highly important and that racialized minority candidates running statewide or in racially mixed or majority white districts typically deploy a deracialization strategy. This approach involves downplaying one’s racial identity and related policies in favour of focusing on issues of common concern like the environment and economy (Collet 2008; Gillespie 2022; McIlwain 2013; Orey and Ricks 2007; Strickland and Whicker 1992). However, unique features of the U.S. political system raise questions about the degree to which these findings can be transposed to other national contexts. First, voting in the U.S. tends to be characterized by extreme racial polarization. That is, racialized minorities tend to vote in a substantial block for candidates who are members of the same racial group, while white voters are substantially unwilling to vote for minority candidates. ³ In a less racially polarized national context, such as Canada, Black candidates may see little need to de-emphasize racial issues. Second, the weak party system in the U.S. gives individual candidates considerable independence, which results in enormous variation in the types and extent of racial messages Black candidates can invoke in their campaign communications (McIlwain and Caliendo 2011). While it uses the same single-member electoral system as the U.S., Canada’s more party-centred parliamentary system should impose stricter constraints on campaign messaging at the local level.

Canada’s 2021 federal election provides an ideal opportunity to study this question. While the Canadian parliamentary system does not normally facilitate highly individualized local campaigns, the 2021 election is a “most likely case” where we might see race-focussed messaging among Black candidates for several reasons. First, race became a highly salient issue before the election with the murder of George Floyd on May 25, 2020 in the U.S. and the resulting swell of support for the Black Lives Matter movement. Second, one of Canada’s main political parties was headed by a Black leader for the first time. Both developments raised the issue of Black political representation to a higher profile than in prior elections. Further, the 2021 federal election saw the largest-ever number of Black candidates (70), running across the country as challengers and incumbents from a variety of parties and social backgrounds. This heterogeneity allows us to explore variations in the local messaging strategies of Black candidates. We focus, in particular, on whether these strategies differ in regard to Black candidates’ partisan affiliation and their status as incumbents or challengers.

Our qualitative analysis reveals that Black candidates generally used a deracialization strategy when communicating on Twitter, opting to celebrate the many cultural groups in their riding rather than casting their appeal only to Black voters. Policy-focussed messaging tended to address cross-cutting issues such as the environment and economy rather than Black issues. Black candidates mainly highlighted their racial identities and/or racial issues when world or campaign events gave them the political cover to do so. We also found that partisanship influenced which Black candidates engaged in deracialized communications. Those running for right-wing parties usually did not address race, even implicitly, while candidates for left-wing parties discussed race more openly, though still with considerable nuance. This study is the first to conceptualize the form that Black candidates’ racial appeals take in the Canadian electoral context. Moving beyond the U.S. context helps to advance our understanding of institutional constraints and the role of partisanship in shaping the communication strategies of Black politicians.

Setting the context: Black political representation in Canada

Despite being a leader in incorporating diversity within parliament (Bloemraad 2006; Bird et al. 2010), Black Canadians have been historically underrepresented within the House of Commons. The first Black MP (Lincoln Alexander) was elected at the comparatively late date of 1968, while the first Black woman MP (Jean Augustine) took office only in 1993. Since then, Black Canadians have made incremental gains in terms of political representation (Johnson-Myers and Everitt 2022), though the pace has increased in recent elections. Between 2015 and 2021, the number of Black candidates running for a seat in the House of Commons nearly doubled from 45 to 70, while the number who won seats grew from 5 to 9.

Of the 70 Black candidates who contested the 2021 election, 25 ran for the New Democratic Party, 20 for the Liberal Party, 14 for the Conservative Party, 7 for the Green Party, 3 for the Bloc Québécois, and 1 as an independent. Only 9 candidates were successful in their electoral bids (7 for the Liberals, 1 for the Conservatives, and 1 for the NDP), leaving Blacks holding 2.7% of the seats in the House of Commons. Myers-Johnson and Everitt (2022) argue that, while Black Canadians are generally over-represented among the candidate pool, their continued under-representation among successful candidates is because parties tend to field them in weak or lost-cause ridings. This pattern held for all parties in the 2021 election, though less so for the Liberals, who ran several of their Black candidates in winnable strongholds.

Unsurprisingly, Black candidates were concentrated in 3 of the 4 most populous provinces: Ontario (with 51.4% of all Black candidacies), Quebec (31.4%), and Alberta (12.9%). These numbers reflect two factors. First, they roughly follow the distribution of Canada’s 338 ridings, though the share of Black candidates was lower than the provincial seat share in Ontario and higher than expected in Quebec. Second, they reflect the fact that Canada’s Black population comprises a sizable population share in very few ridings. The largest concentrations are in 12 ridings in Ontario and Quebec (see Table A1). In all, 23 of the 70 Black candidates ran in these 12 ridings, with 2 winning their seats (Emmanuel Dubourg in Bourassa and Ahmed Hussen in York South—Weston). Excluding these 12 ridings, the remaining 47 Black candidates contested ridings where the Black population share on average was only 4.4%.

Finally, Black candidates in 2021 were a socially heterogeneous group. Based on an analysis of their campaign websites and Twitter communications, we identified more than half (55.7%) as first-generation immigrants while less than one-third (30%) were Canadian-born (the birthplaces of the remaining 14.3% are not known). Those born outside of Canada claimed diverse national origins across more than 20 countries throughout the Caribbean and sub-Saharan Africa, plus some European countries. They also exhibited gender diversity: 50% of Black candidates were women, 48.6% were men, and 1.4% were non-binary. Another 4.3% of Black candidates openly identified as LGBTQ+.

Racial identity and Black candidates’ communication strategies

The contextual factors discussed above hint at competing scenarios. On the one hand, a low Black population share in most ridings, together with the institutional constraints of party discipline, make it unlikely that race would figure prominently in the communication strategies of Black candidates. On the other hand, the heightened salience of racial issues during the 2021 Canadian election makes it more likely that we will see racial messaging, though candidates’ partisan affiliation is expected to play a key mediating role.

The first scenario suggests that Black candidates in Canada will use a deracialization strategy. Originally conceived of as a campaign tactic (Hamilton 1977), deracialization involves racialized candidates purposely avoiding explicit references to race and emphasizing issues of concern to all racial groups in an effort to build crossover support and encourage voter mobilization (Collet 2008; McIlwain 2013; Orey and Ricks 2007). Black candidates following this strategy would avoid discussing Black-related positions on issues like policing, welfare, and affirmative action programs and instead emphasize common issues such as the environment, economy, and taxes. This electoral strategy was first advanced in the 1970s by Democratic candidates who hoped to make it harder for Republicans to use race as a wedge issue (Hamilton 1977). A decade later, several African American candidates were shown to use deracialization in their successful mayoral and gubernatorial bids (McCormick 1990; McCormick and Jones 1993; Pierannunzi and Hutcheson 1991; Summers and Klinker 1991). Since then, studies in the U.S. show that while deracialization is a key electoral strategy, winning crossover support is especially challenging for Black candidates in black-white contests where they may face institutional, individual, and cultural racism (Burge et al. 2020; Gillespie 2016; Juenke and Sampaio 2010; Strickland and Whicker 1992; Ward 2016). This task is a challenging one because of the need to avoid triggering white backlash while simultaneously energizing high Black voter turnout at the polls. ⁴

However, not all minority candidates use a deracialization strategy. Charlton McIlwain (2013) asserts that some candidates use a racial distinction strategy to convince voters to support them because of their race, not in spite of it. Racial distinction is similar to deracialization in two ways: (1) candidates minimize any racial or value differences between themselves and voters who do not share their racial identity and (2) candidates emphasize issues of interest to all voters. But racial distinction differs in that candidates deliberately draw attention to their own racial identity. McIlwain argues that Barack Obama used a racial distinction strategy to construct “a seamless racial narrative — deployed through constellations of subtle racial language and imagery — that incorporated Obama’s own personal biography within a broader narrative of the nation, specifically a narrative of American progress” (2013: 120). In other words, Obama subtly used race to his advantage in his bid for the American presidency.

Alternatively, Christian Collet (2008) posits that minority politicians might toggle, or simultaneously pursue a racialization strategy with co-ethnic voters and a deracialization strategy with others. Collet defines toggling as “a strategy that seeks to build or maintain a victorious electoral coalition in a potentially polarized, multicultural environment through the balanced communication of broad- (i.e., ‘mainstream’) and narrow-cast messages, symbols, issue positions, personal characteristics and socio-cultural cues to specific racial and ethnic groups” (2008: 712). Toggling seems to be facilitated when candidates can appeal to co-ethnic voters in a different language using mailouts and ethnic-language media. It thus seems less useful than deracialization for understanding how Black candidates in Canada campaign via a mainstream social media platform. ⁵

The second scenario speaks to political parties as a key factor for understanding the role of race in the communication choices of Black candidates in the Canadian context. The U.S. has a weak-party, strong-candidate system. Although Democrat and Republican candidates must operate within the ideological boundaries of their respective parties, they have considerable control over their campaign and communication strategies. In contrast, Canada has a strong-party, weak-candidate political system. Decision-making power is concentrated in the hands of political parties in general and the party leader in particular. Candidates are mainly brand ambassadors for the party and are expected to reproduce the party’s main talking points in their campaign communications (Marland and Wagner 2020). This expectation applies whether the candidate is an incumbent or challenger, cabinet minister or backbencher, long-time politician or newcomer. Candidates who do not toe the party line face discipline and might not be allowed to run for the party in the next election. Considering that Canadian voters generally do not favour independents or candidates who switch parties for electoral reasons (Snagovsky and Kerby 2018), this punishment effectively brings the politician’s federal career to an end.

Party discipline means candidates within the Canadian party system do not have the same level of local control over their campaign strategies as their American counterparts do. But, of course, parties themselves vary in the degree to which they foreground the needs and interests of diverse social groups. In the U.S., with its sharp ideological and partisan cleavages, the share of racialized minorities among Democratic office holders has moved steadily upward since the early 1990s, while the same is not true for Republicans. The two parties also take strikingly divergent positions on policies regarding minority rights. In this context, racialized minority candidates for the Republican Party face challenges in developing their local campaign platforms, as they need to craft a public persona authentic to who they are as individuals while conforming to the Republican brand (Cooperman and Crowder-Meyer 2018; Fields 2016; Kitchens and Swers 2016; Och 2018, 2020; Wineinger 2021). By comparison, candidates for the Democratic Party enjoy far more latitude to claim their social identities as political assets.

Canadian parties also differ in the degree to which they foreground the interests of diverse social groups. The left-wing New Democratic Party (NDP) prioritizes the interests of the working class and, to a lesser extent, women, sexual minorities, Indigenous peoples, and racialized communities (DeGagne 2019; Lexier 2021). In striving to diversify its slate of candidates, it requires local riding associations to recruit diverse nominees before signing off on their selection (Ashe 2021). NDP candidates therefore have considerable leeway to emphasize race in their campaign communications. In contrast, the right-wing Conservative Party rejects what it calls “identity politics” and declines to take any formal steps to diversify its candidate slate (Thomas 2017). It assumes a more socially conservative stance on issues like abortion and same-sex marriage (Boily and Robidoux-Descary 2019; Farney 2012) and consistently lags behind the other main parties when it comes to nominating female candidates. With respect to immigrants and ethnic minorities, however, it has taken a more nuanced approach. Beginning in the mid-2000s under leader Stephen Harper, Conservative Party efforts to “unite the right” resulted in strategic outreach to immigrant communities whose members were understood to lean right on social and economic issues. The Conservative Party has been reasonably effective in attracting racially diverse candidates (Black 2020), though like other parties it has had less success with Black candidates.

The centrist Liberal Party shifts its issue positions depending upon the electoral threat posed by the main parties to its left and right (Pétry 2014: 155). The party includes identity-related issues in its platform, but is perceived to do so more for strategic reasons than out of genuine commitment to addressing social inequalities (Bird 2019; Findlay 2022). The Bloc Québécois is a sovereigntist party that only contests seats in Quebec with the aim of protecting the province’s interests at the federal level. Identity issues have long been at the heart of Quebec nationalism (Béland et al. 2021; Sharify-Funk 2011; Turgeon et al. 2019). During the 2021 federal election, the BQ found itself having to defend the Quebec government’s decision to pass Bill 21, legislation that expressly forbids government employees from wearing religious symbols at work. The law applies to everyone in the province, but is perceived as mainly affecting Muslims and Sikhs, who also tend to be racialized minorities. The Green Party, for its part, tends to steer away from identity-related issues and focuses instead on environmental concerns. Notwithstanding this, the Green Party faced its own turmoil as leader Annamie Paul, a Black Jewish woman, confronted internal party challenges to her leadership that she claimed were driven by racism and sexism.

Our study asks whether race is an aspect that Black candidates in Canada try to optimize or minimize in the course of an election campaign. The constraining features of the Canadian parliamentary and party system notwithstanding, we expect that partisan differences will produce considerable variation in candidates’ communication strategies. Black Conservative candidates are expected to pursue a deracialization strategy that bypasses racial issues in favour of generic messages; however, more nuanced strategies might be evident among other parties.

Methodology

This study stems from a larger project exploring the social media strategies of women, Indigenous, racialized, and LGBTQ+ candidates in the 2021 Canadian election. To understand the role of race in candidate communications, this paper focuses specifically on the Twitter activity of 19 Black candidates who sought a seat in Parliament. The candidates were selected to represent different genders, sexualities, ethnicities, incumbency status, political positions, and partisan affiliations. Our intersectional analysis probes the data to determine what role, if any, other factors might play in a Black candidate’s decision to use a (de)racialization strategy in their campaign communications.

Of the 19 candidates, 11 are female, 7 are male, and 1 is non-binary while 2 are also publicly declared members of the LGBTQ+ community. Ethnicities include, but are not limited to, Trinidadian, Haitian, Nigerian, Somalian, and Jamaican, while religious faiths are mainly Christianity, Islam, and Judaism. Only 3 of the candidates we study were incumbents while the rest were challengers. The incumbents featured a cabinet minister, parliamentary secretary (an assistant to a prime minister or cabinet minister), and a backbench member of Parliament. The challengers included the leader of the Green Party, a minor but increasingly important party. In terms of partisan affiliation, 8 candidates ran for the NDP, 4 for the Liberals, 4 for the Conservatives, 1 each for the Green Party and Bloc Québécois, and 1 as an independent. The candidates competed in ridings located primarily in Ontario and Quebec with 1 in Manitoba. See Table A2 for descriptive information about the candidates.

Twitter is a useful medium for examining candidate self-presentations. The platform is popular among politicians seeking to quickly broadcast information about themselves, their policies, and their activities. Twitter also offers a cost-effective way of disseminating one’s campaign message and is especially effective in reaching journalists, who routinely monitor it in the course of campaign coverage. Twitter’s 280-character limit constrains what politicians can say but such brevity coincides with legacy media’s preference for ever-shorter soundbites.

As a source of data, however, Twitter has some drawbacks that could influence the analysis. Twitter users are not representative of the general population. They tend to be younger, male, well-educated, financially better off, and more politically interested. In other words, they are more likely to “reflect the interests, values, skills, priorities, and biases of elites” (Blank 2017: 691). Nor are tweets representative of general public discourse. Compared to communications on other platforms, they tend to hew more to the ideological extremes. Researchers should exercise caution, especially in using Twitter data to assess public opinion (Barberá, and Rivero 2015; Blank 2017; Blank, and Lutz 2017; Mellon, and Prosser 2017). Our study of Black candidates is largely exempt from these concerns because we are not trying to assess public opinion toward these candidates. However, one possibility is that candidates might pitch their message to appeal to the “youthful” social demographics of Twitter users—in which case, we would expect greater attention to anti-racism and other progressive issues, when compared to traditional communication platforms. On the other hand, the capacity of the parties to monitor social media might lead candidates to exercise more constraint in how they talk about race.

Our data set includes all tweets, quote tweets, and retweets issued by selected candidates from July 20, 2021 to September 30, 2021. This timespan, which exceeds the unusually short 36-day election campaign (August 15 to September 20), was selected to explore the possibility that identity issues might be discussed differently before or after the official campaign period. While we initially selected 19 Black candidates for study, by the end of the campaign, 4 of these candidates either did not have a Twitter account or did not issue any tweets during the election. Our analysis therefore focuses on the Twitter activities of 15 candidates. For the purpose of this analysis, all French-language tweets were machine translated into English. A total of 3,314 tweets were analyzed.

We apply critical discourse analysis (CDA) to explore more closely how Black candidates deployed their racialized identities within a local campaign. We invoke three of the core theoretical tenets of CDA in addressing our research question. First is the general presumption that a political actor’s discourse is structured according to different patterns across various domains of social and political life (van Dijk 1997). In this case, we focus on campaign tweets, which we assume to be structured by constraints and norms of Twitter, and also patterned in ways different from parliamentary discourse or private interpersonal discussions among individuals. Second, we accept that discourse is implicated in relations of power and persuasion (Fairclough 2003; Wodak 1989). Given the centralization of the Canadian party system, parties wield power over what candidates can say, though the limits of what is acceptable for a Black candidate to say about race might differ across parties and local contexts. Candidates themselves might vary in the power they hold vis-à-vis their party; notably, an incumbent cabinet minister might face fewer constraints than a first-time candidate. Thirdly, CDA asserts that actors do not neutrally reflect the world, identities, and social relations but rather play an active role in creating them (van Dijk 1993; Wodak 1989). We thus investigate ways in which Black candidates discursively push back against party constraints and conventions. We aim to uncover what such discursive acts of resistance (or compliance) look like and what they are intended to do. While our ability to infer power relations is limited because of the brevity of Twitter communications, our use of CDA can shed light on the way that Black candidates navigate these challenges and offer hypotheses that can be tested in quantitative research.

Twitter data were analyzed in two stages, following Yin (2011), who describes coding as an iterative process that involves compiling, disassembling, reassembling, and interpreting data. First, we created separate Excel files of the Twitter activity for each of the 15 candidates. Tweets were then read and hand-coded in a reflexive, inductive fashion focused on meaning, context, and timing of how candidates discussed social identities and related issues over the 9-week period of data collection. This approach enabled us to examine shifts in candidates’ communication styles as well as analyze differences among candidates, especially in terms of their party affiliation and political status.

The second stage of analysis was designed as a data check and involved importing all of the candidates’ tweets into MAXQDA, a software for qualitative and mixed-methods analysis. We conducted a series of lexical searches to identify all mentions related to social identities and related issues. Race mentions included terms such as Haiti, Somalia, Trinidad, Black, India, Afghan(istan), and other country names. Indigenous mentions included not only the word Indigenous but also Aboriginal, residential schools, and reconciliation. Gendered content was identified via words such as gender, woman/women/womxn, girls, feminism/feminist(s), childcare, abortion, reproductive rights, and gun(s). Sexuality was captured through terms such as LGBTQ+, trans, conversion therapy, and blood ban. ⁶ All mentions of religion were captured, including specific religions (such as Judaism and Islam), religious sites (church, mosque, temple), and religious events (Yom Kippur). Disability mentions included the word disability, disabilities, and PWD (people with disabilities) as well as specific conditions such as autism. To identify other mentions not covered by the lexical searches, we read through each individual tweet. This fine-grained coding uncovered mentions of notable Black individuals and organizations. The MAXQDA data were used to strengthen the intersectional analysis. Table A3 provides more details on how the different categories were coded.

Results

Our CDA revealed that Black candidates mainly used race in a strategic manner during the 2021 Canadian election, undertaking a racial distinction strategy only when campaign conditions permitted but otherwise following a deracialization strategy in which they discussed general issues like the economy. In essence, Black candidates behaved like other candidates in acting as brand ambassadors for their party in the local riding. Most parties emphasized issues of concern to senior citizens, a group that votes in much higher numbers than other generational cohorts but especially young adults (Gidengil et al. 2004). Youth-related issues like student loans were rarely discussed in comparison to more candidate-focused considerations like problems with on-campus voting. Candidates for all but the Conservative Party vowed to address the climate change emergency, while Liberal and NDP candidates also gave strong attention to affordable housing, pandemic responses, health care, and economic matters. Conservative candidates overwhelmingly tweeted about campaign events and get-out-the-vote activities. When they did tweet about policy, they focused on typical party topics such as seniors, taxes, and the economy.

Candidate position-taking on various issues typically followed the party line, but it was clear from the language in some tweets that many candidates strongly supported their party’s platform. NDP candidates drew upon their own personal experiences when discussing an issue. For example, Michelle Spencer, who ran for the NDP in the Ontario riding of Scarborough—Guildwood, admitted in a four-part Twitter thread that she was facing a possible diagnosis of an invisible disability after having been tested, and ruled out, for multiple sclerosis years earlier:

…It was a terrifying time for me. At that time I was a young mother with two little ones (1) ⁷

and the thought that I might not be able to be actively engaged in their lives or be able to support them, was devastating. There are treatments for the diagnosis I may be facing, and it’s not life-threatening, but no known cures. Most of those treatments are not covered (2) ⁸

I am hoping for the best, while bracing for the worst. The #Liberals’ failure to include a plan for #Pharmacare in their platform is a slap in the face to all Canadians, but especially to those who have no choice but to cover the cost of potentially life-saving meds (3) ⁹

The #NDP is the only party that is committed to bring pharmacare to all Canadians. We need to do better, for each other (end). #cdnpoli #Elxn44 #ReadyForBetter ¹⁰

While most candidates simply repeated their party’s talking points, Twitter threads like this one demonstrate that some candidates are sincere brand ambassadors for their party. They have chosen to run for that party because they believe in its overall approach to policy.

Black candidates were reserved when it came to discussing their racial identity or racial issues during the election. They generally did not discuss their specific ethnicity or Black issues like policing and the criminal justice system. Nods to Black culture were subtle and strategically inclusive. NDP candidate Paul Taylor exemplified this approach. He tweeted about local restaurants and businesses in his Parkdale—High Park riding. But for a hashtag, an uninformed reader would be unaware that one of those restaurants served Jamaican patties ¹¹ while a vintage pop-up store offered albums by reggae acts. ¹² Other candidates tweeted about notable Black individuals such as Toronto Raptors president Masai Ujiri, who unveiled an art installation inspired by South Africa’s Nelson Mandela at Union Station in Toronto in August 2021. Black candidates more often took a multicultural approach to tweeting about race. Jan Doering describes this tactic as “multicultural moments” when “candidates celebrate the cultures of various ethno-racial groups to mobilize a broad range of voters” and argues it is used more often in Canada than the United States because of the latter’s greater racial polarization (2020: 1055). Black candidates sent congratulatory tweets to other ethnic groups celebrating milestones or making strong contributions to the community. The American withdrawal from and Taliban takeover in Afghanistan inspired a flurry of tweets from Black candidates urging the Canadian government to help individuals flee the country.

Black candidates were most comfortable explicitly addressing their race when world or campaign events made it appropriate to discuss such matters. An earthquake in Haiti near the start of the election campaign drew sympathetic tweets from most Black candidates, while anniversaries such as the International Day of Remembrance of the Victims of Slavery and the Transatlantic Slave Trade or Trinidad and Tobago’s Independence Day were marked by some candidates. Serious discussion of Blackness mainly took place when one of the major parties faced accusations of racism. When Conservative leader Erin O’Toole said at the start of the campaign that his party would “take inequality seriously,” Liberal cabinet minister Ahmed Hussen retorted that the Conservative platform did not outline specific plans to address racism, Islamophobia, and antisemitism. ¹³ Hussen, who ran for re-election in Ontario’s York South—Weston riding, then issued a nine-part Twitter thread outlining what the Liberal government has done and would do for Black individuals if it returned to power. These initiatives included addressing the over-representation of Blacks in the Canadian criminal system, launching a $200-million philanthropic endowment fund, a $50-million fund to help racialized journalists and creators, and diversifying the federal public service.

Hussen’s thread was the only clear instance (in our sample) of a Black candidate discussing Black-related policy at any length during the campaign, and three aspects of this are notable. First, the Somali-born Hussen was the only Black MP to have served in Trudeau’s cabinet between 2015 and 2021. And while his portfolios and mandate letters to that date (minister of immigration, refugees and citizenship; minister of families, children and social development) did not directly concern race, it would seem to augment the government’s credibility to have him address this policy commitment. ¹⁴ It is thus reasonable to assume that this tweet would have been fully vetted and perhaps even orchestrated by the party itself. Second, Hussen represents a riding with a large Black population share, which suggests that such a tweet would incur less risk of “backlash” and little threat to his re-election. Finally, it is remarkable the tweet did not explicitly reference Hussen’s personal identity but rather highlighted his party’s platform. As “exceptional” as this tweet is, we interpret it as supporting our hypothesis that Black candidates overwhelmingly choose to play it safe when it comes to tweeting about race.

We saw this pattern play out even when the candidate’s party faced accusations of racism. During an all-leaders debate at the mid-point of the campaign, BQ leader Yves-François Blanchet declared support for the Quebec government’s decision to pass Bill 21, legislation banning provincial civil servants from wearing religious symbols at work. Blanchet also rejected the idea that systemic racism was an issue in Quebec and accused critics of bashing the province. Marie-Eve-Lyne Michel, the lone BQ candidate in our study, did not defend her party against racism charges. As the challenger for Laurier–Sainte-Marie, a riding with a relatively low Black population share, she remained silent on the issue and kept her identity talking points focused on the BQ’s usual topics of protecting the French language and culture. Her lone mention of Quebec’s controversial Bill 21 came days before the debate, when she retweeted another BQ candidate’s comment that their party was the only one to defend secularism. ¹⁵ For their part, Conservative Black candidates were similarly silent when their leader was under attack for the party’s weak platform on addressing discrimination and racial inequality.

Of the 15 candidates we examined, only 2—the Liberals’ Hussen and NDP challenger Camille Esther Garon in the Quebec riding of Beauport–Limoilou—explicitly addressed their party’s or leader’s position with respect to Black and racialized Canadians and anti-Black racism. Following the leaders’ debate referenced above, Garon praised her party leader Jagmeet Singh for his composure in the face of the BQ’s accusations of Quebec bashing:

Quand je vois Jagmeet ne pas avoir tombé dans le piège de Blanchet concernant la question du racisme et Jagmeet sera toujours la pour dénoncer les enjeux des minorités au Québec et le Québec Bashing. Les jeunes Noir.e.s et Autochtones ont été entendu ❤¹⁶

While other Black candidates mentioned once or twice the need to address racism in Canadian society, Hussen and Garon discussed it at length. Others also broached the issue, though often in the form of retweeting other people’s comments.

Several references to identity were made by Black women candidates calling for greater diversity in Canadian politics. Marci Ien, who sought re-election in Toronto Centre, participated in an Instagram event with other two other Liberal Black women candidates. Annamie Paul, the Green Party leader who unsuccessfully contested the riding of Toronto Centre, drew attention to her identity during the English-language leaders’ debate. She noted that “I’m proud to be here tonight. I may be the first of my kind, but I know I won’t be the last.” ¹⁷ This might have been a strategic choice on Paul’s part not to explicitly mention race, but rather let audiences draw their own inferences as to whether she was referencing her identity as Black, Jewish, or female. In another rare discussion of racial identity, the BQ’s Mare-Ève-Lyne Michel expressed a similar desire for Quebec political life to become more diverse:

Fille d’un père Haïtien et d’une mère Québécoise, je crois fermement que nous, personnes issues de la #diversité, devons ns investir dans la vie politique si on veut que le Québec, que se construit un peu plus tous les jours, nous ressemble aussi. #BlocQC ¹⁸

This passage sees Michel signalling her mixed identity as a francophone and a daughter with Haitian and Quebecois roots. As we have discussed, racism is a touchy subject in Quebec so the intersection of race and gender might open opportunities to address identity in ways that cast a broader appeal.

While their wish for more diverse representation is no doubt sincere, Black women also have an instrumental motive, and perhaps an advantage, in addressing these two facets of their identity in a way that does not too overtly or exclusively emphasize race. Finally, while Black candidates were generally reluctant to address Black identity issues, they had no such reticence when it came to Indigenous issues. As Canada begins to reconcile centuries of state-sponsored atrocities against its First Peoples, Indigenous issues are viewed not as the sole concern of one racial group but rather of the entire country. In light of archeological confirmation of unmarked graves of Indigenous children on former residential school sites in the weeks leading up to the federal election, candidates of all backgrounds tweeted about Indigenous matters. But as with other identity-related topics, we found partisan differences here as well. Black Conservatives were most silent on Indigenous issues, and NDP candidates were most vocal, while Black Liberal candidates fell in-between.

Discussion and conclusions

In Canada’s party-centred political system, voters typically cast their ballot for a party rather than for a candidate while parties play a central role in articulating campaign platforms and themes. Candidates, for their part, are largely brand ambassadors with strict limits on campaign spending and little margin of manoeuvre from their leader’s messaging.

With this in mind, a CDA of campaign Twitter feeds found Black candidates generally used a deracialization strategy in the 2021 Canadian election—except when campaign events provided political cover to address racial identity and racial topics. In essence, Black candidates behaved like other federal candidates in that they acted as brand ambassadors for their party in the riding, emphasizing their party’s stance on major issues like affordable housing, climate change, and pandemic responses rather than highlighting specific Black positions on issues like anti-racism or policing. The Canadian political context thus creates challenges for Black candidates who want to discuss their racial identity and racial issues, but these are not insurmountable. Some Black candidates took advantage of campaign events to highlight issues of concern to the Black community and other identity groups without fear of backlash or accusations of being a single-issue candidate. Other candidates found ways to subtly signal race to Black voters, such as talking about a Black musical group, restaurant, or public figure. Black women, furthermore, could more “safely” address the need for greater diversity in parliament without explicitly referencing their racial identity.

Party discipline notwithstanding, we found that partisan ideology mediated the degree to which Black candidates were comfortable discussing race. The Conservative Party and Bloc Québécois reject the idea that systemic racism is a problem in Canada, leading their Black candidates to follow a deracialization strategy. They rarely discussed race in their Twitter communications, not even to defend their parties from accusations of racial bias. In contrast, social and racial inequalities are a prominent issue on the NDP platform, and this was reflected in their candidates’ greater willingness to use a racialization strategy. Liberal candidates fell between these two extremes.

Electoral incentives appear to play a minor role in how Black candidates in Canada talk about race, at least in the broadcast format of the Twitter platform. In the vote-rich big-city ridings where Black candidates tend to run, they need to appeal to a multi-racial electorate and thus avoid narrowly targeting any single group via their tweets. We anticipated that the youthful social demographics of Twitter users might provide greater scope for Black candidates to foreground their racial identity or address the “progressive” topic of anti-Black racism when compared to traditional broadcast platforms such as newspaper, radio, or television. Preliminary evidence offers little support for this hypothesis. However, we are undertaking further research to assess whether these candidates’ communication strategies vary across different media platforms.

Our examination of the Twitter communications of Black candidates in the 2021 Canadian election makes several substantive contributions to the literatures on deracialization, political communication, and Canadian politics. First, despite Canada’s image as a multicultural nation, our study illustrates the constrained political space in which Black candidates can address their racial identities and racial issues—even in the aftermath of the George Floyd murder and worldwide attention to #BLM. Second, by applying the deracialization and racial distinction theories to Black politicians in a non-U.S. setting, our findings reveal the importance of institutional context. In particular, we see that in Canada’s Westminster-style parliamentary system, party discipline, and partisan ideology are important mediating factors in shaping the local communications strategies of Black candidates. Black candidates have recently gained ground in elections in Canada and other countries that share similarities with the Canadian parliamentary system (e.g., Britain, New Zealand). This is a positive development. Nevertheless, our study raises important questions about these candidates’ capacity to address race issues during their campaigns as well as the fate that awaits them if they assume a more vocal stance once elected.

Finally, our small sample of 15 Black candidates and our intensive focus on one social media platform (Twitter) has allowed us to examine closely the nuances and subtle shifts in how Black candidates address race. Our qualitative methodological approach, and especially the application of CDA, is well-suited to probing the discursive strategies of Black candidates, given both the novelty of this research in Canada and the more “coded” way in which Canadians tend to talk about race and racial issues compared to the U.S. While this approach obviously limits our ability to reach generalizable conclusions, the insights generated should be considered as a starting point for future work.