Hostname: page-component-cd9895bd7-p9bg8 Total loading time: 0 Render date: 2024-12-26T08:51:43.245Z Has data issue: false hasContentIssue false

Israel: a novel wedge issue in Canadian electoral politics

Published online by Cambridge University Press:  30 June 2023

Yannick Dufresne
Affiliation:
Département de science politique, Université Laval, Quebec City, QC, Canada
Jamie Levin
Affiliation:
Department of Political Science, St. Francis Xavier University, Antigonish, NS, Canada
Jonathan Paquin
Affiliation:
Département de science politique, Université Laval, Quebec City, QC, Canada
Marc-Antoine Rancourt*
Affiliation:
Department of Political Science, University of Toronto, Toronto, ON, Canada
*
Corresponding author: E-mail: marcantoine.rancourt@mail.utoronto.ca
Rights & Permissions [Opens in a new window]

Abstract

This article investigates the relationship between partisan foreign policy positions on Israel and the voting behavior of religious minorities in Canada. It discusses Stephen Harper's strong pro-Israeli stance in foreign policy when the Conservatives were in power and focuses on two main explanations accounting for such politicization of Israel, namely moral obligations and political clientelism. These hypotheses are tested using the 1968–2015 Canadian Election Study (CES) surveys and the 2011–2015 Vox Pop Labs election data. The results suggest that the Israeli issue had an impact on the support for the Conservatives among voters from religious minorities. Considering the effect of this foreign policy positions, Jewish Canadians are shown to be more supportive of the Conservatives, while the opposite pattern is observed among Muslim Canadians. The implications of these findings are then discussed.

Type
Article
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
Copyright © The Author(s), 2023. Published by Cambridge University Press on behalf of the Religion and Politics Section of the American Political Science Association

The politicization of religion has been an important feature of Canadian politics since the beginning of the confederation (Godbout and Høyland, Reference Godbout and Høyland2013). The division between Catholics and Protestants is foundational (Gidengil, Reference Gidengil1992; Rayside et al., Reference Rayside, Sabin and Thomas2017) and was shortly absorbed into the Federal party system (Johnston, Reference Johnston2017). Although remnants of these historical divisions still faintly persist, they do not account for the contemporary religious cleavages in Canadian politics (Johnston, Reference Johnston1985). Indeed, in the last decades, the question of Israel, and now, especially, Islam, became among the most contentious issues in modern Canadian politics. In a historically surprising twist of faith—akin to the one observed in the United States—the political left is said to have moved away from their Jewish voters and move toward Muslim voters, while the political right did the opposite.

In the wake of the 2011 Canadian federal election, a flurry of articles appeared in the Canadian press suggesting that Jewish voters had suddenly migrated rightward in unprecedented numbers.Footnote 1 Under Stephen Harper's leadership, the Conservatives made extensive inroads amongst various ethnic voting groups, breaking apart an electoral base historically acquired to the Liberals (Blais et al., Reference Blais, Gidengil, Nadeau and Nevitte2002). Not only had the Canadian Jewish laity shifted their allegiance, the media also reported that numerous prominent Jewish Canadians—including Robert Lantos, Heather Reisman, and Gerald Schwartz—had publicly quit the Liberals, casting their support and fundraising weight behind Stephen Harper's Conservatives (Clark, Reference Clark2006). Most journalistic accounts attributed this rightward shift to the pro-Israel positions staked out by the Harper government.Footnote 2

If true, this new religious cleavage is remarkable for two reasons. First, Jews and Muslims represent small shares of the Canadian population—1.1 and 3.2% of the population, respectively. Second, the reported pattern of politicization is orthogonal to the dominant axes of religious division. The political right is said to have aligned against Muslims, who are reported to be among the most devout and morally conservative religious groups in the country (Cochrane, Reference Cochrane2011, Reference Cochrane2015), which is to say that Muslims disproportionately agree with moral policies that the political right has spent the past five decades promoting, and the left opposing. The political left, meanwhile, is said to have moved away from Jewish voters, who are one of the most liberal and secular religious groups in the country (Cochrane, Reference Cochrane2015), which is to say that a substantial number of Jewish voters generally agree with the moral positions that the left has spent half a century supporting, and the right opposing.Footnote 3 As such, there does not appear to be any good reason for Canadian parties to have politicized these minority religions in the first place, and even less reason for them to have politicized these religious issues in the topsy-turvy way that they did.

There are two observable implications of the hypothesis that a new religious cleavage has emerged in Canada. The first implication is that the voting behavior of the Jewish and Muslim communities is increasingly aligned with parties' positions on Israel. The second implication is that, if not for the positions of voters on these issues, voting behavior would be different. This article tests two straightforward hypotheses deduced from these implications: (1) Stephen Harper's consistent and vocal pro-Israeli policy increased the support of Jewish-Canadian voters for the Conservatives Party, and (2) this political stance had the opposite effect among Muslim-Canadian voters.

The only comprehensive study on Jewish-Canadians' voting patterns was conducted more than 30 years ago by Jean Laponce. The research found that Jewish Canadians leaned leftwards, most of them casting their votes for the Liberals (Laponce, Reference Laponce1988).Footnote 4 Our article therefore offers a significant contribution to this topic by focusing on recent transformations of Jewish and Muslim Canadians' voting patterns. Our research shows that despite important and systematic heterogeneity within the Jewish and Muslim communities, a significant number of Jewish-Canadian voters did shift their vote for supporting the Conservatives, while the opposite behavior can be observed among Muslim-Canadian voters. As we show in this article, public opinion evidence from multiple large election surveys over the past decade is consistent with the emergence of a new cross-cutting religious dimension in the Canadian political landscape. We test prominent and plausible explanations for the emerging cleavage. We give particular attention to the impact of the pro-Israeli stance of the Harper government (2006–2015) on Jewish-Canadian and Muslim-Canadian voting behavior.

As such, we first present an analysis on the politicization of the Israeli issue in Canada and discuss the Conservative government pro-Israeli policy, which was part of a broader foreign policy transformation under Stephen Harper. We also introduce two plausible explanations—moral obligation and political clientelism—that might help account for this shift. In relation to these explanations, we briefly address the notion of political segmentation. We then look at the effect of proposed government policies on Israel—the Israeli issue—on the evolution of Jewish-Canadian and Muslim-Canadian vote using large-N data. The analysis shows a substantial increase in Jewish-Canadian support for the Conservatives and a decrease in Muslim-Canadian support for the party over the last decade. The results suggest that this increase might be explained by the Conservatives' repositioning on the Israeli issue. We conclude by discussing implications for theories of voting behavior, issue publics' conception of public opinion, party competition, and Canadian politics more generally.

1. The Conservatives and the politicization of Israel

During its 10 years in power, the Harper government demonstrated remarkable support for the state of Israel (Morin and Roussel, Reference Morin and Roussel2014; Paris, Reference Paris2014; Paquin, Reference Paquin2018). Although Canada's policy regarding the settlement of the Israel-Palestine conflict did not officially change under the Conservative government, the intensity and consistency of Harper's support for Israel were unprecedented and represented a diplomatic shift in the handling of this issue. Indeed, many observers described Harper's government as the most pro-Israeli government in Canadian history.Footnote 5 A few examples are in order: after the Conservatives' election in 2006, the government in fact stated that Canada was Israel's “best friend” and suspended support for the Palestinian Authority following Hamas—a group considered a threat to Israel and designated as a terrorist organization in Canada (Chapnick, Reference Chapnick2012, 151)—election to the Palestinian Legislative Council.Footnote 6 Later that year, Harper publicly sided with Israel during its war against Hezbollah in Lebanon, one of the few Western leaders to do so.

In 2009, the Harper government boycotted a UN international conference on racism to protest Iranian President Ahmadinejad's presence, given his well-known aggressive stance toward Israel. That same year, following Hugo Chavez's decision to expel the Israeli ambassador to Venezuela as a protest against Israel's military offensive in the Gaza Strip, Canada accepted Israel's request to represent its diplomatic interests in Caracas. The Harper government also shifted Canada's voting patterns at the UN and other affiliated agencies to consistently oppose, or at least “abstain” from, supposedly anti-Israel resolutions where it had previously taken a more balanced approach. Most notably, in 2012, the Conservative government opposed the bid by the Palestinian Liberation Organization (PLO) for statehood in the UN, making Canada one of only nine countries to do so. Prime Minister Harper personally urged other world leaders to vote against the bid.

Finally, the Canadian government was highly skeptical of the P5 + 1 talk with Iran on its nuclear military program that eventually led to the Joint Comprehensive Plan of Action in 2015. While three core Canadian allies (the United States, the United Kingdom, and France) worked around the clock to find an agreement with Tehran, the Conservatives argued that Canada could not trust a regime that sponsors terrorism and that threatens the state of Israel (MacDonald, Reference MacDonald2014). So radical were these departures from the mainstream Canadian diplomatic tradition that it has been widely argued that Canada's loss of its 2010 bid for a seat on the UN Security Council was a direct result (Ibbitson and Slater, Reference Ibbitson and Slater2010; Levitz, Reference Levitz2013).

1.1 Liberals, New Democrats, and Israel

We tend to forget that it is under Paul Martin's Liberals (2004–2005) that the Canadian government initiated a recalibration of Canada's positions on Israel. Liberal MPs including several Cabinet ministers known as the “Liberals for Israel” pushed for the modification of Canada's traditional positions with respect to the Israel/Palestine conflict. At the UN, the Martin government abstained or opposed resolutions that did not promote a constructive dialogue; essentially resolutions condemning Israel's behavior. The Martin government abstained on a resolution supported by 150 states condemning Israel's West Bank wall even if the International Court of Justice concluded that the wall violated international law. Canada also voted against a UN resolution urging Israel to respect international law and condemning its occupation of the West Bank and the Gaza Strip. However, since Paul Martin was in power for only two years and the Harper government adopted such strong pro-Israel positions for nearly 10 years, we tend to forget that the recalibration toward Israel was initiated by the Liberals. Moreover, the Martin government was criticized by the Jewish community for dispatching its Foreign Minister, Pierre Pettigrew, to attend PLO Chairman Yasser Arafat's funeral in 2004. Pettigrew was also criticized for laying a wreath on Arafat's grave while on an official visit to Ramallah a year later (Merkley, Reference Merkley2012). This somehow gave the impression that the Liberals had a mixed record on Israel. Then, as the Harper's Conservatives unequivocally ranked behind Israel from 2006 on, the Liberals appeared more critical of Israel. This culminated with Michael Ignatieff's statement according to which the Israeli government behaved as a war criminal during the 2006 Lebanon war (Gordon, Reference Gordon2009).

Canada's traditional third party, the New Democrats, underwent no particular change during this time, long having been more sympathetic to the Palestinian cause. The New Democrats, for example, has recognized Hezbollah as a legitimate political interlocutor since 2006, despite that organizations being placed on the terror watch list by the Chrétien government in 2002, and former senior New Democrat leaders, such as Alexa McDonough and Peggy Nash, have called for the inclusion of Hamas in peace negotiations, a move that the Liberals and Conservatives both reject. The New Democrats have also suggested amending the Canada-Israel Free trade agreement to exclude Israeli exports from the occupied territories. And unlike the Conservatives and Liberals, the New Democrats supported the Palestinians 2012 bid for statehood in the UN. While the New Democrats attempted to reposition itself under the leadership of Thomas Mulcair, it has not managed major changes to its platform, though New Democrat MPs voted unanimously in 2016 against a call to condemn the Boycott, Divest, and Sanction movement (BDS) against Israel, a policy that was actively endorsed only a few weeks earlier by the federal Greens. And, under the current leadership, no changes have been proposed.

2. Explaining the Conservative pro-Israel stance

The reasons behind the position that the Conservatives staked out vis-à-vis Israel are open for debate, though two main explanations emerge in the literature (Flanagan, Reference Flanagan2011; Nossal, Reference Nossal, Smith and Sjolander2013). The first is the moral obligation argument, which holds that Canada has a duty to support Israel because it shares Canadian values such as freedom, democracy, and human rights in a region marked by authoritarianism, hatred, and intolerance. At the 2011 Conservatives' convention, Prime Minister Harper declared,

In the world, we must also have a purpose and that purpose is no longer just to go along and get along with everyone else's agenda. It is no longer to please every dictator with a vote at the United Nations… Moral ambiguity, moral equivalence are not options; they are dangerous illusions… we know where our interests lie and who our friends are (Robertson, Reference Robertson2011).

This led the Harper government to distance itself from the liberal internationalism stance—emphasizing UN multilateralism, international rules and norms, and peacekeeping missions and mediation (Nossal, Reference Nossal, Smith and Sjolander2013)—that was the consensual foundation of Canada's foreign policy since the end of the World War II. The Conservatives were highly critical of this stance as well as of previous Liberal governments who were the standard bearers of that approach. They argued that liberal internationalism compromised Canada's core principles and values. As Paris (Reference Paris2014) points out, “The Conservative narrative about foreign policy portrayed liberal internationalism not only as a failure, but also as morally flawed.” These transformations led some observers to argue that the Conservative government had initiated a “big shift” (Bricker and Ibbitson, Reference Bricker and Ibbitson2013; Ibbitson, Reference Ibbitson2014) in Canadian foreign policy (see also Nossal, Reference Nossal, Smith and Sjolander2013; Paris, Reference Paris2014).

A similar line of argumentation suggests that Israel stands on a civilizational fault line, a bulwark of Judeo-Christian values in a largely Muslim region. On this view, Canada should align itself with Israel because both countries share the same civilizational values—Judeo-Christian beliefs in a region defined by Islam. This argument is said to be in line with Stephen Harper's strong personal beliefs (Sasley and Jacoby, Reference Sasley, Jacoby, Heinbecker and Momani2007; Martin, Reference Martin2011). Such an ideological stance stresses that Canada has to do what is right as opposed to what is popular (Chapnick, Reference Chapnick, Chapnick and Kukucha2016, 106). As a result, Canada must stand up for Israel rather than adopting a position of moral relativism or acting as a so-called honest broker in the Israel-Palestine conflict (Baird, Reference Baird2012).

The second explanation for the Conservatives newfound position on Israel is electoral clientelism; that the Harper government tried to instrumentalize Canada's foreign policy to mobilize support from specific segments of the Canadian population to both strengthen and diversify their base of support (Flanagan, Reference Flanagan2011; Nossal, Reference Nossal, Smith and Sjolander2013; Paris, Reference Paris2014). This explanation is in line with Bow and Black's (Reference Bow and Black2008, 17) comment that “each party wants to try to use foreign policy issues as political wedges, to try to turn voters away from the other party and attach them to itself.” By emphasizing the promotion of uncompromising moral principles and values, particularly civilizational or religious values, the Conservatives brought their foreign policy in line with the values of their core voters (Lagassé et al., Reference Lagassé, Massie, Roussel, Castro-Rea and Boily2014), particularly conservative-Christians (Merkley, Reference Merkley2012). And by improving their branding on minority issues, the Conservatives hoped to broaden their base of support.

3. Targeting small electorate segments

Ethnic communities are said to be profitable electoral segments for parties to target, particularly when community members are often concentrated in specific geographic areas where they share a community space and media sphere. Conventional wisdom has historically considered immigrants and minorities to be Liberal partisans (Bilodeau and Kanji, Reference Bilodeau, Kanji, Anderson and Stephenson2010). Indeed, immigrant voters have contributed to the past electoral success of the Liberals (Blais, Reference Blais2005). However, immigrants and racial minorities are of particular strategic interest to the Conservatives since many of these voters hold conservative positions on issues such as the economy, law and order, and moral questions (Flanagan, Reference Flanagan2011; Bricker and Ibbitson, Reference Bricker and Ibbitson2013). After failed attempts to rally francophone Quebecers in the 2006 and 2008 elections (Flanagan, Reference Flanagan2011), the Conservatives began courting immigrant voters on the eve of the 2011 election (Flanagan, Reference Flanagan2011; Friesen and Sher, Reference Friesen and Sher2011), leading them to pull ahead of the Liberals at the polls while remaining close to their ideological core (Paré and Berger, Reference Paré and Berger2008). As Seligman (Reference Seligman2016, 311) argues,

The Conservative Party's success in 2011, particularly in ridings home to many Jewish voters (i.e. Thornhill), emboldened the party to more aggressively appeal to Jewish and pro-Israel voter in Canada. In this regard, if the changes introduced in 2006 represented divergent foreign policy priorities then those introduced after 2011 represented an effort to further drive a political wedge between the Conservatives and the Liberals.

It is reasonable to assume that voters who identify with Jewish or Muslim communities are likely to care more about the government's position on the Israeli question than are members of other ethnic or religious groups. Members of these two religious communities are more likely to form what is called an “issue public” based on that particular issue (Krosnick, Reference Krosnick1990; Soroka, Reference Soroka2002). This does not mean that all members of those two ethnic groups care about the issue nor that all non-members do not care. Following the literature on political targeting strategies (Alvarez et al., Reference Alvarez, Hopkins and Sinclair2010), we argue that the Conservative government made a concerted effort to attract additional Canadian-Jewish voters by adopting a pro-Israel policy. However, the consequence of such one-sided positions would likely lead to the alienation of Muslim voters (Flanagan, Reference Flanagan2011). What explains the calculus? Martin (Reference Martin2011, 82) remarks, “Though Muslims outnumbered Jews by two to one in Canada, the Jewish community was more politically impactful.”

4. Methodology and data

Our two research hypotheses are tested using two datasets: the Canadian Election Study (CES) surveys and data from the 2011 and 2015 Vox Pop Labs (VPL) election data.Footnote 7 The data from 13 waves of the CES, from 1968 to 2015, allow the analysis to extend over a wide time span and reveal changes in trends. Despite there being some variance in the wording of questions, the different CES waves were conducted using similar methods of investigation and include comparable sets of questions. These questions provide the sociodemographic and attitudinal information required to conduct the subsequent analysis. The questions of particular importance—those regarding ethnicity and year of immigration—were asked in all CES waves since 1974. Finally, the longevity of the CES gives it an advantage over other data sources since it is amenable to the pooling of data samples and permits cross-time comparisons. The data are weighted using classic sociodemographic variables. Thus, the CES data used in this paper are representative of both Canadian population and the different population subgroups (for more information on the 2011–2015 CES, see https://ces-eec.arts.ubc.ca/english-section/home). For recent information and the 2019 dataset, see http://www.ces-eec.ca/).

In addition to the CES data, the VPL data include 30 positional questionsFootnote 8 as well as users' sociodemographic backgrounds and political preferences. Nearly two million usable responses were collected during the 2011 and the 2015 Canadian federal election campaigns making VPL useful for studying small electoral subgroups.Footnote 9 The VPL data, from the Vote Compass of 2011 and 2015, are collected from an online application that surveys users' views on a variety of public policy issues germane to a given election campaign, and then offers users an estimation of their position in the political landscape and, by extension, their alignment with each of the political parties contesting said campaign. The application is wildly popular, drawing millions of users worldwide, and is generally run in partnership with major media organizations in the jurisdiction where an election is being held.

5. Results

The small number of Jewish and Muslim voters present in each CES sample requires a pooling of many CES samples taken over time.Footnote 10 Despite the lack of precise information for each election, some trends are apparent. First, Jewish and Muslim voters have contrasting political behavior. Second, the Conservatives' long-standing difficulty amongst minority voters is apparent. Being a Muslim voter is a very strong predictor of not voting for the Conservatives. This result is stronger and more statistically significant during the period for which we have the most recent data.Footnote 11 Jewish voters similarly exhibited a consistent and relatively negative relationship with the Conservatives, oscillating between the Liberals and the New Democrats. For all periods before 2008–2015, results from multinomial regressions show that being a member of the Canadian-Jewish community is associated with greater support for either the Liberals or the New Democrats than for the Conservatives (the baseline category). However, CES data show an increased propensity for Jewish-Canadians to vote for the Conservatives during the last decade (see Table 1), though these results do not rise to a level of statistical significance. For that reason, we turn to the larger VPL data to create a voting model including a much larger sample of voters.

Table 1. The Jewish- and Muslim-Canadians vote (1968–2015)

Source: Canadian Election Studies (1968–2015). Vox Pop Labs election data (2011, 2015).

Methods: Multinomial logistic regression (base category: Conservatives). The coefficients represent effects on log odds.

Note: The models also include a set of sociodemographic controls (not shown).

†Significant at p < 0.10; ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001.

One main advantage of VPL data is its ability to allow for the examination of subgroups. Specifically, the 2011 and 2015 VPL data include 30,940 respondents identifying as Jewish or Muslim. In comparison, there are only 754 of these respondents in all 13 CES waves conducted since 1968.Footnote 12 It should thus come as no surprise that VPL data can be used to detect statistically significant effects of smaller groups better than can surveys the size of the CES. Nevertheless, some scholars have already shown that the validity of multivariate results estimated with VPL data is comparable to the results obtained from more traditional surveys (see Fournier et al., Reference Fournier, Gélineau and Harell2014).

Indeed, the VPL model is nearly identical to the findings based on CES data described above; the results of which are shown in Figure 1.Footnote 13 To ease comparison with the CES models, the same results are also presented in Table 1.

Figure 1. Jewish- and Muslim-Canadians vote (2011–2015).

Methods: Multinomial logistic regression (base category: Conservatives). The coefficients on the y-axis represent effects on log odds.

Note: Error bars indicate 95% confidence intervals.

Source: Vox Pop Labs election data (2011, 2015).

The results of the VPL model support the idea of recent polarization between Jewish-Muslim voters.Footnote 14 All other factors held constant, Muslim voters clearly oppose the Conservatives, while Jewish voters clearly support them. The inverse relationship is found for support for the Liberals. The effect of being Muslim is particularly substantive; it is the strongest sociodemographic predictor for voting for the Liberals and against the Conservatives both in 2011 and 2015. It is impossible to determine precisely from these results that the polarization of Jewish and Muslim voters is a result of changing Conservative policy toward Israel. While some scholars believe that this to be the case, the data presented above only provide indirect evidence. For more direct evidence, we need to study a question tapping more precisely into the Israeli issue. However, this issue question is rarely, if ever, asked in election surveys. Also, a very large sample is required to allow investigation at such a micro-subgroup level. Fortunately, the 2015 Canadian VPL data include just such a question—How supportive should Canada be of Israel?—allowing us to study the relationship between party policy on Israel and voter preference. The distributions of attitudes on that question and their impact on support for the Conservatives for Jewish, Muslim, and other Canadians are presented in Figure 2 and Table A1.

Figure 2. Effect of the Israeli issue on Jewish- and Muslim-Canadians 2015 vote intentions for Harper's Conservatives.

Methods: Ordinary linear regression.

Independent variable: “How supportive should Canada be of Israel?”

Dependent variable: “Regardless of the party you intend to vote for in this election, in general how likely are you to support the Conservative Party of Canada?” (On a scale from 0 to 10).

Note: For more details, see Table A1 in Appendix.

Source: Vox Pop Labs election data (2011, 2015).

The attitudes toward the Israeli question show that Canadians are generally not supportive of increasing Canada's support for Israel. Not surprisingly, perhaps, Jewish Canadian voters most support such an option. However, these results do not explicitly tell us the degree of support for the current position on Israel. That being said, most Canadians appear divided on the question whether Canada should reduce its support to Israel or keep it as it is. This last result is particularly significant considering the pro-Israeli position of the Canadian government at the time. The same division does not exist among Muslim Canadians. Indeed, nearly 85% of Muslim Canadians are in favor of reducing Canada's support for Israel. But do such attitudes have any impact on the vote?

The findings shown in Figure 2 and Table A1 are consistent with the interpretation that the Israeli issue matters. Given the Conservatives' vocal and consistent position on the Israeli issue, the relationship appears to be in the right direction: the greater one's support for a pro-Israeli stance, the more likely one is to vote Conservative. One should note, however, that despite the fact that the model shows a strong effect among Muslim voters, very few members of that community hold that opinion in the population. Figure 2 and Table A1 show the great discrepancy.

Figure 2 shows just how different Jewish-Canadians, Muslim-Canadians, and other religious groups are on the Israel issue in 2015. The fact that the two religious groups show a very clear stance on the Israel issue, compared to the rest of the population, is striking. It also shows how that greatly impacts the vote for the Conservatives that year. This is confirmed by the results in Table A1. The interaction between both religious groups and the issue of Israel is statistically significant while controlling for sociodemographic variables, left/right ideology, and 29 other issues. This comports with the core of the issue-public hypothesis. The significant interaction between being a member of the Jewish or the Muslim communities and one's position on the Israeli issue also supports the idea of salience-based heterogeneity in the effect of positional issues: the same attitude on an issue can have a different effect on vote choice for different segment of voters. In this situation, the personal salience of the issue operates as the key to make that issue matters for voters' choice.

6. Discussion: a new politics of religion

These findings suggest that Harper's more vocal, unequivocal, and consistent stance on the issue of Israel might have cost his party some voters. The sharp polarization between Jewish- and Muslim-Canadians on the Israeli issue appears to be reflected in their vote choice. Considering that there are more than twice as many Muslims than Jews in Canada, the Conservatives' strategy of the last few years might be electorally costly in the middle or long term.

More generally, these results can also have substantial consequences for studies of political behavior. In fact, in most conventional voting models, the effect of being a member of a particular community can be obscured by the fact that respondents having religions other than Protestantism and Catholicism are often grouped into the same “other religions” category. A single variable indicating all of these voters would never capture the alleged widening polarization between voters of Jewish and Muslim faiths. By definition, if there was a polarization, the positions of Jews and Muslims would simply cancel out because the effects are in opposite directions.Footnote 15

There are therefore clear advantages to considering different ethnic or religious groups independently of one another. However, it is important to avoid essentialist conceptualizations of social groups. These groups are vehicles of socialization and their members should not be considered as having fixed characteristics. Ultimately, these conceptual challenges add to the problems of sample size that are often encountered when dealing with subgroups. Conclusions about the efficacy of the Conservative targeting strategy must therefore rest on indirect evidence, that is, on the polarizing effect that the Conservatives' pro-Israeli policy is alleged to have had on Jewish- and Muslim-Canadians voters. Moreover, some further analyses would contribute to a better understanding of the phenomenon at hand. With the right data, time-series analyses would allow for a clearer understanding of whether “Stephen Harper's consistent and vocal pro-Israeli policy” may explain the increase in support for the Conservative Party among the Jewish community and the decline in support for such a party among the Muslim community.

We can nevertheless suggest that the Israeli issue has become a “wedge issue” in that it splits Conservatives and Liberals on dividing lines. We do not mean here that Jewish-Canadians and Muslim-Canadians are split among themselves. While it may be the case, the focus of this paper is the discrepancy between the two groups and the polarization that followed the Harper government decisions on Israel. Furthermore, the dynamic use of the issue of Israel by the Harper government cemented its status in the Jewish-Canadian community. Jewish-Canadian voters who might have previously been Liberals or New Democrats partisans massively turned toward the Conservatives because of this particular issue. As such, it rendered this singular issue a key role in any understanding of subgroups political and voting behavior in Canada.

Since assuming the leadership of the Liberal Party, Justin Trudeau has made a deep and concerted effort to win back Jewish voters. In speaking to the Jewish community, Trudeau has played up his visits to Israel and has stridently rejected Conservative claims that the Liberal Party is a lesser friend to Israel. In the process he has won back several prominent Jewish fundraisers who had previously defected to the Conservatives, such as Stephen Bronfman (Taylor-Vaisey, Reference Taylor-Vaisey2013).

Jewish-Canadian voters have little electoral impact, given the community's small size, and, if anything, Muslim-Canadian voters have even less impact, despite the community's larger size. If the rationales for politicizing Israel are eventually uncovered by researchers, the votes of Muslim and Jewish religious minorities will turn out, we suspect, to have had nothing to do with it.

Yannick Dufresne is a professor and chair of the Chaire de leadership en enseignement des sciences sociales numériques at the Department of Political Science at Université Laval. He is a member of the RCMPP, the chair of Democracy and Parliamentary Institutions, the Center for Studies on Democratic Citizenship and the Center for Public Policy Analysis. He holds a Ph.D. in political science from the University of Toronto (2015). His fields of expertise are the study of public opinion, electoral studies, political psychology, and political communication. His research focuses mainly on democratic processes and political strategies in the era of big data. His publications include the British Journal of Political Science, Parliamentary Affairs, and the Canadian Journal of Political Science.

Jamie Levin is an assistant professor in the Department of Political Science at St. Francis Xavier University in Nova Scotia, Canada. He earned his Ph.D. in political science at the University of Toronto. His work focuses on the role of weapons in the resolution of civil wars. Jamie's research demonstrates how weapons can be a stabilizing force in the resolution of conflict. His work has appeared in both the Academic Press and Popular Media, including International Peacekeeping, Journal of Peace Conflict and Development, International Politics, International Studies Review, the Palestine Israel Journal, Harper's Magazine, the Washington Post's Monkey Cage, the Walrus, Haaretz, the Jerusalem Post, the Toronto Star, and the National Post.

Jonathan Paquin is a professor of political science at Université Laval, Canada. His research focuses on U.S. and Canadian foreign policy, foreign policy analysis, and foreign interventions.

Marc-Antoine Rancourt is a graduate student at the Department of Political Science at the University of Toronto. His fields of expertise are the study of public opinion, media studies, and political methodology. He is an affiliate member of the Chaire de leadership en enseignement des sciences sociales numériques.

Appendix

Table A1. Effect of the Israeli issue on Jewish- and Muslim-Canadians 2015 vote intentions for Harper's Conservatives

Table A2. Issues and question wording of the 29 issues included in the “positional issues” variable

Footnotes

1. See Caplan (Reference Caplan2009), Gordon (Reference Gordon2009), Tracy (Reference Tracy2009), Ibbitson and Slater (Reference Ibbitson and Slater2010), Schnoor (Reference Schnoor2011), Simpson (Reference Simpson2011), Illouz (Reference Illouz2013), Iveson (Reference Iveson2013), Levitz (Reference Levitz2013), Medved (Reference Medved2013), Taylor-Vaisey (Reference Taylor-Vaisey2013). While there is a lively debate about Jewish identity (c.f., Boyarin and Boyarin, Reference Boyarin and Boyarin1993) we sidestep these issues here. “Jewish” is a self-identified category in the data we use herein.

3. For a discussion of these attitudes in Canada, see notably Laponce (Reference Laponce1988) and Cochrane (Reference Cochrane2011).

4. Laponce (Reference Laponce1988) explains the propensity of Jewish voters to vote for parties on the left variously to “values” (Fuchs, Reference Fuchs1956), “status discrepancy” (Lipset and Raab, Reference Lipset and Raab1984), “anti-Semitism,” “concern for Israel,” and others.

5. See, Gordon (Reference Gordon2009), Ibbitson and Slater (Reference Ibbitson and Slater2010), Illouz (Reference Illouz2013), Levitz (Reference Levitz2013), Taylor-Vaisey (Reference Taylor-Vaisey2013).

6. Ezra Levant explains that the Jewish community treated the precursors to the Conservative Party—the Canadian Alliance and Reform Parties—with “suspicion” because of their links to the Social Credit Party, which had been inflected with anti-Semitism (Preston Manning's father was the head of Social Credit). Similarly, Adelman notes that the Christian Conservatism of these parties was equally likely to frighten away Jewish voters (Adelman, Reference Adelman2013).

7. The VPL data were collected through the Vote Compass tool, a popular voter engagement application promoted in collaboration with CBC/Radio-Canada. See https://voxpoplabs.com/ and http://www.votecompass.com for details.

8. The “Positional Issues” in the final regression model (see Table A1) include 29 out of the 30 positional questions as controls for the other social and economic issues that were more or less salient at the time. See Table A2 for a list of these issues, their wording and answer choices.

9. Despite its obvious advantages, VPL data must be treated with care because their participants opt-in. In order to deal with potential selection bias, analyses using these data are corrected based on users' sociodemographic backgrounds such as age, education, income, gender, language, size of town, and country of origin.

10. When models are estimated for every CES wave separately, the results are similar, despite the large standard errors of the coefficients for the variables representing Jews and Muslims. Linear regression models, which can be preferred in some circumstances for cross-time comparisons, also provide similar results.

11. Reflecting Canadian demography at the time, very few Muslim voters were included in older CES surveys.

12. There are 1782 Jewish-Canadians in 2011 in the VC dataset and 11,859 in 2015. There are 205 between 1968 and 1988 in the CES, 115 between 1993 and 2000, 48 between 2004 and 2006, and 119 between 2008 and 2015. There are 972 Muslim-Canadians in 2011 in the VC dataset and 16,327 in 2015. There are 11 between 1968 and 1988 in the CES, 52 between 1993 and 2000, 55 between 2004 and 2006, and 230 between 2008 and 2015. This number of respondents is one of the reasons why the Vox Pop Labs election data are used. Similar analyses on the small samples from the CES would not allow the detection of significance effects. Moreover, a bigger sample size allows for a smaller margin of error.

13. Figure 1 uses the data from Vox Pop Labs in 2011 and 2015 because we are mostly interested in how the behavior of Jewish-Canadians and Muslim-Canadians changed at that time regarding the Israel issue and the Harper government decisions on Israel. While we are also interested in the period from 1968 to 2008, it is not the focus of this paper. Moreover, the low number of Jewish-Canadians and Muslim-Canadians respondents makes it so we cannot reliably conduct the analyses presented in Figure 1 before 2011. The data from Vox Pop Labs only go back to 2011.

14. We cannot make a causal inference claim with observational data. We have thus to rely on statistical controls. See Table A1 for more information on statistical controls.

15. Of course, canceling-out effects also depend on the strengths of the relationships and the sizes of each of the social groups.

Source: Vox Pop Labs election data (2011, 2015).

Methods: Ordinary linear regression.

Note: The “positional issues” included in the final model combines 29 Likert-scale type of questions related to various issues salient during the 2015 Canadian federal election. Contact the authors for more details. See Table A2 for the wording of the questions related to each issue. No coefficient is on purpose. The issues are used as controls only.

†Significant at p < 0.10; ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001.

References

Adelman, H (2013) Canadian Jewish voting and Israel. Unpublished manuscript.Google Scholar
Alvarez, RM, Hopkins, A and Sinclair, B (2010) Mobilizing Pasadena democrats: Measuring the effects of partisan campaign contacts. The Journal of Politics 72(1), 3144.CrossRefGoogle Scholar
Baird, J (2012) Address by Minister Baird to the American Jewish Committee. Government of Canada. URL https://www.canada.ca/en/news/archive/2012/05/address-minister-baird-merican-jewish-committee.htmlGoogle Scholar
Bilodeau, A and Kanji, M (2010) The new immigrant voter, 1965–2004: The emergence of a new Liberal partisan?. In Anderson, CD and Stephenson, LB (eds), Voting Behaviour in Canada. Vancouver: UBC Press, 6585.Google Scholar
Blais, A (2005) Accounting for the electoral success of the Liberal party in Canada presidential address to the Canadian political science association London, Ontario June 3, 2005. Canadian Journal of Political Science/Revue canadienne de science politique 38(4), 821840.CrossRefGoogle Scholar
Blais, A, Gidengil, G, Nadeau, R and Nevitte, N (2002) Anatomy of a Liberal Victory: Making Sense of the Vote in the 2000 Canadian Election. Peterborough: Broadview Press.Google Scholar
Bow, B and Black, D (2008) Does politics stop at the water's edge in Canada? Party and partisanship in Canadian foreign policy. International Journal 64(1), 727.Google Scholar
Boyarin, D and Boyarin, J (1993) Diaspora: Generation and the ground of Jewish identity. Critical Inquiry 19(4), 693725.CrossRefGoogle Scholar
Bricker, D and Ibbitson, J (2013) The Big Shift: The Seismic Change in Canadian Politics, Business, and Culture and What it Means for our Future. Toronto: Harper Collins.Google Scholar
Caplan, G (2009) Is the Harper government playing the anti-Semitic card? The Globe and Mail. URL http://www.theglobeandmail.com/news/politics/is-the-harper-government-playing-the-anti-semitic-card/article4296503/Google Scholar
Chapnick, A (2012) A diplomatic counter-revolution: Conservative Foreign Policy, 2006–11. International Journal 67(1), 137154.Google Scholar
Chapnick, A (2016) Stephen Harper's Israel policy. In Chapnick, A and Kukucha, CJ (eds), The Harper Era in Canadian Foreign Policy, Parliament, Politics, and Canada's Global Posture. Vancouver: UBC Press, 324.Google Scholar
Clark, C (2006) Liberal power couple back Harper on Mideast. The Globe and Mail. URL http://www.theglobeandmail.com/news/national/liberal-power-couple-back-harper-on-mideast/article4109511/Google Scholar
Cochrane, C (2011) The structure and dynamics of Muslim-Canadian opinions about same-sex marriage, abortion, and welfare. Association Canadienne de Science Politique/Canadian Association of Political Science.Google Scholar
Cochrane, C (2015) Left and Right: The Small World of Political Ideas. Montreal & Kingston: McGill-Queen's Press-MQUP.Google Scholar
Flanagan, T (2011) The emerging Conservative coalition. Policy Options. URL https://policyoptions.irpp.org/magazines/the-winner/the-emerging-conservative-coalition/Google Scholar
Fournier, P, Gélineau, F and Harell, A (2014) Are voting advice applications useful for the study of elections? Working Paper.Google Scholar
Friesen, J and Sher, J (2011) How courting the immigrant vote paid off for the Tories. The Globe and Mail. URL https://www.theglobeandmail.com/news/politics/how-courting-the-immigrant-vote-paid-off-for-the-tories/article578608/Google Scholar
Fuchs, LH (1956) The Political Behavior of American Jews. Glencoe: Free Press.Google Scholar
Gidengil, E (1992) Canada votes: A quarter century of Canadian national election studies. Canadian Journal of Political Science/Revue canadienne de science politique 25(2), 219248.CrossRefGoogle Scholar
Godbout, JF and Høyland, B (2013) The emergence of parties in the Canadian House of Commons (1867–1908). Canadian Journal of Political Science/Revue Canadienne de Science Politique 46(4), 773797.CrossRefGoogle Scholar
Gordon, S (2009) Conservatives wooing traditionally Liberal Canadian Jews. The Forward. URL http://forward.com/news/121602/conservatives-wooing-traditionally-liberal-canadia/Google Scholar
Ibbitson, J (2014) The big break: The Conservative transformation of Canada's foreign policy. Centre for International Governance Innovation Papers, no. 29.Google Scholar
Ibbitson, J and Slater, J (2010) Security Council rejection a deep embarrassment for Harper. Globe and Mail. URL https://www.theglobeandmail.com/news/politics/security-council-rejection-a-deep-embarrassment-for-harper/article1370239/Google Scholar
Illouz, D (2013) A fresh perspective: Harper and Israel. The Jerusalem Post. URL http://www.jpost.com/Opinion/Columnists/A-Fresh-Perspective-Harper-and-Israel-335538Google Scholar
Iveson, J (2013) Harper's Israel trip spurred on by the one constituency that still treats him like a rock star. The National Post. URL http://news.nationalpost.com/full-comment/john-ivison-harpers-israel-trip-spurred-on-by-the-one-constituency-that-still-treats-him-like-a-rock-starGoogle Scholar
Johnston, R (1985) The reproduction of the religious cleavage in Canadian elections. Canadian Journal of Political Science/Revue Canadienne de Science Politique 18(1), 99113.CrossRefGoogle Scholar
Johnston, R (2017) The Canadian Party System: An Analytic History. Vancouver: UBC Press.CrossRefGoogle Scholar
Krosnick, JA (1990) Government policy and citizen passion: A study of issue publics in contemporary America. Political behavior 12(1), 5992.CrossRefGoogle Scholar
Lagassé, P, Massie, J and Roussel, S (2014) Le néoconservatisme en politique étrangère et de défense canadiennes. In Castro-Rea, J and Boily, F (eds), Le fédéralisme selon Harper: la place du Québec dans le Canada conservateur. Québec: Les Presses de l'Université Laval, 4981.Google Scholar
Laponce, JA (1988) Left or centre? The Canadian Jewish electorate, 1953–1983. Canadian Journal of Political Science/Revue canadienne de science politique 21(4), 691714.CrossRefGoogle Scholar
Levitz, S (2013) John Baird: Won't sell “Canada's Principled Foreign Policy” for another shot at UN Security Council seat. The National Post. URL http://news.nationalpost.com/news/canada/canadian-politics/wont-sell-canadas-principled-foreign-policy-for-another-shot-at-un-security-council-seat-bairdGoogle Scholar
Lipset, SM and Raab, E (1984) The American Jews, the 1984 elections, and beyond. The Tocqueville Review 6(2), 401419.CrossRefGoogle Scholar
MacDonald, LI (2014) Q&A: A conversation with John Baird: “we promote Canadian values”. Policy: Canadian Politics and Public Policy 1(2), 49.Google Scholar
Martin, L (2011) Harperland: The Politics of Control. Toronto: Penguin Press.Google Scholar
Medved, M (2013) Jews, Conservatives, and Canada: North of the border, why have they gone to the right? Commentary 136(4), 3032.Google Scholar
Merkley, PC (2012) At issue: Reversing the poles: how the pro-Israeli policy of Canada's Conservative government may be moving Jewish voters from left to right. Jewish Political Studies Review 23(1–2), 3960.Google Scholar
Morin, D and Roussel, S (2014) Autopsie de la politique étrangère de Stephen Harper: un examen préliminaire. Canadian Foreign Policy Journal 20(1), 18.CrossRefGoogle Scholar
Nossal, KR (2013) The Liberal past in the Conservative present: Internationalism in the Harper era. In Smith, HA and Sjolander, CT (eds), Canada in the World: Internationalism in Canadian Foreign Policy. Oxford: Oxford University Press, 2135.Google Scholar
Paquin, J (2018) US partisan perceptions of Stephen Harper's shift in foreign policy. International Journal 73(2), 282298.Google Scholar
Paré, DJ and Berger, F (2008) Political marketing Canadian style? The Conservative Party and the 2006 federal election. Canadian Journal of Communication 33(1), 3963.CrossRefGoogle Scholar
Paris, R (2014) Are Canadians still Liberal internationalists? Foreign policy and public opinion in the Harper Era. International Journal 69(3), 274307.Google Scholar
Rayside, D, Sabin, J and Thomas, PE (2017) Religion and Canadian Party Politics. Vancouver: UBC Press.CrossRefGoogle Scholar
Sasley, BE and Jacoby, TA (2007) Canada's Jewish and Arab communities and Canadian foreign policy. In Heinbecker, P and Momani, B (eds), Canada and the Middle East: In Theory and Practice. Waterloo: Wilfrid Laurier University Press, 185204.Google Scholar
Schnoor, R (2011) The contours of Canadian Jewish life. Contemporary Jewry 31(3), 179197.CrossRefGoogle Scholar
Seligman, S (2016) Canada and the United Nations general assembly (1994–2015): Continuity and change under the Liberals and the Conservatives. Canadian Foreign Policy Journal 22(4), 276315.CrossRefGoogle Scholar
Simpson, J (2011) How the political shift among Jewish voter plays in Canada. The Globe and Mail. URL http://www.theglobeandmail.com/opinion/how-the-political-shift-among-jewish-voters-plays-in-canada/article4199522/Google Scholar
Soroka, SN (2002) Issue attributes and agenda-setting by media, the public, and policymakers in Canada. International Journal of Public Opinion Research 14(3), 264285.CrossRefGoogle Scholar
Taylor-Vaisey, N (2013) Israel's best friend: Stephen Harper. Maclean's Magazine. URL http://www.macleans.ca/news/canada/israels-best-friend-stephen-harper/Google Scholar
Tracy, M (2009) Canada PM Harper continues drive for Jewish vote. Tablet. URL http://www.tabletmag.com/scroll/22886/canada-pm-harper-continues-his-drive-for-the-jewish-voteGoogle Scholar
Figure 0

Table 1. The Jewish- and Muslim-Canadians vote (1968–2015)

Figure 1

Figure 1. Jewish- and Muslim-Canadians vote (2011–2015).Methods: Multinomial logistic regression (base category: Conservatives). The coefficients on the y-axis represent effects on log odds.Note: Error bars indicate 95% confidence intervals.Source: Vox Pop Labs election data (2011, 2015).

Figure 2

Figure 2. Effect of the Israeli issue on Jewish- and Muslim-Canadians 2015 vote intentions for Harper's Conservatives.Methods: Ordinary linear regression.Independent variable: “How supportive should Canada be of Israel?”Dependent variable: “Regardless of the party you intend to vote for in this election, in general how likely are you to support the Conservative Party of Canada?” (On a scale from 0 to 10).Note: For more details, see Table A1 in Appendix.Source: Vox Pop Labs election data (2011, 2015).

Figure 3

Table A1. Effect of the Israeli issue on Jewish- and Muslim-Canadians 2015 vote intentions for Harper's Conservatives

Figure 4

Table A2. Issues and question wording of the 29 issues included in the “positional issues” variable