Defining Reconciliation

The concept of reconciliation is frequently applied in contexts where the state has a long and ongoing history of large-scale human rights violations toward selected social groups (El-Masri et al. Reference El-Masri, Lambert, Quinn, El-Masri, Lambert and Quinn2020: 3; Walker Reference Walker2018). Reconciliation emerges, either as an ongoing process or as an end state, when governments adopt policies and processes that try to rectify these violations. The goal is to repair relationships between the state and its citizens by removing the structural causes of discrimination and inequality within the country, restoring dignity to marginalized groups, and empowering them to become full and equal citizens (Seils Reference Seils2017: 1; Murphy Reference Murphy2017). The result is a perpetual state of “non-alienation”, in which the state and society actively protect all “members against social relationships and structures in which inequalities in power, status differences, institutional mechanisms, social practices or cultural artifacts” lead to “the explicit or systematic disrespect of the affected person’s (or group’s) status as socially free and equal” (Schuppert Reference Schuppert2015: 451). To achieve non-alienation in the social world, reconciliation processes operate at multiple levels, including the individual (e.g. victims coming to terms with the trauma they experienced), the interpersonal (e.g. victims and perpetrators working together to repair relationships), the socio-political (e.g. reconciliation between groups), and the institutional (e.g. the restoration of trust between the state and its citizens) (Seils Reference Seils2017: 4–6).

Canadian efforts at Indigenous reconciliation reflect many of these sentiments and have evolved over time. Early efforts at reconciliation were assimilative and colonial in nature and focused on non-alienation (e.g. Indigenous peoples are no different than any other social group) (Flanagan Reference Flanagan2008). Over time, Canadian policies shifted to recognizing Indigenous peoples as “Citizens plus” (e.g. Indigenous peoples are Canadian citizens with special rights) (Cairns Reference Cairns2001) and then as nations deserving of greater control and autonomy over their lives (RCAP 1996). Indeed, current federal reconciliation policies talk about the need to work “in partnership with Indigenous Peoples to address past harms, support strong and healthy communities, and advance self-determination and prosperity” (Government of Canada 2023). From the perspective of the Crown, however, reconciliation can only occur under the presumption of Canada’s unquestioned and unchallenged sovereignty over all lands and peoples within its borders. There is no room for co-sovereignty with Indigenous constitutional orders (Ladner Reference Ladner, Asch, Tully and Borrows2018). Global Affairs Canada, for instance, defines reconciliation as working with Indigenous communities to foster “strong, healthy and sustainable Indigenous nations within a strong Canada” (emphasis added), meaning reconciliation can only occur within the constitutional framework of Canada (Canada 2022; McCrossan and Ladner Reference McCrossan and Ladner2016).

In contrast to these conceptions of reconciliation, many Indigenous scholars and their allies argue that reconciliation must go beyond non-alienation towards an ongoing process in which Indigenous peoples, settlers, and the state work together to create sustainable and positive relationships with each other, grounded in Indigenous traditions and constitutional orders (Borrows Reference Borrows2010; Borrows and Tully Reference Borrows, Tully, Asch, Tully and Borrows2018). According to Tsartlip Elder May Sam, “reconciliation for her people is a relationship of mutual respect and sharing with Mother Earth and all our kin – human and more than human—that we have responsibilities to acknowledge and enact in every breath and step we take” (Borrows and Tully Reference Borrows, Tully, Asch, Tully and Borrows2018: 13). This means that settlers and the settler state should return land back to Indigenous peoples and empower and give them the space to restore and strengthen their political, economic, and social systems (Alfred Reference Alfred2023; Coulthard Reference Coulthard2014). It also means transforming the Canadian state so that Indigenous constitutional orders are recognized as equal to the Canadian one, not only in the courtroom, but also in its political, economic, and social arenas (Borrows Reference Borrows2010). Further, it means rejecting any sort of “convergence of nations” approach to instead embrace activities that transform our collective consciousness about Indigenous peoples and the historical and ongoing harms that the settler state and settlers have and continue to perpetuate towards them.

In short, the concept of reconciliation inhabits a continuum. At the one end is non-alienation or assimilation, in which Indigenous peoples are treated no differently than any other group, and so they do not have any special programs or rights that are attached to their nationhood or claims to land (Flanagan Reference Flanagan2008). At the other end is Indigenous resurgence, in which Indigenous communities are independent entities with their own constitutional orders and economic and social systems, derived not from the state but rather from their own rights to self-determination (Couthard Reference Coulthard2014). In between these extremes are reconciliation models such as “Citizens Plus”, in which Indigenous people are first and foremost citizens but with additional rights that recognize their Indigeneity (Cairns Reference Cairns2001), and nation-to-nation (RCAP 1996), in which Indigenous communities are province-like entities with significant autonomy that other social groups do not enjoy, but which still exist within the constitutional order of the settler state. In all cases, except for non-alienation, meaningful reconciliation requires the state to recognize the traumatic experiences it inflicted on Indigenous peoples before engaging in processes and policies that lead to the recognition and protection of Indigenous identity (TRC 2015). As Ladner (Reference Ladner, Asch, Tully and Borrows2018: 249) notes, settlers/citizens need to come “to terms with those great myths that deny the past, and in so doing continue to deny Indigenous humanity and rights, both in the past and present” and Canada must “confront its mythologized exceptionalism … [which] constructs Canada as the good colonizer, a peaceful nation that did not engage in Indian wars but has instead always dealt justly with Canada’s Indigenous peoples” (see also Frank et al. Reference Frank, Kelly and Mitchell1983; Macklem Reference Macklem2001).

Attitudes Towards Reconciliation

Despite the growing salience of Indigenous reconciliation, we know relatively little about settlers’ attitudes toward Indigenous peoples. What we do know is that Indigenous peoples frequently endure many instances of prejudice, pernicious stereotypes, and discriminatory behavior across settler societies (Environics 2010; Findling et al. Reference Findling2019; González et al. Reference González, Carvacho and Jiménez-Moya2022). Much of the literature on attitudes towards out-groups stems from the United States, rooted in work by Kinder and Sears (Reference Kinder and Sears1981) on symbolic racism, among others (see also, Henry and Sears Reference Henry and Sears2002). In the United States, most studies have focused on categorizing and describing these instances and expressing Native American feelings about them in a broad range of political and social spaces (Findling et al. Reference Findling2019; Robertson Reference Robertson2015; Williams Reference Williams2007). In contrast, only a handful of studies (Foxworth and Boulding Reference Foxworth and Boulding2022) have focused on public attitudes towards Native Americans, and when they do, the focus is usually on public opinion in relation to the mascots of college and professional sports teams (Williams Reference Williams2007). Jimenez et al. (Reference Jimenez, Arndt and Helm2023), for instance, found that prejudicial attitudes towards Native Americans increased substantially after Cleveland’s major league baseball team decided to retire its Native American mascot, Chief Wahoo.

In Australia and Canada, it is only relatively recently that scholars have turned to empirically documenting the existence of anti-Indigenous sentiment. In Australia, Shirodkar (Reference Shirodkar2019) found that Australians broadly held negative unconscious views about Indigenous peoples in their country but that these opinions are becoming more positive over time (Levy and McAllister Reference Levy and McAllister2022). Similarly in Canada, recent papers by Beauvais (Reference Beauvais2021, Reference Beauvais2022) and Harell et al. (Reference Harell, Soroka and Ladner2014, Reference Harell, Soroka and Iyengar2016) found evidence of strong resentment toward Indigenous peoples, but this resentment may be decreasing over time. Figure 1 depicts these trends by presenting a short scale of Indigenous resentment over the course of six surveys, one in 2012 (Harell et al. Reference Harell, Soroka and Iyengar2012) and five conducted between 2019 and 2021 (Harell et al. Reference Harell, Stephenson, Rubenson and Loewen2020, Reference Harell, Stephenson, Rubenson and Loewen2021, Reference Harell, Stephenson, Rubenson and Loewen2022; Stephenson Reference Stephenson, Harell, Rubenson and Loewen2020, Reference Stephenson, Harell, Rubenson and Loewen2023).Footnote ² It shows that feelings of resentment seem to have decreased over time, with a large dip in late May/early June 2021 (e.g. 2021 Democracy Checkup (DC) data were collected between May 20–June 7) and a significant increase in the fall of 2021 when the 2021 Canadian Election Study data were collected. The dip in 2021 coincided with the discovery of multiple mass graves at the sites of former residential schools (Gulliver Reference Gulliver2021), suggesting that making the consequences of residential schools salient may have helped to decrease resentment towards Indigenous peoples (see also Williamson, Reference Williamson2024).

Figure 1. Indigenous resentment over time.

In terms of the substance and nature of these negative attitudes, they are often grounded in colonialism and a lack of knowledge and understanding of the current realities of Indigenous peoples. For example, Fryberg and Eason (Reference Fryberg and Eason2017) suggest that contemporary prejudice toward Native Americans is often grounded in “commissions” of romanticized stereotypes and “omissions” that render Native people largely absent from contemporary media coverage and settlers’ own awareness and contact (see also Davis-Delano et al. Reference Davis-Delano, Folsom, McLaurin, Eason and Fryberg2021). In terms of contemporary attitudes grounded in the on-going experience and consequences of colonialism, Foxworth and Boulding (Reference Foxworth and Boulding2022) document how resentment toward Indigenous peoples is often framed in terms of Native Americans having special rights that are unfair or unequal, such as the right to build and manage casinos on their territories. In Canada, Beauvais (Reference Beauvais2021, Reference Beauvais2022) argues that Indigenous resentment differs from other forms of prejudice as it is often grounded in conflicts over Indigenous lands, language rights, and jurisdictional autonomy, all of which are rooted in historical and ongoing processes of colonization. In New Zealand, negative attitudes toward the Māori are similarly grounded in ideologies of “colonization, land alienation, and political inequality” because these are “undeniable features of post-colonial societies that must be addressed (and mitigated) by ideologies developed by the ruling class” (Sibley and Osborne Reference Sibley and Osborne2016, 117; see also Satherley and Sibley Reference Satherley and Sibley2018).

In sum, negative attitudes toward Indigenous peoples are often grounded in perceived conflicts that emerge out of colonization as well as intergroup conflict more generally. Yet, we know surprisingly little about how settlers feel specifically about addressing the legacies of these processes through reconciliation initiatives. The one exception we are aware of is Campo et al. (Reference Campo, Mastin and Somjen Frazer2004), which found that while race predicted American attitudes towards certain kinds of reparations for slavery (e.g. compensation and community-based programs for Black individuals), race failed to predict support for reparations for other racial groups, including Japanese Americans, Holocaust survivors, and Native Americans. Other studies have looked at the relationship between public support for Indigenous reconciliation and attitudes about nature and levels of critical historical knowledge, finding that they are correlated, but these studies rely on university student samples ranging in size (approximately 20, 200, and 1200; Neufeld et al. Reference Neufeld, Starzyk, Boese, Efimoff and Wright2022; Starzyk et al. Reference Starzyk, Fontaine, Strand and Neufeld2021; Woolford et al. Reference Woolford, Starzyk, Sinclair, Muller and MacDonald2022), which means their generalizability is likely limited. The relatively small number of studies on this topic is surprising given that in many settler contexts, such as in Australia, Canada, New Zealand, and the United States, Indigenous issues have become salient to different degrees, especially as Indigenous communities have become more successful at politically mobilizing, organizing, and negotiating institutional and policy changes that strengthen their communities (Miller Reference Miller2018; Hobbs Reference Hobbs2020; Wilkins and Stark Reference Wilkins and Kilwetinepinesilk Stark2010).

While the literature on public opinion towards Indigenous reconciliation is scarce, the same is not true for research on reconciliation attitudes towards other marginalized, non-Indigenous populations. Some studies have developed rich descriptions of how citizens conceive of reconciliation (El-Masri et al. Reference El-Masri, Lambert, Quinn, El-Masri, Lambert and Quinn2020). Rettberg and Ugarriza (Reference Rettberg and Ugarriza2016), for instance, found that Colombians prefer reconciliation to be targeted towards improving social relations more generally rather than targeting particular groups for redress. Colombians want the state to provide perpetrators with the opportunity to accept responsibility and blame, rather than subject them to punitive punishments. Finally, Colombians seem “to care more about their well-being and their relations with others in their immediate context than for abstract processes of rebuilding historical memory at a central level. In fact, only a few respondents explicitly explained reconciliation as a process of dealing with the past or the future” (Rettberg and Ugarriza Reference Rettberg and Ugarriza2016: 531–32).

Other studies have focused on assessing the determinants of support for the general idea of reconciliation. James Gibson’s (Reference Gibson2004) research on South Africa has found that those who acknowledge and accept the truth about their country’s ignoble past are more likely to hold more positive attitudes towards reconciliation. Aguilar and colleagues (Reference Aguilar, Balcells and Cebolla-Boado2011: 1398) examined whether a broader range of individual factors (e.g. age, interest in politics, education, religiosity, and ideology), socialization experiences (whether family identity was linked to the previous regime, politics discussed at home, ideology of father, and whether the family or individual was a victim), and contextual factors (whether a region was strongly repressed due to their linguistic or cultural identity) influence attitudes toward creating truth commissions, prosecuting perpetrators, and removing all symbols tied to the Franco regime. They found that “ideology and religiosity are decisive determinants of support”; specifically, highly religious conservatives were less likely to support transitional justice (TJ) efforts (Aguilar et al. Reference Aguilar, Balcells and Cebolla-Boado2011: 1418–19). Age and education only had modest effects, with older respondents (and especially those in rural settings) less supportive of TJ policies. In terms of socialization experiences and contextual factors, they found that family ties to the previous regime and whether you lived in a culturally/linguistically distinct region reduced public support for TJ measures (Aguilar et al. Reference Aguilar, Balcells and Cebolla-Boado2011: 1419; see also Bobowik et al. Reference Bobowik2017).

Another stream of research on reconciliation has focused on developing models that predict support for different kinds of reconciliation initiatives, including political apologies, symbolic (e.g. museums and monuments) and financial reparations, and public policies that purport to eliminate all forms of discrimination or which mandate the preferential treatment of marginalized groups. In most studies, the dominant social group is more likely to support symbolic gestures and policies that promote non-alienation and less likely to support distributive and preferential reconciliation measures (Campo et al. Reference Campo, Mastin and Somjen Frazer2004; Torpey and Burkett Reference Torpey and Burkett2010). Reichelmann and Hunt (Reference Reichelmann and Hunt2022), for instance, find that variation in white American attitudes towards reparations for slavery in 2016 depended mainly on the type of reparation being considered, with respondents more likely to support symbolic efforts (e.g. memorials) and policies of non-alienation in the workplace, and less likely to support financial reparations and preferential treatment policies for Black Americans. Other studies suggest that these effects are likely driven by in-group pride and perceptions of deservingness for the marginalized, aggrieved group (Reichelmann et al. Reference Reichelmann, Michael Roos and Hughes2022). Similarly, Bobowick et al. (Reference Bobowik2017) found that support for institutional apologies in three South American countries depends on the extent to which victims and non-victims perceive the structural conditions in their country to be favourable to collective hope and security. In short, the effects of different kinds of reconciliation policies may be overcome by individual and contextual factors (Aguilar et al. Reference Aguilar, Balcells and Cebolla-Boado2011; Bobowick et al. Reference Bobowik2017; Reichelmann et al. Reference Reichelmann, Michael Roos and Hughes2022).

Predicting Attitudes toward Indigenous Reconciliation

The comparative literature above suggests that support for reconciliation is likely influenced by the types of reconciliation policies under consideration (Reichelmann and Hunt Reference Reichelmann and Hunt2022; Torpey and Burkett Reference Torpey and Burkett2010), but also by a range of individual, socialization, and contextual factors that can outweigh considerations specific to the policies being proposed (Aguilar et al. Reference Aguilar, Balcells and Cebolla-Boado2011; Banting et al. Reference Banting, Harell and Kymlicka2022; Bobowick et al. Reference Bobowik2017; Reichelmann et al. Reference Reichelmann, Michael Roos and Hughes2022). It is also important to consider existing research about attitudes towards Indigenous communities in settler contexts where colonialism and dispossession were crucial to state formation. In one systematic review of 20 articles examining public opinion towards Indigenous peoples in Australia, the authors found that age had mixed effects, but men and individuals with lower levels of education were generally more negative towards Indigenous peoples (Falls and Anderson Reference Falls and Anderson2022). They also found that those who demonstrated a “social-dominance orientation and right-wing authoritarianism” were also more likely to hold negative attitudes towards Indigenous peoples (see also Heaven and Quintin Reference Heaven and St. Quintin2003), as did those who belonged to the Anglican Church (Falls and Anderson Reference Falls and Anderson2022). In the Canadian context, several studies have found that negative evaluations of Indigenous peoples correlate with an increased desire to be socially distanced from Indigenous peoples, increased opposition to Indigenous-focused government spending (Beauvais Reference Beauvais2021), increased opposition to welfare for Indigenous beneficiaries (Harell et al. Reference Harell, Soroka and Ladner2014, Reference Harell, Soroka and Iyengar2016), and increased support for the building of new pipeline infrastructure (Beauvais Reference Beauvais2022).

Taking all the existing research together, as well as general expectations related to policy attitudes, there seem to be three key categories of factors to consider. First, since reconciliation policies are explicitly designed to recognize wrongs committed by non-Indigenous persons against Indigenous peoples, general attitudes towards Indigenous peoples are likely to be relevant. Someone who has negative attitudes toward Indigenous peoples is unlikely to be supportive of policies designed for that group’s benefit.

Beyond attitudes about the groups identified by a policy, there are more general considerations known to shape policy opinions that might structure baseline inclinations toward new policies. Research on public opinion makes it clear that group identities are important when people are developing their own opinions about new or complex issues, as groups can provide individuals with cues about what a consistent stance would be (Lau and Redlawsk Reference Lau and Redlawsk2001). Since we are interested in reconciliation policies, which are both relatively newer additions to Canadian political debate as well as complicated in terms of understanding rights, responsibilities, and obligations, partisan and religious identities could be important for understanding policy stances. With respect to partisanship, those who identify with a political party often develop opinions in line with the stances of their preferred party. Especially for individuals who are ambivalent or who have not considered the issue of reconciliation policies, heuristics like party cues can be extremely useful (Bullock Reference Bullock2011). In Canada, conservative and ideologically right-wing parties have been generally less supportive of policies that recognize the distinctive rights and claims of Indigenous communities relative to the Liberal, New Democratic, and Green parties at the ideological center and left. Indeed, reflecting these trends, Indigenous voters have been found to disproportionately support the two main federal parties on the left, the Liberals and the NDP, relative to the main federal party on the right, the Conservatives (Harell and Panagos Reference Harell and Panagos2013; Dabin et al. Reference Dabin, Daoust and Papillon2019: 48). While partisan and ideological differences partly capture how parties “sort” the population along values, when confronted with specific policy proposals, we expect that Canadians may be particularly likely to take cues from elites from relevant social and political groups. We expect, therefore, those on the left of the political spectrum and those who support parties on the left will be more likely to support Indigenous reconciliation policies relative to those who support right-wing parties or are themselves positioned on the right.

In terms of religion, although residential schools were created by the federal government, they were run and administered by the Catholic church and other Christian denominations (TRC 2015; Miller Reference Miller2017; see also Aguilar et al. Reference Aguilar, Balcells and Cebolla-Boado2011: 1418–19). Blame for the atrocities that were committed has been leveled in part at Christian churches. For their part, the churches have typically responded with penitence but with limited financial support for reparations (Austen Reference Austen2022; Grant and Cardoso Reference Grant and Cardoso2021). Therefore, we expect Christian respondents, whose own religions are implicated, to be less likely to support reconciliation policies. Finally, given the very public verbal support for reconciliation provided by the government of Canada, including by then-Prime Minister Justin Trudeau after the final report of the Truth and Reconciliation Commission in 2015,Footnote ³ we think that it might be seen as the “Canadian” thing to do to support such policies. Therefore, those who value their identity as Canadians may take the government’s cue and support reconciliation policies.

Finally, related policy preferences are likely to shape one’s opinions on efforts toward reconciliation as well. Since reconciliation involves government actions for a minoritized group, it is likely that general attitudes towards government spending and activity, especially targeted policies for minority groups, will be relevant. It is not uncommon for individuals to develop opinions on new issues that are consistent with their existing policy preferences, especially if they are based on underlying values or orientations. For example, those who have a strong sense of justice or believe in support for minorities in general might likewise support Indigenous reconciliation efforts. Based mainly on Falls and Anderson (Reference Falls and Anderson2022), we therefore expect that attitudes about inequality in other domains may inform preferences over reconciliation policies. Similarly, we expect that support for government spending to support minorities is a relevant consideration as it captures a general willingness to proactively support minority-targeted policies and inclusion more generally.

In the next section below, we empirically explore the impact of a range of individual-level determinants that various literatures suggest should be important for influencing public opinion, with a specific focus on how group prejudice, partisan and religious identities, and other policy attitudes may shape public support for reconciliation in Canada.

Data and Methods

Our analysis relies on the 2021 Canadian Election Study (Stephenson et al. Reference Stephenson, Harell, Rubenson and Loewen2023), a survey of the general population conducted online with panelists from the Leger Opinion panel. The sample was drawn to be balanced on gender and age with region and language quotas. Canadian citizens and permanent residents aged 18 and over were eligible. The survey had two waves—one in the weeks leading up to the election, and one immediately after. 20,968 people answered the campaign period wave and 15,069 answered the post-election wave. The 2021 CES is ideally suited for this study because it contains the necessary sociodemographic, political, and attitudinal questions, including a battery of questions to gauge Indigenous resentment as well as another to gauge support for reconciliation, which we use to build Indigenous resentment and reconciliation scales. Developing single measures from multiple items enables us to capture underlying attitudes and, for Indigenous resentment at least, is consistent with the approach of classic (Henry and Sears Reference Henry and Sears2002) and recent scholarship on this and other related topics (Beauvais Reference Beauvais2022; Reference Beauvais2021).

Developing an Indigenous Reconciliation Scale

Before moving to evaluate the correlates of attitudes about reconciliation, we construct a scale using the four Likert-type questions from the 2021 CES. The questions capture attitudes on government actions and policies towards Indigenous peoples, particularly within the context of those impacted by residential schools. These questions span the topics of Canada’s Truth and Reconciliation Commission: accelerating progress on the calls to action, federal funding for unmarked graves, court actions directed at residential school survivors, and the renaming of buildings for those individuals associated with the residential schools.Footnote ⁴ At the same time, these questions do not capture the full range of possible reconciliation policies or outcomes. Questions relating to more expansive forms of self-government, land restitution, and constitutional change, for instance, might produce a wider range of public opinion. Nonetheless, the CES questions that are available do capture a reasonable range of reconciliation policies, from symbolic (e.g. accelerate progress; rename buildings) to more substantive (e.g. federal funding and ceasing court actions) that vary in terms of the kind of costs imposed on the Crown and the public more generally. The specific wording of the four questions is displayed in Table 1.Footnote ⁵ Respondents were presented with four response options—strongly oppose, somewhat oppose, somewhat support, and strongly support—to rate their level of agreement with the statements. For each of the four questions, opposition to the statement indicated less support for reconciliation and support for the statement meant greater support for reconciliation. Because support for reconciliation can be understood as a latent, continuous concept, a summated rating model provides a valuable framework for capturing the construct. The summated rating model allows us to transform multiple ordinal indicators into a single-level measure of support for reconciliation policies. To create the scale, we calculated the mean score across the four items (alpha=.73).Footnote ⁶ Figure 2 displays the distribution of reconciliation scale scores for 2021 CES respondents, showing that, on average, respondents were more supportive than unsupportive of reconciliation.Footnote ⁷

Table 1. Reconciliation scale items

Figure 2. Indigenous reconciliation scale.

As this scale has not been used in previous academic work (to the best of our knowledge), validating the scale is important, which means assessing the construct’s dimensionality, monotone homogeneity, reliability, and criterion validity. We conducted an exploratory factor analysis (EFA) and based upon the results of a scree plot analysis and loadings, the variables load on a single dimension (see Appendix Figure A1). All items are retained in the index as the factor loadings are all above the accepted .3 threshold (see Appendix Table A1).Footnote ⁸ The scale’s Cronbach’s alpha, which provides an estimate of the true reliability of the reconciliation scale, is .73.Footnote ⁹ We also looked at the Cronbach’s alpha for each item in relation to the rest scores, which exclude that specific item; the scores range from .61 to .78, indicating that each item is a reliable indicator of the concept (Appendix Table A2).Footnote ¹⁰

Finally, we examined criterion validity using an Indigenous resentment scale created from three indicators in the 2021 CES Study data (as shown in Figure 1).Footnote ¹¹ Theoretically, if our reconciliation measure truly captures attitudes in favour of policies that would lead to more reconciliation with Indigenous peoples, it should be negatively correlated with indicators of Indigenous resentment.Footnote ¹² The wording of the three Likert-scale questions designed to measure resentment are displayed in Table A3 in the Appendix.Footnote ¹³ Respondents were tasked with selecting one of the following: strongly disagree, somewhat disagree, neither agree nor disagree, somewhat agree, or strongly agree. For the “Favors” item, agreement with the statement meant greater Indigenous resentment. We reverse-coded “Colonialism” and “Deserve” so that higher levels of the item also indicated greater resentment. As with the reconciliation scale, the resentment scale was created using a summated rating model and we ran the same validation tests on it, which are included in the appendix (Figure A2, Tables A3-5); the tests suggest that the Indigenous resentment scale is valid and reliable. Our reconciliation scale has a significant negative relationship with the Indigenous resentment scale (correlation is -.62, p<.001), further supporting the validity of the Indigenous reconciliation scale.

Analyzing Determinants of Support for Reconciliation Policies

Having established the validity of the Indigenous reconciliation scale, we can use it as the dependent variable in regression models to identify the correlates of support for reconciliation policies. We focus on three sets of individual characteristics: group attitudes, partisan and religious identities, and existing policy preferences. We analyse these blocs separately and then in a final combined model. The first bloc, group attitudes, uses the Indigenous resentment scale described in the previous section. The second bloc includes various political and religious identities, in this case, partisanship, ideology, religion, and Canadian identity. To capture political orientation, we created a categorical variable indicating partisans of each major political party in Canada (Conservative, Liberal, NDP, Bloc Quebecois, Green/Other, as well as Non-partisan) and a continuous variable that captures the left-right political orientation of the respondents, measured on a 0-10 scale. For religion, we identified the groups implicated in the residential schools’ genocide (Catholics and some Christian religions), as well as other Christian religions.Footnote ¹⁴ The baseline is those who identify with non-Christian religions or who do not have a religious affiliation (Other). For Canadian identity, we used responses to a question about how important being a Canadian is to the person (1–4 scale). The third bloc is made up of three variables that capture support for equal rights (higher values mean less support), support for spending on immigrants (coded to indicate support for less spending), and a preference for doing less for racial minorities.

We also consider several control variables. We expect those with lower education to be less informed about the historical abuses that occurred at residential schools and therefore less likely to support policies that seek to address those experiences (Aguilar et al. Reference Aguilar, Balcells and Cebolla-Boado2011: 1419; Falls and Anderson Reference Falls and Anderson2022). Research also suggests women are more likely to be empathetic and altruistic than men and so we expect them to be more inclined to support reconciliation policies (Brañas-Garza et al. Reference Brañas-Garza, Capraro and Rascón-Ramírez2018; Löffler and Greitemeyer Reference Löffler and Greitemeyer2023; Mehravar et al. Reference Mehravar, Alcantara and Roy2023). New research by Albaugh et al. (Reference Albaugh, Harell, Loewen, Rubenson and Stephenson2024) finds that non-binary individuals tend to be more left-leaning than either men or women; in keeping with left-wing preferences for more government intervention and support for minorities, we expect them to be more supportive of reconciliation policies. Finally, we expect Indigenous respondents to be more likely to support Indigenous reconciliation initiatives, given previous studies that found that groups are more likely to support policies that target their group but not others (Campo et al. Reference Campo, Mastin and Somjen Frazer2004; Williams Reference Williams2007). In coding racial background, we also create a variable for racialized minorities, as we expect there may be cross-group support for minority-targeted policies like reconciliation (white is the baseline). The coding details of the variables included in the models are summarized in Table 2 of the Appendix.

To assess the influence of each bloc on support for reconciliation policies, we ran sequential OLS models to test the direct impact of the different factors. We run separate models for each set of predictors (Model 1 group attitudes, Model 2 political and religious identities, and Model 3 policy attitudes) while controlling for a small set of sociodemographic characteristics. We then include a final model with all variables (Model 4).Footnote ¹⁵ As a reference, we also include a parallel model where resentment is the outcome variable (see Appendix Table A7). In the appendix, we probe further by running models for each scale item individually (Appendix Table A8). Even though the scale is a valid measure of the underlying support for reconciliation policies that informed responses to the individual questions, the variation in the nature of the questions and their loadings led us to be curious as to whether a specific question, analyzed on its own, might have different correlates. This choice builds upon existing research that suggests attitudes towards reconciliation may vary with the type of policy being considered.