Antisemitism and polarization: the political dynamics of American Jewish concerns about traditional and Israel-related antisemitism

ABSTRACT

Notwithstanding the increasing alignment between social and political identities, political divisions exist within social groups as well as between them. Despite their shared traditions, American Jews face similar political divisions as Americans in general. But could the perception of a common threat, such as antisemitism, alter the dynamics of political polarization within a social group? American Jews’ concerns about antisemitism might themselves be filtered through political identities, with liberal Jews more concerned about “traditional” antisemitism (long-standing anti-Jewish stereotypes) emanating from the political right, and conservative Jews concerned about “Israel-related” antisemitism (blaming individual Jews for the actions of Israel) associated with the political left. Using a sample of over 2000 Jewish young adults, we find that, although liberal and conservative Jews are equally concerned about traditional antisemitism, conservatives are significantly more concerned about Israel-related antisemitism than liberals. We also find that, after a series of high-profile attacks on American Jews related to the 2021 Israel–Hamas conflict, liberal and conservative Jews’ concerns about Israel-related antisemitism increased at similar rates. These results suggest that, although concerns about antisemitism are influenced by political identity, rising antisemitism is unlikely to exacerbate political polarization between liberal and conservative Jews.

KEYWORDS:

• Antisemitism
• polarization
• political identity
• American Jews
• ideology
• Israel

Introduction

Political polarization in the United States, especially affective polarization (Iyengar et al. 2019) is often linked to growing divisions between identity groups. In the United States, the left and the right have each become more homogenous in terms of race, geography, religion, and education, with partisans increasingly viewing social groups linked to their political opponents in negative terms (Achen and Bartels 2016; Robison and Moskowitz 2019). At the same time, political divisions also occur within social identity groups, resulting in a complex negotiation of similarity and difference. To what extent do shared traditions, narratives, and hardships reduce ideological and partisan animosity?

This question is especially relevant for American Jews. Although they are connected by traditions, culture, and religion, liberal and conservative¹ American Jews, like liberal and conservative Americans in general, tend to see the world in dramatically different ways. But regardless of their political identities, Jews face a growing threat associated with a resurgence of antisemitism in the United States. Although the common threat of antisemitism may help bridge political divisions in the American Jewish community, Jews’ concerns about antisemitism might themselves be filtered through ideology (Wright et al. 2021; Becker 2020). Liberal Jews may express greater anxiety about “traditional” antisemitism based on long-standing anti-Jewish bigotry, stereotypes, and tropes that are frequently associated with the political right. In contrast, conservative Jews may mostly focus on “Israel-related” antisemitism, which involves the use of antisemitic tropes to express criticism of Israel or the practice of blaming Jews (individually or as a group) for the actions of the Israeli government, and is often associated with the political left. If political attitudes shape concerns about antisemitism, then liberal and conservative Jews might not see antisemitism as a common threat at all, but rather as an ideologically coded conflict that is primarily driven by the “other” side of the political divide. Because Jewish young adults are especially likely to be the targets of antisemitic harassment (Cohen 2018a; Smith and Schapiro 2018) and are still developing emotionally and politically (Dinas 2013), they provide an especially useful window into this dynamic.

This paper explores the relationship between American Jewish young adults’ symbolic ideological identity (Ellis and Stimson 2012) and their concerns about traditional and Israel-related antisemitism. Understanding this relationship will shed light on the nature of affective polarization in the US Jewish electorate. Using a sample of over 2000 Jewish young adults, we find that while liberal and conservative Jews are equally concerned about traditional antisemitism, conservative Jews are significantly more concerned about Israel-related antisemitism. We also find that in the wake of a series of high-profile attacks on American Jews related to the 2021 conflict between Israel and Hamas, liberal and conservative Jews’ concerns about Israel-related antisemitism increased at similar rates. This suggests that the ideological disparity in concerns about Israel-related antisemitism is less likely a result of motivated reasoning in response to cognitive dissonance and is more likely driven by ideological disagreements about criticism of Israel (between those who view criticism of Israel as a reflection of their love for the country and those who believe criticism of Israel is unacceptable). These results suggest that, despite their differences, liberal and conservative American Jews see rising antisemitism as a common threat that can transcend partisan politics.

Affective polarization and American Jews

Affective polarization is one of the most serious challenges facing United States politics in the twenty-first century (Iyengar et al. 2019). American politics have become increasingly “tribal” – with those on the left and right ever more likely to fear and loath those who hold divergent political identities, regardless of how much they actually disagree about particular political issues (Mason 2018a; Mason and Wronski 2018; Iyengar, Sood, and Lelkes 2012; Kingzette et al. 2021). Affective polarization is driven in part by the coupling of political identities with other social identities, such as race (Mason 2016; Westwood and Peterson 2022). However, affective polarization can also cause divisions within identity groups.

Recent research indicates that the rise in political polarization that has impacted Americans, in general, has also divided American Jews (Pew Research Center 2021). Currently, a majority of American Jews identify as politically liberal, although a substantial minority, especially those who identify religiously as Orthodox, tend to identify as politically conservative (Tighe et al. 2021; Pew Research Center 2021). Many of the key flashpoints in American politics, such as the actions of former President Trump, appear to have divided American Jews along similar ideological lines as Americans in general (Tobin 2020; Vulliamy 2018). In 2020, 88% of Jewish Republicans approved of President Trump’s performance as President; among Jewish Democrats, only 3% approved (Pew Research Center 2021, 162).

Rising antisemitism may, however, alter the dynamics of affective polarization among American Jews. In addition to recent high-profile examples of antisemitic violence (Van Sant and Doubek 2019; Turkewitz and Roose 2018; Wu, Desiderio, and Zanona 2021), there appears to be a broader increase in antisemitic hostility and anti-Jewish attitudes in the general public (Lipstadt 2019; Smith and Schapiro 2018; AMCHA Initiative 2018). American Jews are aware of the growing threat posed by antisemitism in the United States and have become increasingly concerned about it (Wright et al. 2021; Cohen 2018b; Pew Research Center 2021).

The threat of antisemitism seems particularly acute for Jewish young adults. Nearly two-thirds of American Jews ages 18–29 attend college (Pew Research Center 2021), where acrimonious conflicts about antisemitism, especially as it relates to criticism of Israel, have taken place in the past several years (Pollack 2010; Bauer-Wolf 2018; Saxe et al. 2016). Although other forms of prejudice are often more evident among older adults (Stewart, Hippel, and Radvansky 2009), recent research finds that expressions of antisemitism are more common among younger adults, possibly due to the lower salience of the Holocaust, or elevated hostility toward Israel among younger Americans (Cohen 2018a; Smith and Schapiro 2018; Royden and Hersh 2022; Hersh and Royden 2022). This dynamic suggests that Jewish young adults may be especially likely to encounter antisemitism from non-Jewish peers. Furthermore, for these young Jews, exposure to rising antisemitism comes at a critical time in their emotional and political development. “Emerging adulthood” (Arnett 2014) represents an important transition period, where political opinions, beliefs, and identities are challenged, revised, and eventually crystalized (Sears and Valentino 1997). Indeed, the political attitudes of young adults seem especially likely to be altered in response to political events (Dinas 2013).

So how will politically divided American Jews, and Jewish young adults, in particular, respond to rising antisemitism? Will they grow closer together or become even more divided? We might expect that the common threat of antisemitism represents an important “cross-cutting” concern (Lipset and Rokkan 1967; Mason 2016, 2018b) that will reduce affective polarization in the American Jewish community. American Jews on opposite sides of the American political divide may be less likely to see one another as “the enemy” if they feel similarly targeted by a common adversary. Insofar as Jews perceive a need for solidarity in the face of antisemitic threats, Jews on the left and right might find themselves working together toward a common cause, which in turn may help bridge social divisions (Pettigrew 1997; Pettigrew and Tropp 2006)

At the same time, if American Jews’ concerns about antisemitism are themselves filtered through ideology, antisemitism might not reduce polarization among American Jews and may even exacerbate it. It has long been known that political social identity, including partisan identity (Green, Palmquist, and Schickler 2002) and “symbolic” identification as liberal or conservative (Ellis and Stimson 2012), can influence perceptions of social reality (Bartels 2002; Jerit and Barabas 2012; Kahan et al. 2017). The power of the psychological mechanisms, such as motivated reasoning (Kunda 1990; Kahan 2013), that enable us to manage cognitive dissonance and threats to social identity, have been well documented (Festinger 1965; Branscombe et al. 1999; Nauroth et al. 2017).

One way in which motivated reasoning can ease cognitive dissonance is by managing attributions of threat and blame: It is not dissonant to believe that a hated out-group poses a serious threat to our way of life, but admitting that a group that we identify with poses a similar threat would cause far more psychological discomfort (Festinger 1965; Cooper 2007). Thus, in the United States, Democrats and Republicans agree that democracy is in danger of collapse but disagree about the source of this danger – with Democrats casting the blame on “MAGA Republicans” and Republicans pointing the finger at “socialist Democrats” (Kanno-Youngs and Shear 2022). It seems plausible, then, that a similar dynamic may divide liberal and conservative Jews, in terms of how they think about the threat of antisemitism.

Existing research has already shown that when the question of who is “responsible” for antisemitism is explicitly framed as part of the broader conflict between the left and right, American Jews answer this question using existing political cues. Becker (2020) finds that, when asked whether the left or right bears greater responsibility for antisemitism in the United States, liberal Jews tend to blame the right, while conservative Jews blame the left.² But this leaves open the more important question of whether political identity influences American Jews’ concerns about antisemitism in general – even when they are not being prompted to think about antisemitism in terms of existing partisan or ideological conflicts. Although existing literature suggests that political identity may influence Jewish concerns about antisemitism, it provides little guidance as to the nature or even direction of this relationship. Cohen (2018b) finds that perceptions of antisemitism are significantly higher among political conservatives, while Wright et al. (2021) find that concern about antisemitism is significantly higher among political liberals.

Two forms of antisemitism

One possible explanation for these ambiguous results, and a reason to question whether Jewish concerns about antisemitism constitute a “cross-cutting” issue, is that existing scholarly and popular discourse suggests that American Jews distinguish between two forms of antisemitism, each associated with a different side of the political spectrum. Like many other forms of prejudice toward historically marginalized groups, “traditional” or “classic” antisemitism based on long-standing anti-Jewish stereotypes (e.g., allegations of “blood libel,” neo-Nazi racial theories, or conspiracy theories about world domination) has historically been associated with right-wing ideas and groups (Dunbar and Simonova 2003; Dunbar 1995; Adorno et al. 1950). Contemporary discourse frames “traditional” antisemitism as a problem primarily among those on the political right, including the QAnon movement and extremist supporters of former President Trump (Moshin 2018; Weisman 2018).

At the same time, there have been long-standing debates over the existence and nature of a “new antisemitism” (Rosenfeld 2015) that appears to be intimately connected to criticism of the modern state of Israel (Judaken 2008; Klug 2013; Cohen et al. 2009; Shenhav-Goldberg and Kopstein 2020). Setting aside the question of whether it really is “new,” what we will refer to as “Israel-related antisemitism,” has been defined (International Holocaust Remembrance Alliance 2016) as including criticism of Israel that invokes or reflects antisemitic stereotypes or tropes (e.g., the use of Nazi images or references to the Holocaust to characterize the actions of the Israeli military) as well as expressions of animosity toward individual or groups of Jews driven by hostility to Israel (e.g., blaming individual Jews for the actions of the Israeli government).³ In contrast to traditional antisemitism, Israel-related antisemitism is more often framed (accurately or not) as a particular problem of the political left (Lipstadt 2019; Barclay 2020). For example, accusations of antisemitism related to criticism of Israel have been leveled against the Boycotts, Divestments and Sanctions movement (Marcus 2015; Sheskin and Felson 2016), the Occupy movement (Arnold 2015), the British Labour Party (Barclay 2020), and members of the Democratic Congressional Caucus (Nelson 2019).

The distinction between traditional and Israel-related antisemitism can blur, depending on the context. For example, one “traditional” stereotype about American Jews is their purported “dual loyalty” to Israel and the United States (Golinkin 2019). Similarly, discussions of “hegemonic Zionism” associated with the left-wing criticism of Israel are sometimes seen as having antisemitic overtones precisely because they play into traditional antisemitic conspiracy theories about Jewish control of the media and an international Jewish cabal bent on world domination (Schraub 2019). In fact, no prior research has attempted to empirically determine the extent to which traditional and Israel-related (or “new”) antisemitism in fact represent distinct concerns in the minds of American Jews.

If American Jews do see traditional and Israel-related antisemitism as distinct threats, then we might expect liberal and conservative Jews to be more concerned about the form of antisemitism associated with the “other side” of the political divide and less concerned about the form associated with their own political “team,” in the same way that Americans express greater concerns about the threats to democracy posed by their political opponents. For example, in order to mitigate cognitive dissonance, American Jews who identify as liberal might employ motivated reasoning to discount the threat of Israel-related antisemitism when left-aligned political groups and figures make disparaging, Israel-related statements. For similar reasons, liberal Jews might also be more likely to express concerns about “traditional” antisemitism, which reinforce their existing beliefs about bigotry on the political right. The same dynamics might lead conservative Jews to express less concern about charges of traditional antisemitism, often leveled against their political allies on the right, and to express greater alarm about Israel-related antisemitism, associated with their political opponents on the left.

But aside from these psychological dynamics, there are more substantive reasons why liberal and conservative Jews might express different levels of concern about traditional and Israel-related antisemitism. Political scientists have long argued that political conservatives are more likely than liberals to express “uncritical patriotism” toward their home country (Huddy and Khatib 2007),⁴ meaning that they are uncomfortable with criticism of their country for any reason. In contrast, liberals are more likely to express “constructive patriotism,” in which criticism, motivated by a desire for positive change, is viewed as an essential component of what it means to care about their country. Recent research (Wright et al. 2022) has found that the same dynamics characterize young American Jews’ feelings about criticism with regard to both the United States and Israel. Young adult conservative Jews are more likely to feel that criticism of America or Israel is inherently illegitimate, while young adult liberal Jews are more likely to agree that they “only criticize [Israel/America] because they love it.” Since they have less tolerance for criticism of Israel in general, conservative Jews may be more likely than liberal Jews to feel that particularly harsh criticism of Israel “crosses a line” into antisemitism. For example, a liberal Jew might view the claim that Israel is an “apartheid state” as legitimate criticism of the Israeli occupation of Palestinian territory (even if they disagree with the claim itself), while conservative Jews might see the same claim as reflecting an antisemitic “double standard” toward Israel, given the dearth of similar claims about other countries that restrict the rights of particular ethnic groups.

At the same time, traditional antisemitism is often conceptualized as a form of racism (Dunbar 1995; Dunbar and Simonova 2003), and liberal Americans are more likely to view racism as a pervasive problem in American society. Since the 1960s, liberalism as a political identity has been increasingly associated with the recognition of discrimination against traditionally marginalized groups, while Americans who identify as conservative have been more likely to downplay the continued prevalence of racism and discrimination in America (Enders 2019; Ellis and Stimson 2012). Seeing traditional antisemitism as a serious problem may amount to an admission that “old fashioned” racism and discrimination based on racial or ethnic categories still play a major role in American life, a belief more aligned with ideological liberalism.⁵

If political identity is related to how American Jews think about antisemitism, then the common threat of antisemitism may not reduce political divisions between liberal and conservative Jews, even though it endangers them both. Liberal Jews will see the threat of antisemitism as mainly due to right-wing racism, while conservative Jews will see antisemitism as primarily driven by left-wing critics of Israel. In both scenarios, antisemitism will be viewed as a problem of “the other side,” and an extension of existing political conflicts, rather than as a common concern for all Jews, regardless of their ideological identity.⁶

It is, however, possible that ideological identity influences concerns about one form of antisemitism but not the other. For example, liberal and conservative Jews may be equally concerned about traditional antisemitism but deeply divided in their level of concern about Israel-related antisemitism (or vice versa). In this example, concerns about traditional antisemitism would be a cross-cutting issue that might reduce polarization, even as concerns about Israel-related antisemitism continue to further divide American Jews politically. It is important to note that in this situation, liberal and conservative Jews would still differ in their degree of concern about traditional antisemitism relative to Israel-related antisemitism (i.e., compared to conservative Jews, liberal Jews would be more likely to say that traditional antisemitism is a “bigger problem” than Israel-related antisemitism). However, this disparity in relative concern about traditional antisemitism versus Israel-related antisemitism would be entirely due to the relationship between ideology and concerns about Israel-related antisemitism. To prevent this conflation of effects, it is important to examine the effect of ideology on the absolute level of each form of concern separately, rather than analyzing relative concern about one form versus the other.

Furthermore, if a relationship between ideology and either form of concern is found, it is important to understand the stability of this relationship; in particular, the extent to which the magnitude of the relationship changes in response to an exogenous increase in that form of antisemitic threat. One example of such an exogenous shock came in May 2021, when there was a spate of attacks against American Jews during a period of conflict between Israel and Hamas. Media accounts explicitly noted that the perpetrators of the incidents often framed their own actions as a response to Israel’s actions during the conflict (Breslow 2021; Shapiro 2021). Because these attacks involved individual Jews being blamed for the actions of the Israeli government, they clearly represented an example of Israel-related antisemitism.

We would expect that Jewish concerns about Israel-related antisemitism would be higher in the aftermath of these high-profile attacks. Yet, if as hypothesized, liberal and conservative Jews differ in their baseline levels of concern about Israel-related antisemitism, might they also differ in their reaction to an increase in the threat posed by Israel-related antisemitism? The answer likely depends on the mechanism driving the ideological divide. Insofar as disagreements about the threat of Israel-related antisemitism are primarily driven by politically motivated reasoning to address cognitive dissonance, we would expect that any increase in concern about Israel-related antisemitism after the 2021 attacks would be concentrated among conservative Jews – since an increase in this type of concern would produce less cognitive dissonance for conservatives than for liberals. However, if differences in views about the threat of Israel-related antisemitism are primarily due to disagreements about the extent to which criticism of Israel crosses the line into antisemitism, then we would expect the 2021 attacks on individual Jews – a clear-cut case of Israel-related antisemitism that cannot be construed as “constructive” criticism of Israel’s policies – to produce an increase in concern about Israel-related antisemitism among liberal Jews that is at least as large, if not larger, than that observed among conservative Jews.

Research questions

This paper assesses the relationship between American Jewish young adults’ political identity and their concerns about traditional and Israel-related antisemitism using survey data collected in winter and spring 2021 from two independent samples of Jewish young adults. We first examine the correlation between these two concerns to understand the extent to which they represent distinct constructs in the minds of American Jewish young adults. We then analyze the relationship between each form of concern about antisemitism and symbolic ideological identity. We hypothesize that, after controlling for background characteristics, Jewish young adults who identify as politically liberal will be more likely to be concerned about traditional antisemitism, compared to those who identify as politically conservative. We also hypothesize that conservative Jewish young adults will be more likely to be concerned about Israel-related antisemitism, compared to liberal Jews. The relationship between symbolic ideological identity and concerns about these two forms of antisemitism may be explained by politically motivated reasoning, or by substantive political disagreements about criticism of Israel or the prevalence of ethnic and religious bigotry in American society. We attempt to adjudicate between these competing causal explanations by analyzing how young Jews’ concerns about Israel-related antisemitism changed in response to the 2021 Israel–Hamas conflict and the related attacks on American Jews. If an ideological disparity in concerns about Israel-related antisemitism is primarily driven by motivated reasoning employed to address cognitive dissonance, we would expect to see a larger increase in concerns about Israel-related antisemitism among conservatives than among liberals after Israel-related antisemitic incidents.

Data and methods

Data for these analyses come from two online surveys administered to separate samples of US applicants to the Birthright Israel program, which has offered tens of thousands of Jewish young adults free, peer educational trips to Israel (Saxe and Chazan 2008). The first survey was conducted in February 2021 and targeted eligible⁷ applicants who applied to summer 2020 trips that were canceled due to the COVID-19 pandemic. The survey was administered to a random sample of 7091 applicants. Individuals were randomly offered either a guaranteed incentive in the value of $10 (a choice between an Amazon gift card and a donation to one of two charities) or an opportunity to enter a lottery to win a $100 prize (a choice between an Amazon gift card and a donation to one of two charities). The achieved sample for this survey was 1034 (AAPOR RR2 = 15%). The second survey was conducted in June 2021 and was sent to all 12,435 eligible applicants to summer 2021 trips, before the trips departed. All respondents were offered an opportunity to enter a lottery to win one of ten $100 prizes (a choice between an Amazon gift card and a donation to one of two charities). The achieved sample for this survey was 1622 (AAPOR RR2 = 14%). None of the respondents to either survey had participated in the Birthright Israel program at the time of the survey.

One limitation of this study is that our sample, being limited to young Jews who applied to Birthright Israel in 2020 and 2021, does not represent a random sample of Jewish young adults in general. The threat to external validity posed by the use of this nonprobability sample is mitigated somewhat by the enormous scale of the program: As of 2019, approximately one quarter of all American Jews ages 25–34 reported having been on a Birthright trip (Pew Research Center 2021). Consequently, although we expect certain characteristics, such as positive attitudes towards Israel, to be associated with the decision to apply, the program still attracts applicants from a wide variety of backgrounds and attitudes, including those who are highly critical of Israel.⁸ Analyses have shown that during the program’s dramatic expansion between 2007 and 2018 differences between Birthright applicants and other young adult Jews with regard to background characteristics (such as exposure to formal Jewish education or having had a bar/bat mitzvah ceremony) have declined dramatically, although they have not disappeared entirely (Wright, Hecht, and Saxe 2019, 39).

In addition, methodologists have begun to take a more nuanced view of the utility of nonprobability sampling, recognizing that such samples may be perfectly appropriate for some research questions but deeply problematic for others (Simmons and Bobo 2015; Zack, Kennedy, and Long 2019). For questions that involve the relationship between variables (as opposed to the estimation of population values), the most critical questions are not whether the sample characteristics exactly match those of the population, but whether selection into the sample is correlated with potentially confounding variables, and whether critical sub-groups are simply missing entirely (Mercer et al. 2017).

Consequently, although remaining differences between the Birthright applicant population and the broader Jewish population are likely to bias overall estimates of the prevalence of concerns about traditional or Israel-related antisemitism, they are less likely to bias estimates of the relationship between these concerns and symbolic ideology, especially after the addition of statistical controls for differences in Jewish background. That is, given the size and diversity of the Birthright applicant population, it seems unlikely that the relationship between symbolic ideology and concerns about antisemitism would operate dramatically differently among those individuals who applied to the program and those who did not. Nevertheless, future work should attempt to extend these analyses to other samples of Jewish young adults, as well as to older American Jews.

A second limitation of this analysis is the lack of robust longitudinal data for investigating whether concerns about antisemitism actually changed in the wake of the 2021 Gaza conflict. The February and June surveys were conducted on largely independent samples of Birthright applicants: Only a small number of individuals (N = 272) applied to Birthright in both time points and responded to both surveys.⁹ Although the number of applicants who completed both surveys is too small to allow for a reliable, longitudinal test of our research question about differential change in concerns about antisemitism by symbolic ideology, this small sub-sample enables us to determine whether there was an overall increase in concerns about each form of antisemitism among Birthright applicants between these two time points. To help address the possibility that, in the full sample used to investigate our research question, observed changes in concerns about antisemitism between the two time points are due to changes in the applicant or sample composition (and to adjust for nonresponse bias more generally), we separately weighted the achieved samples of both surveys to demographic and Jewish characteristic targets from the same population: Summer 2021 applicants. Weights adjusted on gender, number of Jewish parents, age, and identifying with an Orthodox Jewish denomination, and weighting targets were drawn from Birthright registration data available for all population members. These weights are applied to all analyses discussed below.

Measures

To measure concerns about antisemitism, respondents in both surveys were asked two separate questions: “In the US, how concerned are you about antisemitism related to criticism of Israel?” and “In the US, how concerned are you about antisemitism related to traditional anti-Jewish stereotypes?” Both questions had ordinal response options ranging from “not at all” to “very much.” As shown in Table 1 “very much” was the modal response to both questions, with concern about traditional antisemitism being somewhat higher.

Table 1. Frequency of two forms of concerns about antisemitism.

	Israel-related	Traditional
Not at all	8%	3%
A little	20%	14%
Somewhat	28%	30%
Very much	43%	52%
Total	100%	100%

To establish that these two forms of concern represent distinct constructs, we first examine the correlation between them. The full bivariate relationship between the variables is shown in Table 2, which presents cell percentages and marginal distributions for each form of concern. Thirty-four percent of respondents reported being “very much” concerned about both forms of antisemitism. In contrast, hardly any respondents reported being “not at all” concerned about either form.

Table 2. Cell percentages and marginal distributions for concerns about Israel-related and traditional antisemitism.

		Israel-related
		Not at all	A little	Somewhat	Very much	Total
Traditional	Not at all	2%	1%	0%	0%	3%
	A little	2%	8%	3%	2%	14%
	Somewhat	2%	7%	14%	7%	30%
	Very much	2%	6%	10%	34%	52%
	Total	8%	20%	28%	43%	100%

Note: Pearson’s r = 0.5.

Table 3 summarizes this relationship more clearly by dividing the entire sample into four mutually exclusive categories: those who are “very much” concerned about both forms of antisemitism, those who are “very much” concerned about one form but not the other, and those who are not “very much” concerned about either. Just over one quarter of respondents (28%) reported that they were “very much” concerned about one form of antisemitism but not the other, indicating that, although highly correlated, concerns about these two forms of antisemitism do represent distinct constructs.

Table 3. Concern about each form of antisemitism (mutually exclusive summary).

	%
“Very much” concerned about both forms	34
“Very much” concerned about traditional antisemitism only	19
“Very much” concerned about Israel-related antisemitism only	9
Not “very much” concerned about either form	38
Total	100

Self-identified political ideology was measured by a question with a standard seven-point scale, ranging from “extremely liberal” to “extremely conservative.” Responses were grouped into three categories of liberal, moderate, and conservative (see Table 4). We measure the effect of the summer 2021 Israel–Hamas conflict and subsequent attacks on American Jews by a variable indicating the time period in which the survey was conducted: “Before the conflict (February 2021)” and “After the conflict (June 2021).”

Table 4. Descriptive statistics for independent variables.

		%
Political identity	Liberal	59
	Moderate	21
	Conservative	20
Highest level of formal Jewish education	None	37
	Supplementary school	38
	Day school	26
	Participated in informal Jewish education growing up	46
	Visited Israel	27
Date of survey	February 2021	44
	June 2021	56
	Belong to Orthodox denomination	19
Gender	Female	53
	Mean	Std. Dev.	Min	Max
Activism index	0.07	1.03	−1.07	3.61
Age	23.23	4.31	18	34

Models

Our initial hypotheses are tested using a pair of ordered logit models. The dependent variable in the first model is the original four-category measure for concern about traditional antisemitism and the dependent variable in the second model is the equivalent measure for concern about Israel-related antisemitism. Because coefficients from ordered logit models have no intuitive interpretation, we also present predicted probabilities, showing the estimated likelihood of being “very much” concerned about each form of antisemitism by political ideology, holding all other factors constant.¹⁰ To test our second set of hypotheses (that ideology moderates the effect of the Israel–Hamas conflict on levels of concern), we add an interaction term between ideology and time period to each model and again present predicted probabilities to illustrate the magnitude of the effect, if any, of ideology and the time of survey.

The models include a variety of control variables to address potential confounding relationships and biases associated with the sample design. Due to the eligibility requirements of the program, all respondents are between the ages of 18 and 34, but we still include a continuous control for age because applying to Birthright at different ages (e.g., applying immediately after one becomes eligible, versus waiting until one has almost aged out of eligibility) might reflect different levels of interest in travel to Israel. Because prior research finds that women rank the Israeli-Palestinian conflict as a less important policy issue (Royden and Hersh 2022), and that, among American Jews, women are more concerned about antisemitism in the United States (Wright et al. 2021), we also include a control for gender. To account for bias in the composition of the sample, and because concerns about antisemitism are related to the strength of respondents’ Jewish identity and experiences (Cohen 2018b; Wright et al. 2021), the models also control for variables related to respondents’ Jewish characteristics (identifying their Jewish denomination as Orthodox, participation in informal Jewish education while growing up,¹¹ highest level of formal Jewish education, and whether or not the respondent had ever visited Israel). Because political engagement is likely to be associated with symbolic ideology, and because it may also increase exposure to antisemitism, we also control for political activism, as represented by an index summarizing the number of political activities that the respondent participated in.¹² Because the two surveys had different time references for these questions¹³ we computed z-scores separately for each survey and used those scores in all analyses.¹⁴ Table 4 shows descriptive statistics for all independent variables and controls.

Results

Hypotheses about the relationship between political identity and each form of concern about antisemitism were tested with ordered logit models, each using the original four-category concern variables as the dependent variable (Table 5). Model 1 indicates that contrary to our hypothesis, there is no statistically significant relationship between political identity and concerns about traditional antisemitism. The model also indicates that concerns about traditional antisemitism were significantly higher after the 2021 Israel–Hamas conflict. In contrast, Model 2 finds that, after controlling for other factors, moderate and conservative Jewish young adults are significantly more likely to express concerns about Israel-related antisemitism than liberals, in line with our hypothesis. The model also indicates that, as expected, concern about Israel-related antisemitism was significantly higher in June 2021 than in February 2021.

Table 5. Ordered logit model for being very concerned about traditional and Israel-related antisemitism.

Main models		Model 1 Concern about traditional antisemitism		Model 2 Concern about Israel-related antisemitism
		Coef.	Robust SE	Coef.	Robust SE
Political identity	Liberal	–	–	–	–
	Moderate	−0.044	0.108	0.631***	0.102
	Conservative	0.104	0.113	1.056***	0.126
Date of survey	February 2021 (omitted)	–	–	–	–
	June 2021	0.505***	0.080	0.606***	0.078
Index for political activism		0.351***	0.045	0.146***	0.045
Highest level of Jewish education	None	–	–	–	–
	Supplementary school	0.035	0.102	0.137	0.098
	Day school	0.111	0.134	0.213	0.131
Participated in informal Jewish education growing up		0.002	0.099	0.040	0.091
Belonging to Orthodox denomination		0.009	0.108	0.192	0.112
Visited Israel		0.092	0.102	0.445***	0.103
Gender (female)		0.365***	0.083	0.592***	0.084
Age		−0.013	0.010	−0.035***	0.010

***p < 0.001.

**p < 0.01.

*p < 0.05.

To illustrate the magnitude of these effects, Figure 1 presents the predicted probabilities of being “very much” concerned about each form of antisemitism (and 95% confidence intervals) by ideology, derived from Models 1 and 2, with all other factors held at mean or modal values. As can be seen, all three ideological groups have an approximately 50% likelihood of being “very much” concerned about traditional antisemitism, holding other factors constant. In contrast, liberal Jews have only a 31% probability of being “very much” concerned about Israel-related antisemitism, compared to 56% for conservative Jews.

Figure 1. Predicted probability of being “very much” concerned about traditional and Israel-related antisemitism by symbolic ideology (error bars indicate 95% confidence intervals).

Note: Predicted probabilities derived from Models 1 and 2 in Table 5. Other variables held at mean or modal values.

Models 1 and 2 suggest that concerns about both forms of antisemitism were higher in June of 2021 compared to February. Because the samples are largely independent, we conducted an additional analysis limited to the 272 individuals who responded to both surveys. This analysis provides a longitudinal assessment of the changes in concerns about both forms of antisemitism between February and June 2021. Figure 2 shows that there was a dramatic, and statistically significant increase in the proportion of these individuals who reported being “very much” concerned about each form of antisemitism between these two time points.

Figure 2. Weighted proportion of respondents to both surveys who are “very much” concerned about traditional and Israel-related antisemitism by survey date (error bars indicate 95% confidence intervals).

Note: Analysis limited to 272 respondents who complete both February and June surveys. Design-adjusted chi-square tests were significant at p < .05 for both forms of concern.

Having verified that there was an overall increase in concerns about antisemitism among this population between February and June of 2021, we return to the full sample of respondents and test whether this change is moderated by ideology. The models in Table 6 investigate this question by adding an interaction term between ideology and survey date to each of the ordered logit models described above. The main effects for conservative and moderate political identity are positive and significant in Model 4 but non-significant in Model 3. This indicates that in February 2021 (before the Israel–Hamas conflict) liberal Jews were less likely to be concerned about Israel-related antisemitism than moderate and conservative Jews, but no more or less likely to be concerned about traditional antisemitism. In both models, the main effect of the survey date is positive and significant, indicating that liberal Jews were more concerned about both traditional and Israel-related antisemitism in June, compared to February. In Model 3, the interaction term for conservative identity is positive and significant, indicating that the increase in concerns about traditional antisemitism was more acute for conservatives than for liberals. In Model 4, the interaction terms are non-significant, indicating that the increase in concern about Israel-related antisemitism was not significantly different for moderates or conservatives than it was for liberals.¹⁵

Table 6. Ordered logit model for being very concerned about traditional and Israel-related antisemitism with interaction terms.

		Model 3 Concern about traditional antisemitism		Model 4 Concern about Israel-related antisemitism
		Coef.	Robust SE	Coef.	Robust SE
Political views	Liberal	-	-	-	-
	Moderate	−0.142	0.143	0.554***	0.138
	Conservative	−0.123	0.158	0.978***	0.184
Date of survey	February 2021(omitted)	-	-	-	-
	June 2021	0.372***	0.105	0.546***	0.101
Political views*June 2021	Liberal	-	-	-	-
	Moderate	0.193	0.198	0.151	0.185
	Conservative	0.425**	0.213	0.147	0.230
Index for political activism		0.343***	0.045	0.142**	0.046
Highest level of Jewish education	None	-	-	-	-
	Supplementary school	0.027	0.102	0.134	0.098
	Day school	0.092	0.135	0.207	0.131
Participation in Jewish camp or Jewish youth group		0.005	0.100	0.039	0.092
Belonging to Orthodox denomination		0.066	0.110	0.213	0.115
Visited Israel in the past		0.086	0.102	0.445***	0.103
Gender (female)		0.364***	0.083	0.593***	0.084
Age		−0.013	0.010	−0.035***	0.010

*** p < 0.001.

**p < 0.01.

*p < 0.05.

To visualize these relationships, Figures 3 and 4 present the predicted probability of being “very much” concerned about each form of antisemitism for liberal, moderate, and conservative Jews at each time point, derived from Models 3 and 4, respectively. Figure 4 shows that although liberals were less concerned than moderates and conservatives about Israel-related antisemitism at each time point, the level of concern among each group increased by a similar magnitude. In contrast, Figure 3 shows that while the proportion of liberals who were very concerned about traditional antisemitism only increased by around 10 percentage points, the proportion of conservatives who were “very much” concerned increased by almost 20 percentage points. The increase for moderates was 12 percentage points. However, as the 95% confidence intervals show, even after the Israel–Hamas conflict, liberal, moderate, and conservative Jews did not significantly differ in their likelihood of being “very much” concerned about traditional antisemitism.

Figure 3. Predicted probability of being “very much” concerned about traditional antisemitism by survey date and symbolic ideology (error bars indicate 95% confidence intervals).

Note: Predicted probabilities derived from Model 3 in Table 6. Other variables held at mean or modal values.

Figure 4. Predicted probability of being “very much” concerned about Israel-related antisemitism by survey date and symbolic ideology (error bars indicate 95% confidence intervals).

Note: Predicted probabilities derived from Model 4 in Table 6. Other variables held at mean or modal values.

Discussion

Although concerns about antisemitism are influenced by political ideology, the findings of the present study suggest that rising antisemitism is unlikely to exacerbate the divide between liberal and conservative Jewish young adults in the United States. Contrary to our initial hypothesis, concerns about traditional antisemitism do not appear to be related to political ideology. As hypothesized, however, conservative and moderate young Jews are more likely to express concerns about Israel-related antisemitism than liberal Jews. Although disagreements about the dangers of Israel-related antisemitism may divide liberal and conservative Jews, traditional antisemitism clearly represents a cross-cutting concern, where Jews of divergent political identities may find a point of agreement. Furthermore, the fact that concerns about traditional and Israel-related antisemitism relate to political identity in different ways provides further confirmation that these two concerns represent distinct constructs in the minds of American Jewish young adults.

At the same time, even though liberal and conservative Jews differ in their baseline levels of concern about Israel-related antisemitism, they do not seem to differ dramatically in their reaction to an exogenous increase in the threat it poses. Comparing data from February and June 2021 strongly suggests that liberal, moderate, and conservative Jews became more concerned about Israel-related antisemitism in the aftermath of the 2021 Israel–Hamas conflict and the related attacks on American Jews, and did so at similar rates.

This suggests that the ideological disparity in the overall level of concern about Israel-related antisemitism is unlikely the result of liberal Jews failing to notice or acknowledge Israel-related antisemitism due to its association with their own political “team.” Rather, this disparity is more likely driven by differences in how liberal and conservative Jews feel about criticism of Israel. Since conservative Jews are more likely to see any criticism of Israel as illegitimate, they appear more likely to see especially hostile or virulent criticisms of Israel as crossing the line into antisemitism. Liberal Jews, by contrast, may be less likely to code the same forms of criticism as antisemitic, since they are more likely to see criticism as a healthy part of “constructive” patriotic engagement with both the United States and Israel (Wright, Saxe, and Wald 2022). Nevertheless, the fact that liberal Jews had a 31% probability of being very concerned about Israel-related antisemitism before the Israel–Hamas conflict illustrates that the connection between hostility to Israel and antisemitism is not merely a concern among politically conservative Jews.

Given these different perspectives, it is natural for liberal and conservative Jews to disagree about the prevalence of Israel-related antisemitism in the United States and the threat it poses to American Jews. However, when American Jews were attacked or harassed because of the actions of the Israeli government, both liberal and conservative Jews became more concerned about Israel-related antisemitism. Although Becker (2020) finds that Jews follow cues from their own political identity when forced to choose which side – the left or right – bears the greatest responsibility for antisemitism in the United States, our results suggest that concerns about antisemitism, regardless of its source, can still transcend ideological divisions. We find no evidence that an exogenous increase in the threat of Israel-related antisemitism would drive liberal and conservative Jews further apart, at least insofar as such an increase involves violence and harassment targeted at individual American Jews.

Jewish concerns about traditional antisemitism also increased between February and June 2021. One possible explanation is that young Jews’ perception that they are more vulnerable to one form of antisemitism increases their concerns about antisemitism writ-large. Another possibility is that although they clearly represent distinct constructs and are related to ideology in distinct ways, young American Jews see traditional and Israel-related antisemitism as deeply connected to one another. As noted above, the distinction between traditional and Israel-related antisemitism can become blurry in some situations, such as when purported Jewish “loyalty” to Israel is treated as a marker of disloyalty to America, or when the word “Zionists” is used as code for “Jews” in contemporary restatements of classic antisemitic conspiracy theories (Rosenberg 2022). Indeed, although media reports about antisemitic violence and rhetoric in summer 2021 explicitly situated them in the context of criticism of Israel, they also noted the use of Nazi rhetoric (e.g., “Hitler was right”) and “age-old antisemitic tropes” (Breslow 2021).

The findings also indicate that the increase in concerns about traditional antisemitism over summer 2021 appear to have been concentrated among Jewish conservatives. One piece of context for this result is the fact that there is no significant relationship between ideology and levels of concern about traditional antisemitism, either before or after the Israel–Hamas conflict. It may be that despite the popular and scholarly connection between traditional antisemitism and the political right, young conservative Jews do not make this association. For them, traditional antisemitism may be a less ideologically laden form of prejudice, which carries little or no psychological “baggage” in reference to partisan or ideological social identity. In sum, these results suggest that if we had conducted similar studies before and after a high-profile example of antisemitic violence motivated by traditional anti-Jewish stereotypes (such as the shootings in Poway or Pittsburg), we would expect to see a similar increase in concerns about traditional antisemitism among both liberal and conservative Jews.

Conclusion

Even those who share a religion, culture, and ethnic background can be divided by politics. Like the American electorate in general, American Jews have become increasingly divided into liberal and conservative camps and see one another as political “others,” notwithstanding their many commonalities. In the midst of this crisis, American Jews face another: a dramatic rise in antisemitism in the United States. Although antisemitism remains an urgent threat to American Jews, these results suggest that rising antisemitism is unlikely to further divide Jews politically. Despite some disagreements about the exact point at which criticism of Israel crosses the line into antisemitism, young liberal and conservative Jews are, by and large, united in their concerns about traditional and Israel-related antisemitism. The recognition of a common threat may help American Jews both effectively resist antisemitism and mitigate some of their own political divisions.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Additional information

Funding

This work was supported by Brandeis University and Birthright Israel.

Notes

1 Throughout this paper, the term “conservative” refers to the American political ideology, as opposed to the “Conservative” Jewish denomination.

2 In one of the few efforts to experimentally study the relationship between antisemitism and political ideology in the United States, Hersh and Royden (2022) find that, although antisemitic attitudes were present on both the left and the right, antisemitic attitudes were far more prevalent among those on the right, especially among younger conservatives.

3 The IHRA does not itself distinguish between “traditional” and “Israel-related” antisemitism but includes both of the Israel-related examples cited here alongside other numerous examples (e.g., perpetuating the “the myth about a world Jewish conspiracy or of Jews controlling the media, economy, government, or other societal institutions”) in its “working definition” of antisemitism (International Holocaust Remembrance Alliance 2016).

4 Others (Schatz, Staub, and Lavine 1999) refer to this same construct as “blind patriotism.”

5 Conservative Jews might still believe that discrimination toward other minority groups (e.g., Blacks or Hispanics) is not a major problem, while still believing that discrimination toward Jews is a problem. But other work (Cohen 2018b) has argued that Jewish perceptions of antisemitism are linked to perceptions of discrimination toward other marginalized groups and that “when Jews perceive a high degree of discrimination against non-Jews, they also are more likely to perceive a high level of discrimination against themselves” (p. 408).

6 Other work has argued that perceptions of discrimination (e.g., antisemitism) could actually alter political identities (Kuo, Malhotra, and Mo 2016; Huddy, Mason, and Aarøe 2015), meaning that a correlation between liberalism and concerns about traditional antisemitism might be due to perceptions of traditional antisemitism causing young Jews to identify as liberal. This research, however, is primarily based on immigrant populations (namely, Hispanic and Asian Americans) whose connection to American partisan politics is likely less robust than that of American Jews, who, by and large, are not recent immigrants. Furthermore, recent work has failed to confirm a causal relationship between perceived discrimination and political identity, suggesting that, even among immigrant groups, the observed correlation may be due to political identity influencing perceptions of discrimination (Hopkins et al. 2020).

7 Applicants to Birthright were eligible to participate if they met the following criteria: were between the ages of 18 and 32, had at least one Jewish birth parent or had completed Jewish conversion through a recognized Jewish denomination, and had not participated in an educational trip since they turned 18 and/or lived in Israel past the age of 12.

8 See, e.g., Wright, Saxe, and Wald (2022, 46), who note that 11% of 2018 Birthright applicants “strongly agreed” that Israel “is guilty of violating the human rights of the Palestinian people.”

9 These 272 individuals thus represent 544 observations across the two surveys. To account for the clustering of observations within individuals, cluster robust standard errors are reported for all analyses reported below.

10 In these models, we intentionally do not control for the respondents’ concern about the “other” form of antisemitism. Adding such a control would transform the model into an analysis of relative concern, which, as discussed above, would make it impossible to determine if ideology was associated with one form of concern but not the other. Even if liberal and conservative Jews had identical levels of concern about traditional antisemitism (but different levels of concern about Israel-related antisemitism), we would still expect to see a significant coefficient for ideology in a model of traditional antisemitism that also controlled for Israel-related antisemitism.

11 Attending a Jewish overnight camp while growing up and/or belonging to a Jewish youth group in high school.

12 The political activities presented to respondents include participating in a protest march, rally, or demonstration; signing a petition about a political or social issue; called, written, or visited an elected official; boycotting a company or brand; donating money to any organization concerned with a political or social issue; attending a meeting of any level of government; submitting a letter to the editor or opinion piece to a media publication; posting/sharing political content on social media.

13 The February 2021 survey asked respondents to check all activities in which they participated in the last three months, while the June 2021 survey asked participants about all activities since January 2021 (representing five months between the time indicated and the date of data collection).

14 Cronbach’s alpha was greater than 0.7.

15 Although the results may not be reliable due to the small sample size, we also re-ran Models 3 and 4 on the sub sample of individuals who responded to both February and June surveys (N = 468). In neither model were the moderate or conservative interaction terms statistically significant at the 95% level.