Abstract
Most research on solidarity focuses on democratic, low-repressive contexts. However, support for Palestinians in the Global North shows that solidarity can also emerge in democracies with issue-specific repression, where costs and risks for solidarity are higher, and dominant narratives limit alternative perspectives. This article explores which beliefs, emotions, and attitudes predict low-cost (i.e., low effort/risk) and high-cost (i.e., high effort/risk) solidarity in such contexts. We conducted three studies during major Israel/Palestine escalations: a 2009 German convenience sample (N = 305) and two 2024 representative samples from Germany (N = 412) and the United Kingdom (N = 409). Perceived peaceful intentions and guilt toward Palestinians predicted both types of solidarity. Perceived injustice and moral outrage were more linked to low-cost solidarity, while perceived collective ownership of the land was stronger for high-cost solidarity. Power imbalance, admiration, sympathy, hate, and antisemitism played no or minor roles for solidarity in these contexts. The results highlight the distinct nature of conflict-related solidarity under issue-specific repression compared to solidarity under low repression.
“Everyone loves diversity and inclusion—until you stick up for Palestinians”—Moustafa Bayoumi (author, journalist, professor)
Introduction
Over the last decade, there has been an increased interest in solidarity practices and their effectiveness to reach equality and justice for oppressed groups. Solidarity refers to actions taken by individuals from advantaged (i.e., allyship) and disadvantaged (i.e., interminority solidarity) groups with the aim to dismantle power inequalities, and to foster inclusion and amplify the voices of oppressed groups within a society (Neufeld et al., 2019; Radke et al., 2020). These actions can also be performed by members of groups outside of the targeted society (i.e., third-party solidarity or transnational solidarity; Davis, 2017). Particularly for intergroup conflict, third-party solidarity can be relevant since it might influence third-party governments to implement policies that could change the course of a conflict (e.g., sanctions; Uluğ et al., 2024). Although third-party solidarity has gained more attention from psychological research recently (Saab et al., 2015; Thomas et al., 2018, 2019), research on it remains scarce and currently focused primarily on democratic, less repressive contexts. However, we have witnessed over the past years (and even before that) a variety of low-cost and high-cost actions of third-party members in support of Palestinians in the Global North. The context of these actions differs enormously from previous work on solidarity since
(1) Many of the democracies in which these solidarity actions have taken place are not typical third-party countries (as in neutral bystanders) but complicit actors in the oppression of Palestinians (e.g., by providing weapons to Israel; Action on Armed Violence, 2024).
(2) Based on this complicity, although these countries emphasize freedom of speech, they actively repress voices in solidarity with Palestinians within their democratic context (Amnesty International, 2024; Browne et al., 2025), thus making Palestine solidarity far more risky and the development of certain beliefs about the conflict in favor of Palestinians (e.g., settler-colonialism) that diverge from the dominant narrative far more difficult for residents in these countries.
(3) Due to this issue-specific repression, solidarity often does not have immediate outcomes, placing another barrier for people to become active (Acar et al., 2024).
(4) Since residents in these democratic countries can express their solidarity for other oppressed groups (e.g., Ukrainians, women) without or lower personal costs, and with a potential for higher immediate outcomes, they might focus on other causes and might be less motivated by predictors that do not elicit a strong moral stance or emotion (e.g., sympathy might be too weak in these contexts).
Considering the above, previous work highlighted that higher costs for action reduce solidarity intentions (Wiltfang & McAdam, 1991). Additionally, other work points out that people are less likely to voice opinions publicly that are not aligned with the perceived mainstream (see spiral of silence, Noelle-Neumann, 1974). Nevertheless, the example of Palestine solidarity shows that people are willing to act and speak out against dominant narratives despite the potential costs. Thus, the question arises: What motivates third parties in democratic, but issue-specific repressive countries to be active for Palestinians despite the higher prevalence of dominant narratives and the potential personal costs of becoming active?
With this contribution, we aim to address this research question by investigating some of the motivators for third-party members to support Palestinians in democratic yet issue-specific repressive contexts (i.e., in Germany and the United Kingdom). For this, we will bring the solidarity/allyship and the intergroup conflict literatures together by identifying predictors for supporting oppressed groups in such contexts. With three studies, we aim to make the following contributions: (a) To identify which of the predictors for solidarity/allyship in less repressive contexts also function as predictors for solidarity for Palestinians in more repressive “third-party” contexts, (b) to introduce novel predictors that might be relevant for solidarity toward conflict contexts and have only been discussed in the intergroup conflict literature so far, (c) to shed more light on the willingness of third parties to show different types of solidarity (low- vs. high-cost) depending on the socio-polticial context (i.e., democraties with different levels of repression), and (d) to scrutinize the role of antisemitism for third-party solidarity for Palestinians since this is often discussed as a core motivator, and one of the justifications for the repression of Palestine solidarity. In the following sections, we shortly introduce the case of Palestine, and then elaborate on diverse predictors of third-party solidarity, and how they may be applicable for solidarity with Palestinians under issue-specific repression.
The Historical and Ongoing Oppression of Palestine
At the time of writing up this paper (in October 2025), we have been witnessing an ongoing genocide in Gaza for over 2 years (B’Tselem, 2025; United Nations, 2024). Simultaneously, there has been a rise in physical and structural violence toward Palestinians in the West Bank (Aly et al., 2025). However, the oppression of Palestinians did not start on October 7th, 2023, when Hamas attacked Israel, which resulted in the death and abduction of hundreds of people (Human Rights Council, 2024). For Israelis, in the early 20th century, many Jewish people, primarily from Europe, fled their own persecution (including their genocide during World War II, i.e., the Holocaust) and their century-long oppression, and sought to create a Jewish state to gain their own independence in a territory of cultural, religious, and historical importance to them, which led to the founding of Israel in 1948 (i.e., Zionism; Herzl, 1896). However, this land, under British Mandate at the time, was already inhabited by Palestinians. Consequently, for Palestinians, this has meant for decades that their land has been stolen, they have been expelled (e.g., around 750,000 during the foundation of Israel, referred to as Nakba) or killed through various settler-colonial means (i.e., ethnic cleansing; Anderson, 2018; B’Tselem, 2011; Pappé, 2008). In addition, Palestinians are deprived of basic human rights on a day-to-day basis (e.g., clean water, healthcare, education; OCHA, 2023; UN, 2022). The West Bank has been labeled by human rights organizations as apartheid (Amnesty International, 2022; UN, 2022), and Gaza has been under siege since 2007 (B’Tselem, 2017).
Palestine Solidarity in Europe: A Case of Issue-Specific Repression in Democracies
Europe is the highest-performing region for democratic practices in the world (International IDEA, 2024). However, recent developments threaten this status due to government and media-related repressions and censorship on specific issues. One of the most repressed issues in Europe is the dissemination of information and support for Palestine (Amnesty International, 2024). In our studies, we focus on two European countries that are particularly relevant when it comes to the repression of Palestine solidarity due to their own historical and remaining complicity in the oppression of Palestinians: Germany and the United Kingdom. At this point, we note that we do not equate issue-specific repression with the repression in authoritarian regimes. Instead, we emphasize that we conceptualize repression as a continuum that can also occur within democracies (cf. Ayanian et al., 2025) and that often contradicts a country’s broader democratic values.
Germany, being responsible for the Holocaust and therefore seeing Israel’s security as Germany’s “reason of state” (L. Fischer, 2019), and the United Kingdom, as the former colonizer of Palestine, have strong historical ties to the region. Both countries cultivate strong political, religious, military, and academic relationships with Israel, thus making them some of Israel’s strongest allies within Europe (L. Fischer, 2019; Foreign, Commonwealth & Development Office, 2023). Criticism of Israel is often perceived as antisemitic in both countries (Fekete, 2024; Veracini, 2024). Based on this claim, repression of Palestine solidarity existed in Germany in the past (e.g., protest bans in early 2023) and has increased since October 7, 2023, both in Germany and the United Kingdom (Amnesty International, 2024; Browne et al., 2025). Among the repression techniques are protest bans, police crackdowns, legal fines, or being classified as antisemitic or a terrorism supporter (Fekete, 2024; Index of Repression, 2025). In addition to direct forms of repression, the Israeli narrative is predominantly endorsed by these governments and the media, censoring Palestinian voices and portraying them negatively (e.g., as terrorists; Kareem & Najm, 2024).
Based on the repression of Palestine solidarity, it is much more difficult to form beliefs that are supportive of Palestinians and diverge from dominant narratives (e.g., Palestinians having bad intentions/being terrorists; equal responsibility and suffering on both sides). Furthermore, actions in support of Palestinians can be relatively costly compared to actions that support other oppressed groups within these democratic societies (e.g., supporting LGBTQ+ in Germany). Although both countries are repressive toward Palestine solidarity, we acknowledge that, during the time of data collection, this issue-specific repression was still much higher in Germany compared to the United Kingdom, making public actions, even those requiring low effort (e.g., wearing symbols), more costly in this context (Amnesty International, 2024; Schreiber, Uluğ, & Drury, 2025).
Furthermore, although we acknowledge that these countries are complicit actors in the oppression of Palestinians, we will refer to the individuals in these countries as third-party members. Many citizens of these countries might not be aware of their countries’ involvement since these countries are not direct actors (i.e., neither Palestinians nor Israelis), and information about the situation and their countries’ complicity is repressed. Thus, the solidarity of the general public might be seen as an “outsider” role.
Motivators of Solidarity and Research Gaps
Previous work classified the motivations of solidarity and allyship in relation to individuals’ identification, beliefs, emotions, and attitudes (Louis et al., 2019; Radke et al., 2020; Woo et al., 2025). So far, much research on the predictors of solidarity has been based on collective action literature, and replicated that identification with the oppressed, perceived injustice, perceived efficacy, morality, and moral outrage are also predictors of solidarity (Agostini & van Zomeren, 2021; Saab et al., 2015; Thomas et al., 2018). Apart from these well-established predictors, research has begun to identify the potential role of positive emotions in fostering solidarity, such as sympathy, hope, and pride (Thomas et al., 2021; van Zomeren, 2021).
In a recent paper, Uluğ et al. (2024) calls for an expansion of the predictors of allyship within conflict contexts beyond the established predictors of allyship in peaceful contexts. They argue that other predictors such as conflict-specific beliefs (e.g., narratives of victimhood, power asymmetries) and moral convictions (e.g., commitment to nonviolence), emotions toward conflicting parties (e.g., moral obligation, hate), and ideological attitudes (e.g., decoloniality) have been neglected in research but should be considered to better explain why people engage in solidarity toward intergroup conflict under different conditions of cost. Moreover, when it comes to the repression of movements, we currently do not know much about the motivators for people to show solidarity, since most work on repression has focused on the motivators for people fighting for their ingroup’s rights (Ayanian et al., 2025). However, we know from peaceful contexts that collective action for ingroup and outgroup injustices differ in their motivations (Saab et al., 2015). Additionally, other work points out that dominant narratives in the Global North, including research in psychology, have labeled the situation in Israel/Palestine as “intractable” or “asymmetric” conflict, although different disciplines consider the context as one of settler-colonialism (Hakim et al., 2023; Pappé, 2006, 2008). This neglect leads to the oversight of relevant constructs when studying people’s perceptions and behavioral motivations toward the context, such as seeing Palestinians as the indigenous population or them resisting their oppression rather than being terrorists (Hakim et al., 2023). We aim to address these gaps by focusing on third-party solidarity for Palestinians in issue-repressive contexts. Specifically, we examine whether common predictors of low-cost solidarity that are also relevant for conflict contexts (e.g., perceived injustice, guilt, sympathy) replicate in this context, while also testing additional conflict-specific factors, including some that might originate from less dominant conflict narratives (e.g., collective ownership beliefs of Palestinians), drawn from the intergroup conflict literature.
Potential Motivators of Third-party Solidarity in Conflict Contexts
Beliefs About the Conflict and the Conflict Parties
Similar to conflict parties, third-party members have narratives about external conflicts that might explain their support for one group (Cohrs et al., 2015; Hill et al., 2018). In the case of Palestine, qualitative work has identified common perceptions that third-party supporters for Palestinians have: (a) Palestinians treatment being unjust, (b) Palestinians being the rightful owners of the land, (c) Palestinians having peaceful intentions and (d) Palestinians having less power compared to Israelis (Hill et al., 2018; Momberg, 2017; Stahel & Cohrs, 2015). While the first has been studied for solidarity under low repression (Saab et al., 2015), the other three motivators have received little attention in quantitative work on third-party solidarity. In the section below, we will discuss the potential role of these four beliefs in third-party solidarity for Palestinians under issue-specific repression.
Perceived Injustice Toward the Oppressed Group
Perceived injustices, such as the attacks on Palestinian civilians by Israel, play an important part in conflict narratives about Israel/Palestine among third parties (Cohrs et al., 2015; Momberg, 2017). Research has consistently found that perceived injustice motivates solidarity (Agostini & van Zomeren, 2021). However, acknowledging the suffering of Palestinians is still within dominant narratives and does not require knowledge about the historical and systematic foundations of these injustices. Neglecting these foundations might lead people to justify the injustice (e.g., seeing the genocide in Gaza as self-defense of Israel and unavoidable when fighting terrorism), or to emphasize that injustices are faced on both sides equally (Stahel & Cohrs, 2015). Consequently, perceived injustice might be too weak to push people to action when dominant conflict narratives make the interpretation of these injustices arbitrary or when personal cost is high. Thus, perceived injustice might be a weaker predictor, predicting only low-cost solidarity in repressive contexts rather than high-cost solidarity.
Rightful Collective Ownership of the Land by the Oppressed Group
Many intergroup conflicts are rooted in disputes over who is the rightful owner of the land. Collective psychological ownership of the land consists of several aspects including (a) who first inhabited the land, (b) whose collective identity was shaped by the land, and (c) who cultivated the land (Verkuyten & Martinovic, 2017). In the context of Israel/Palestine, collective ownership of the land is the core aspect of the conflict since Israelis and Palestinians both claim to be the rightful owners of the land, and settler-colonialism is central to the conflict narrative of Palestinians and some of their third-party supporters (Cohrs et al., 2015). The literature on collective action highlights that perceived rights and their violations can motivate political action (Kutlaca et al., 2017). In line with this, recent findings highlight that collective ownership beliefs impact political preferences in conflict, such as support for outgroups to form their own state (Storz et al., 2022). So far, collective ownership beliefs have only been examined among groups in conflict and only for attitudinal support rather than solidarity actions (e.g., Selvanathan et al., 2021; Storz et al., 2020). Our paper will introduce this construct to the solidarity literature based on previous qualitative work that hints at its importance (Cohrs et al., 2015). Since perceiving Palestinians to be the collective owners of the land diverges from the dominant narratives of complicit third-party countries and acknowledges the historical/systemic foundations on the conflict, we expect people who hold this belief to be more willing to engage in solidarity regardless of cost since: (a) they should be more resistant to repressive tactics that justify the injustice, and (b) they may perceive higher severity of the situation based on the perceived right violations (e.g., settler-colonialism). Thus, in issue-specific repressions, collective ownership beliefs should be more relevant for third-party solidarity than perceived injustice.
Perceived Peaceful Intentions of the Oppressed Group
In the intergroup conflict literature, beliefs about victimhood significantly shape people’s actions toward a group (Uluğ et al., 2024; Vollhardt, 2015). When people determine who is the victim of a conflict, they focus (among other things) on whether a group has peaceful intentions, such as innocence and responsibility for the conflict (Jankowitz, 2018). Thus, perceiving the intention of a group as peaceful increases people’s participation in collective actions on its behalf (Bruneau et al., 2017; Ünal et al., 2022). Qualitative research also showed that some supporters of Palestinians discuss actions by Palestinians as self-defense and do not see them as the initiator of the conflict (Cohrs et al., 2015; Kempf, 2010; Momberg, 2017). Holding this belief challenges the dominant narrative in complicit third-party countries that portrays Palestinians as terrorists or initiators of the conflict (Abu-Laban & Bakan, 2021). Especially for intergroup conflicts, where people can perceive conflicting groups to be both victim and perpetrator, perceptions of a clearer victim should increase support for that group. It may also strengthen third-party members’ resistance to repression tactics, such as dehumanizing the oppressed. We, thus, predict in line with previous solidarity literature that perceiving the oppressed group as having peaceful intentions will be a strong, positive predictor of third-party solidarity, regardless of cost.
Perceived Power Imbalance
The perceived oppression of victim groups (e.g., having lower power) most often lies at the core of third-party support and solidarity (Bruneau et al., 2017; Cohrs et al., 2015). Hereby, power encompasses who has control over resources (e.g., natural, financial, legal, social) and who can shape the outcome of a conflict (Rouhana & Fiske, 1995). Research on intergroup conflict highlights that people not only recognize power imbalances between groups, but also vary in the degree to which they perceive these imbalances (Rouhana & Fiske, 1995). Despite this, no research to date has investigated whether the degree of perceived power imbalance predicts solidarity. The degree seems to be important for third-party solidarity since people might consider the relative disadvantage of a group to judge whether the group needs outside help to overcome oppression or is strong enough to do so independently. However, it is unclear whether perceiving power imbalance is enough to act, or if repression tactics might actually justify this power imbalance (similar to perceived injustice), thus not leading to action. We, therefore, will introduce this concept to the solidarity literature and expect that the lower the relative power of the oppressed group is perceived compared to the oppressor group, the more willing third parties are to engage in solidarity (see Studies 2 & 3).
Emotions Toward the Conflict Parties
Feelings of injustice are among the strongest predictors of (solidarity-based) collective action (Agostini & van Zomeren, 2021; Thomas et al., 2021). Third parties can experience distinct emotions toward several groups involved in intergroup conflict, which lead to different goals for becoming active (Landmann et al., 2025). Therefore, below, we discuss the potential role of injustice-related emotions toward (a) the oppressed (i.e., sympathy, admiration, and guilt) and (b) the oppressor (i.e., moral outrage and hate) for third-party solidarity under issue-specific repression.
Emotions Toward Oppressed Group
Sympathy is felt when people perceive that others face injustice without holding responsibility for their suffering (Feather & Nairn, 2005). Sympathy was found to motivate third-party solidarity actions, such as demonstrating (Saab et al., 2015; Thomas et al., 2018). Nevertheless, pervious work focused on solidarity actions toward issues that at that time bore lower risks because the support of these causes was tolerated, such as supporting Palestinians in the United Kingdom in 2008, participating in vigils for the 1989 Tianmen massacre in Hong Kong in 2009, or demonstrating for Ukraine in Germany in 2022 (Ditrich & Sassenberg, 2024, Saab et al., 2015). In contrast, when solidarity is risky and requires challenging the government’s repression, sympathy might not be enough to mobilize people, as those who feel sympathy may still identify with the authority and therefore not challenge it (Thomas et al., 2009). We will, thus, examine whether we can replicate sympathy as a predictor of third-party solidarity in contexts where solidarity is much more repressed. We expect that for low-cost actions, sympathy will be a positive predictor of third-party solidarity, while it will not be a predictor of high-cost third-party solidarity in these contexts.
Admiration occurs when outgroup members are seen in a very positive light, including their moral elevation (Onu et al., 2016). Some studies in the solidarity and intergroup conflict literature suggest that (certain aspects of) admiration motivates support for policies that favor a victimized group and challenge the status quo (Čehajić-Clancy et al., 2024; Sweetman et al., 2013). Third-party members might admire an oppressed group’s resilience to their oppression, and might be inspired to become active despite obstacles. Thus, we expect to replicate the finding that admiration toward the oppressed will positively predict third-party solidarity, for both low- and high-cost third-party solidarity and should, be a more relevant predictor compared to sympathy.
Guilt occurs from feeling responsible for one’s own or ingroup’s (in)actions that result in the suffering of a group (Thomas et al., 2021). On the one hand, guilt seems to merely encourage actions with lower costs (e.g., donations) instead of promoting genuine, effortful support for social change (e.g., supporting affirmative actions; Iyer et al., 2007; Thomas et al., 2009). On the other hand, some work found that guilt motivates people to restore justice or demonstrate for the oppressed (Selvanathan et al., 2018). Since the literature is mixed on the role of guilt in solidarity, we exploratively examine guilt toward the oppressed as a predictor of third-party solidarity in contexts where solidarity is more repressed.
Emotions Toward the Oppressor Group
Moral outrage is felt anger toward perceived injustices committed by an outside party (Batson et al., 2007; Thomas et al., 2009). Previous research consistently found that moral outrage leads individuals (including third parties) to engage in (solidarity-based) collective action under varying levels of repression (e.g., protest, sign a petition; Dono et al., 2025; Saab et al., 2015). Hence, we expect to replicate moral outrage as a positive predictor for third-party solidarity in contexts where solidarity is more repressed, regardless of cost.
Hate is an intense negative emotion toward a group that has been mainly discussed in accelerating destructive tendencies (Halperin et al., 2009). Apart from its connection to offensive actions, empirical evidence has recently highlighted that hate has a moral component (Pretus et al., 2023), thus making it a potential motivator for engaging in actions aimed at restoring justice. Although the impact of hate on third-party solidarity is regularly discussed in the media (e.g., labeling Palestine solidarity demonstrations as “hate marches”; Fekete, 2024), hate has not been the focus of research on third-party solidarity, but merely gained some attention in the intergroup conflict literature (A. Fischer et al., 2018; Halperin et al., 2009). Due to its close relationship to injustice perceptions and its importance in intergroup conflict, we will introduce hate as a potential predictor to the solidarity literature. While we do not expect hate to be a core driver of solidarity, we acknowledge that hate might be a stronger moral emotion than moral outrage, and thus potentially serves as a weak predictor of third-party solidarity in contexts with issue-specific repressions.
Attitudes Toward the Conflict Parties: Antisemitism
Attitudes toward parties in conflict might also shape solidarity. In many countries, solidarity with Palestinians is claimed to be antisemitic and therefore used as a justification to shut down or socially devalue solidarity with Palestine (Fekete, 2024). However, while it is true that some people’s support for Palestinians might be motivated by antisemitism, support for Palestinians originates in most cases from human rights concerns, critical views of the actions of the Israeli government or anti-Zionism (Beattie, 2017; Kempf, 2015). Due to the high prevalence of the topic of antisemitism in discussions about solidarity with Palestinians, and its use as a justification to repress solidarity with Palestine, it remains relevant to address the relationship between these two constructs empirically. Thus, we test whether antisemitism is a predictor for third-party solidarity with Palestinians, particularly when we consider other predictors that are explicitly directed toward the situation in Israel/Palestine. Similar to previous literature (Gordon, 2024; Kempf, 2015), we expect that antisemitism will not be a predictor of solidarity with Palestinians.
Overview of Studies
The present research aims to synthesize the literature on solidarity and intergroup conflict, and expand previous findings regarding predictors of third-party solidarity, specifically in democratic countries when solidarity is under repression. For this, we conducted two studies in Germany (Studies 1 & 2) and one in the United Kingdom (Study 3) around times of conflict escalation in Israel/Palestine. In Study 1, we examined different groups of previously identified motivators for solidarity, namely beliefs about the conflict (i.e., perceived peaceful intentions, perceived injustice), and emotions (i.e., guilt, sympathy, admiration, moral outrage). We further investigated novel predictors of third-party solidarity from the traditional solidarity and intergroup conflict literature, namely the belief perceived collective ownership of the land and the emotion hate toward the oppressor. Additionally, we examined the relationship between antisemitism and third-party solidarity for Palestinians. We conducted Study 2 to replicate our findings from Study 1 at a different time in the history of the ongoing atrocities, as well as expand our findings by including a novel predictor (i.e., perceived power imbalance), and using a more nuanced solidarity measure that enabled us to distinguish predictors for low-cost and high-cost solidarity intentions. In Study 3, we investigated whether we could replicate our results in another context (i.e., the United Kingdom) that shares similarities with Germany in terms of being a democracy with issue-specific repression toward Palestine solidarity, albeit with this repression being much milder. While research has looked into some of these predictors in isolation, previous work did not consider that many beliefs, emotions and attitudes toward social injustices exist simultaneously. Thus, although these constructs may be predictors or prerequisites of solidarity, some may be more important than others in driving actual action intentions. Therefore, one of our aims was to identify which of these predictors are most influential and which have weaker roles when they are considered together in one model. We controlled for age, gender, religion, political orientation, and self-reported knowledge about the conflict in all our analyses. The studies were not pre-registered. We received ethical approval for all studies from the University of Sussex (ER/JS2146/1, ER/JS2146/7, ER/JS2146/11). All materials of the studies (surveys, data, codes) can be found on OSF: https://osf.io/6jefw/overview?view_only=8a83c0d0a9144458b55998ea6004a73d.
Study 1: Germany in 2009
Study 1 tested our predictors with a sample of Germans shortly after the Gaza war in 2008/2009, during which over a thousand Palestinians were killed, and many actions by Israel were considered to violate international law (Amnesty International, 2009). Although the actions of Israel were critically portrayed in the German media, the media often justified these actions as self-defense and overall portrayed Palestinians more negatively than Israelis (Maurer & Kempf, 2011). Moreover, the German government underscored its (military) support for Israel (Busse, 2012). Overall, the representation of Palestine solidarity in Germany was marginal, and repression of the Palestinian narrative was more indirect through the endorsement of the Israeli narrative and the portrayal of a negative image of Palestine supporters by the government and the media (Schreiber, Uluğ, & Drury, 2025).
Methods
Participants and Procedure
We used secondary data from an online questionnaire conducted in 2009 (Maes et al., 2010), which included a convenience sample of 410 participants. This survey aimed to examine the opinions of people in Germany toward the conflict in Israel/Palestine, and data was originally collected for a thesis that looked into predictors (justice predispositions, emotions, beliefs about origin, responsibility, and norm violation) of support for nonviolent (e.g., involvement of international intermediaries, demonstrations) and violent political actions (e.g., armed fight against the enemy). We excluded seven participants under the age of 18, five participants with substantial amounts of missing data (>20%; Tabachnick & Fidell, 2007), and one participant due to low quality (consistently chose the highest value on the scales). We, furthermore, excluded participants who/whose families were born in Israel/Palestine (n = 25) since we examined third-party solidarity. We included only German citizens since we had no information on whether the foreign citizens were residing in Germany during the time of data collection.
The final sample had 305 participants (68 women, 237 men; age: M = 32.03, SD = 12.18, range = 18–75; education: no/low: n = 6, mid-level: n = 26, high: n = 273). Many indicated being Christian (n = 137) or having no religion (n = 121), followed by Muslims (n = 24), Jews (n = 11), or others (n = 15). Our sample indicated their political ideology to be in the center (10-point Likert scale: 1 = left, 10 = right; M = 4.85, SD = 2.23). People majorly sympathized with Israel (49.2%; scale 1–4), followed by people who were neutral (34.4%; scale 5–7), and the minority sympathized with Palestine (16.4%; scale 8–11; 11-point Likert scale, “With which side of the Israeli-Palestinian conflict do you sympathize more?”: 1 = Israel, 6 = neutral, 11 = Palestine). Participants indicated high knowledge about the situation in Israel/Palestine (M = 4.24, SD = 1.08; 2 items, e.g., “I have knowledge about the Middle East conflict”; 6-point Likert scale: 1 = completely disagree; 6 = completely agree; r = .57, p < .001). A sensitivity test indicated that with our sample, we had a power of 0.8 to detect small effects (Odds Ratio [OR] >= 1.43).
Measures
All items used a 6-point Likert scale: 1 = completely disagree; 6 = completely agree.
Beliefs
Perceived Injustice by Israelis Toward Palestinians
Participants were asked whether they perceived the actions of Israel(is) toward Palestinians as unjust (e.g., “Israeli settlements violate international law”; 6 items, α = .93).
Perceived Peaceful Intentions of Palestinians
The perception of Palestinians’ peaceful intentions was measured with three items (e.g., “I believe that Palestinians would do everything to live peacefully together”; α = .75).
Perceived Collective Ownership of the land by Palestinians
Participants were asked, “Who owns the land?” and indicated to what extent they agreed with four statements regarding Palestinians’ rightful collective ownership of the land (e.g., “The Palestinians, since they have always lived there”; α = .80).
Emotions
Sympathy Toward Palestinians
Sympathy toward Palestinians was measured with three items (e.g., “I feel sympathy for the Palestinians who constantly experience humiliation and hostility”; α = .77; adapted from Schmitt et al., 1995).
Admiration Toward Palestinians
Admiration toward Palestinians was measured with three items (e.g., “I admire the Palestinians’ ability to maintain their dignity in the face of adversity”; α = .89; adapted from Schmitt et al., 1995).
Guilt Toward Palestinians
Guilt toward Palestinians was measured with three items (e.g., “I feel guilty if I do nothing to improve the situation of the Palestinians”; α = .83; adapted from Schmitt et al., 1995).
Moral Outrage Toward Israelis
Moral outrage toward Israelis was measured with three items (e.g., “I could get really outraged about how the Israelis are treating the Palestinians”; α = .87; adapted from Schmitt et al., 1995).
Hate Toward Israel(is)
Hate toward Israel(is) was measured with a short version of two subscales from the Triangular Hate Scale (e.g., “I would never want to have anything to do with Israel”; Sternberg & Sternberg, 2008; 10 items, α = .96).
Attitudes
Antisemitism
Antisemitism was measured with 15 items that encompassed various forms of antisemitism (primary, secondary, Israel-focused, manifest & latent, antisemitic separation; Frindte, 1999; Heyder et al., 2005; Brähler & Niedermayer, 2002; e.g., “Jews have something special and strange to them and do not fit in with us”; α = .93).
High-Cost Solidarity Intentions for Palestinians
Participants indicated their intentions to engage in five solidarity actions for Palestinians: “demonstrate,” “invest a lot of time and money,” “convince others to be an ally for the Palestinian cause,” “convince others to demonstrate for Palestinians,” and “convince others to invest a lot of time and money for Palestinians” (α = .95). We classified our measure as high-cost since the actions required a lot of effort (time, resources) or some risk (e.g., being seen as antisemitic, or going against the dominant narrative; see Uluğ et al., 2024).
Results
Descriptive Statistics and Statistical Procedure
Descriptive statistics and correlations of all variables can be found in Table 1. High-cost solidarity intentions for Palestinians were highly skewed (skewness: 1.17, kurtosis: 0.08), with most participants having no to little intentions to engage in high-cost solidarity (n = 189; mean =< 2). We, therefore, chose to analyze the data with a logistic regression to identify which constructs predict some high-cost solidarity intentions for Palestinians (Boulton & Williford, 2018). For this, we created a binary variable that indicated whether people showed no high-cost solidarity intentions for Palestinians (=0; mean =< 2) or showed some high-cost solidarity intentions for Palestinians (=1; score > 2; n = 116).
Descriptives of and Correlations Between Key Constructs of Study 1.
p < .05. ***p < .001.
Drawing on evidence in support of cognitive appraisal theory (Schreiber, Jaudas, & Maes, 2025; Yeo & Ong, 2024), and on previous research that showed that emotions, including perception about Israel/Palestine, inform attitudes, including attitudes toward Jews, rather than the other way around (Ozer et al., 2025; Vaske et al., 2021), we ran hierarchical logistic regression for which we entered predictors in the order of (1) beliefs, followed by (2) emotions, and finally (3) attitudes. However, we acknowledge that our data were correlational and that the relationships among beliefs, emotions, and attitudes are complex and may be bidirectional (Boden & Berenbaum, 2010). We, therefore, also ran two alternative models with alternate orders of predictors to check the robustness of effects (alternative model 1: step 1. emotions, step 2. beliefs, step 3. attitudes; alternative model 2: step 1. attitudes, step 2. beliefs, step 3. emotions) and reported how these models differed from our main model (for Tables of alternative models see OSF).
While some predictors showed moderate to strong intercorrelations (r > .60; see Table 1), these likely reflect conceptual proximity rather than redundancy since multicollinearity diagnostics were acceptable (Variance Inflation Factor (VIF) < 5), and the constructs were theoretically distinct. Nevertheless, we ran additional analyses where we accounted for potential suppression effects between predictors by removing highly correlated predictors from the model (e.g., running a model for sympathy without admiration and moral outrage).
Predictors of High-Cost Solidarity Intentions for Palestinians
The results of our main hierarchical regression are presented in Table 2, and the results of our alternative models are reported in Tables S1 and S2 on OSF. Control variables significantly explained 18% of the variance of high-cost solidarity intentions, χ2(5, N = 305) = 59.55, p < .001. Being a man, left-wing, and Muslim was associated significantly with stronger high-cost solidarity intentions. Age and self-reported knowledge were not predictors of high-cost solidarity intentions.
Hierarchical Logistic Regression for Predicting the Occurrence of High-Cost Solidarity Intentions for Palestinians in Germany.
Note. OR = Odds ratio.
In the first step, beliefs about the conflict significantly explained additional variance for high-cost solidarity, χ2step(3, N = 305) = 103.58, p < .001, R2 = .42. All measured conflict-related beliefs significantly increased the likelihood of high-cost solidarity intentions; however, perceived injustice became non-significant once emotions were added, across all models. The relationship between perceived injustice and solidarity intentions may be mediated by emotions, thus being a less proximal and weaker predictor compared to the other beliefs that seem to be more robust.
In the second step, emotions explained additional variance to beliefs, χ2step(5, N = 305) = 42.64, p < .001, R2 = .50. Guilt toward Palestinians and moral outrage toward Israelis stayed significant throughout all models, thus making them robust predictors of high-cost solidarity intentions. Sympathy and admiration toward Palestinians, and hate toward Israelis, remained non-significant across models. However, sympathy became significant when emotions were initially entered and highly correlated emotions (admiration, moral outrage) were removed (OR = 1.84 [1.18, 2.86], p = .007), thus sympathy might be a weaker predictor of high-cost solidarity.
In the third step, antisemitism did not significantly predict high-cost solidarity for Palestinians, χ2step(1, N = 305) = 2.97, p = .085, R2 = .50. Similar in alternative model 1, antisemitism was not significant. However, in alternative model 2, where antisemitism was entered first and as the only predictor, antisemitism was initially a significant predictor but became non-significant once beliefs were included. Thus, antisemitism might predict solidarity intentions for some individuals; however, it is not as central as beliefs about and emotions toward the conflict.
Discussion
The results of Study 1 showed that the beliefs about Palestinians—perceived injustice, perceived peaceful intentions, and perceived collective ownership of the land—were associated with some solidarity intentions for Palestinians among Germans. The emotions—guilt toward Palestinians and moral outrage toward Israelis—were furthermore related to a higher chance of solidarity intentions for Palestinians. The emotions—sympathy and admiration toward Palestinians, hate toward Israelis—as well as the attitude, antisemitism, were not significant or very minor predictors of solidarity intentions for Palestinians.
Similar to the literature on solidarity in less repressive contexts (Agostini & van Zomeren, 2021), the most commonly found predictors, perceived injustice and moral outrage, also seem to matter for third-party solidarity in contexts where solidarity is more repressed; however, perceived injustice did not stay significant in our full model, potentially underlining its weaker, less proximal role for third-party solidarity compared to other beliefs and emotions under issue-specific repression. In contrast to generic perceptions of injustice that may still align with dominant narratives, our data underscores that conflict-specific beliefs that diverge from the dominant narratives seem to be more relevant in predicting solidarity intentions. Similar to results in the conflict literature where ownership beliefs motivate conflicting groups’ political intentions (Storz et al., 2022), we found that these beliefs consistently predict third parties’ solidarity intentions. Moreover, we could replicate previous work (Ünal et al., 2022) that perceived peaceful intentions are a consistent predictor of third-party solidarity even when solidarity is more repressed. These results highlight the importance for third-party solidarity to consider beliefs that go beyond dominant narratives, focus on the severity of the situation abroad (i.e., violation of collective ownership rights), and emphasize the victimhood status of the oppressed group (i.e., peaceful intentions).
Unlike some studies in the solidarity literature (Thomas et al., 2009), in our data, guilt was one of the strongest positive predictors of third-party solidarity. It might be that when the situation is severe and people perceive urgency to act, guilt becomes a powerful facilitator for action in the face of repression. Moreover, compared to less repressive contexts (Saab et al., 2015), positive emotions toward the oppressed, sympathy and admiration, were not predictors of third-party solidarity when other emotions and beliefs were considered; however, sympathy was found to be a predictor once highly correlated constructs were removed, thus making it a weaker predictor of third-party solidarity under issue-specific repression. Thus, in contexts where people face more personal risks for being active and need to diverge from dominant narratives, in most cases, positive emotions toward the oppressed might not be strong enough to push people to action.
A crucial finding is that hate toward Israel(is) and antisemitism did not play a crucial role in solidarity with Palestinians among Germans. However, when fewer emotions were entered, hate toward Israelis became a significant predictor, thus pointing to a weak relationship with third-party solidarity. Moreover, when antisemitism was entered into our model as the only predictor, it was initially significant and then became non-significant, indicating that although not central for most people’s third-party solidarity toward Palestinians, it might be relevant for some individuals. Thus, our data support previous research that shows that antisemitism plays only a minor or no role in support for Palestinians (Beattie, 2017; Kempf, 2010). This finding seems plausible since previous work pointed out that antisemitism and anti-Zionism are distinct (Veracini, 2024).
We want to highlight some of the limitations of Study 1. First, we used secondary data and did not use direct measures for all our predictors (e.g., hate). Second, our data consist of a convenience sample with an overrepresentation of men, clear sympathizers toward Israel, and no information on the country of residency during data collection; therefore, we limited our sample to German citizens. However, Germany is a multicultural society with residents from diverse backgrounds who might also participate in solidarity (14% of residents are without a German passport; Bundesregierung, 2024). Third, we also believe our solidarity measure could be more nuanced. In contexts where solidarity is more repressed, it might be relevant to differentiate between solidarity actions that are low-cost (e.g., sign a petition) compared to high-cost (e.g., protest), based on the relative effort and/or risk that people take up to engage in these actions (Uluğ et al., 2024). While the results of Study 1 indicate which predictors might be relevant for high-cost solidarity, this does not automatically imply that third parties would also refrain from low-cost solidarity actions. People might try to find ways to support an outgroup while reducing their own costs and navigating around the hurdles imposed by issue-specific repression by engaging in low-cost actions instead. Nevertheless, even low-cost actions might require developing certain beliefs that deviate from the dominant narrative. Although these actions seem less personally risky, people might still have higher barriers to being supportive compared to less repressed issues, where supportive narratives are more easily accessible.
Study 2: Germany in 2024
To address the limitations of Study 1, we conducted Study 2. First, we examined whether the findings from Study 1 replicated almost 15 years later, again during an ongoing conflict escalation (2024, the Gaza genocide), and heightened repression of Palestine solidarity (Amnesty International, 2024; Fekete, 2024). Similar to Study 1, during the data collection in Study 2, there was high news coverage of the situation in Gaza. Second, we aimed to improve the sample and measurements: We collected data from a representative sample to make predictions that apply to the German population as a whole. We also decided to use a more nuanced measure for solidarity, which distinguishes low-cost and high-cost actions, as well as more nuanced and improved measures for perceived injustice, hate, guilt, and perceived collective ownership. Third, we extended Study 1 by focusing on an additional predictor: perceived power asymmetry. This belief was missing in the previous data collection, although it was found to be a crucial predictor for policy support among conflict parties in asymmetric intergroup conflicts, such as Israel/Palestine (Rouhana & Fiske, 1995), and is one of the most frequently mentioned reasons to support Palestinians among third parties (Cohrs et al., 2015; Momberg, 2017).
Methods
Participants and Procedure
We aimed to collect data from 500 participants based on an a priori power analysis (power = 0.8, OR = 1.3, α = .05). In total, we collected data at the end of May 2024 from 496 participants who were representative of German society regarding age, gender, and education, with the panel provider Bilendi. We excluded 94 participants who failed at least one attention check to improve data quality and avoid Type 1 errors (3 items, e.g., “please click 3 when you read this”; see Abbey & Meloy, 2017; Fullerton & McCullough, 2023).
The final sample had 412 participants (199 women, 210 men, three missing; age: M = 45.06, SD = 15.13, range = 18–99; education: no/low: n = 105, mid-level: n = 131, high: n = 169, other: n = 7). Participants were born (n = 392) or lived in Germany for a long time (n = 18, range: 6–54 years); therefore, they were familiar with the (political and historical) context of Germany. Many were Christian (n = 227) or had no religion (n = 140), followed by Muslims (n = 20), Jews (n = 1) or others (n = 24), with low levels of religiosity (1 = not religious at all, 7 = very religious; M = 2.79, SD = 1.90). Our sample was politically in the center (1 = left, 7 = right; M = 4.02, SD = 1.03). Most people were neutral toward Israel/Palestine (61.9%; scale 5–7), followed by sympathizers with Israel (22.8%, scale 1–4), and a smaller number of participants who sympathized with Palestine (15.3%, scale 8–11). Participants reported having moderate knowledge about the Israel-Palestine conflict (M = 3.87, SD = 1.82; 2 items, e.g., “I have knowledge about the Middle East conflict”; 7-point Likert scale: 1 = completely disagree, 7 = completely agree; r = .85, p < .001). Although we did not reach our planned sample size, a sensitivity test indicated that we still had a power of .8 to detect small effects (OR >= 1.37).
Measures
Participants rated items on a 7-point Likert scale: 1 = completely disagree; 7 = completely agree. We used the same measures as those used in Study 1 to measure perceived peaceful intentions of Palestinians (α = .70), sympathy (α = .84), and admiration (α = .86) toward Palestinians, as well as moral outrage toward Israelis (α = .86). We used a combination of items from Study 1 and additional items for more nuance to measure perceived collective ownership of the land by Palestinians (10 items; α = .96) and guilt toward Palestinians (9 items; α = .93; Al-Qaddoumi, 2024; Schmitt et al., 2000).
Improved and Novel Measures
Perceived Injustice by Israelis Toward Palestinians
Participants rated whether they perceived the treatment of Palestinians by Israelis in the past 100 years as unjust on six items (e.g., “These actions are unjust”; α = .91; Saab et al., 2015).
Perceived Power Imbalance
Participants rated which group has more power overall and over seven resources/actions (e.g., “military power”; α = .86; 1 = Jewish Israelis have all the power, 3 = Both groups have equal power, 5 = Palestinians have all the power).
Hate Toward Israelis
For Study 2, hate toward Israelis was measured directly with three items based on Halperin et al. (2009, 2012; e.g., “I feel hate toward Israelis”; α = .88).
Antisemitism
Antisemitism was measured with nine items (Kempf, 2013; α = .90).
Solidarity Intentions for Palestinians
Participants indicated their solidarity intentions for Palestinians with 20 items. An exploratory factor analysis (EFA) using Maximum Likelihood and oblique rotation (Kaiser-Meyer-Olkin = 0.97; Barlett’s test of sphericity: χ2(190) = 8,930.59, p < .001) suggested two factors (Eigen value > 1; see Table 3). Items with a factor loading greater than 0.4 were considered meaningful (Stevens, 2002). If items loaded on both factors higher than 0.4, they were considered cross-loadings and excluded from the final measure (i.e., four items removed). We also excluded the item “convince others to be an ally for Palestinians” from our main analysis, as it had a different loading in Study 3, and we wanted to maintain consistency in measurements across countries. We labeled the remaining 15 items for our main analysis based on their cost (i.e., the relative effort, risk and/or resources that someone invests; Uluğ et al., 2024): Factor 1 was labeled as low-cost solidarity (4 items, α = .91) since actions expressed more acceptable points of view, had less individual visibility and were not disruptive, thus implied relatively lower effort and risk; Factor 2 was labeled high-cost solidarity (11 items, α = .96) since actions were more disruptive, addressing directly the public and potentially challenging the dominant narrative, thus implied relatively more effort and risk. This classification might also explain why some items cross-loaded: they could be seen as moderate-cost since they might be perceived to be more in line with legal boundaries, but very resource-intensive (Corcoran et al., 2015).
Exploratory Factor Analysis for Solidarity Actions for Palestinians.
However, for transparency, and because we believe that based on contextual differences in the political and repressive climate these deviations in factor loadings between countries might be meaningful (see Kusano et al., 2025, for a critique on measurement invariance for cross-country comparisons), we ran additional analyses for which we based high-cost solidarity on the factor structure that is solely informed by the sample of this study (i.e., included “convince others to be an ally for the Palestinians”; Tables S9–S11 on OSF).
Results
Descriptive Statistics and Statistical Procedure
Similar to Study 1, solidarity intentions for both low-cost (Skewness: 1.20, Kurtosis: 0.16) and high-cost solidarity (Skewness: 2.16, Kurtosis: 3.96) were (extremely) highly skewed, and thus we created a binary variable indicating no to little intention (= 0; mean =< 2; nlow-cost = 264, nhigh-cost = 328) or some intention for solidarity (=1; mean > 2; nlow-cost = 148, nhigh-cost = 84). We ran two hierarchical logistic regressions as our main models, one for low-cost solidarity and one for high-cost solidarity. In the first step, we entered beliefs, followed by emotions, and lastly, attitudes. Again, while some predictors showed moderate to strong intercorrelations (r > .60; see Table 4), multicollinearity diagnostics were acceptable (VIF < 5). We also conducted additional analysis to account for possible suppression effects by removing highly correlated predictors (see Study 1), and mentioned in text whenever we identified any differences in these models.
Descriptives of and Correlations Between Key Constructs of Study 2.
p < .05. ***p < .001.
Predictors of Low-Cost Solidarity for Palestinians
The results of our main hierarchical regression for low-cost solidarity can be found in Table 5, and the results of our alternative models in Tables S3 and S4 on OSF. Control variables explained significantly 21% of the variance of low-cost solidarity, χ2(5, N = 412) = 96.06, p < .001, R2 = .21. Being younger, left-wing, Muslim and having higher self-reported knowledge about the conflict were positively related to low-cost solidarity intentions. Gender was not a significant predictor of low-cost solidarity intentions.
Hierarchical Logistic Regression for Predicting the Occurrence of Low-Cost Solidarity Intentions for Palestinians in Germany.
Note. OR = Odds ratio.
In the first step, beliefs about the conflict increased the explained variance significantly, χ2step(4, N = 412) = 70.20, p < .001, R2 = .34. Among beliefs, perceived injustice and peaceful intentions of Palestinians significantly predicted low-cost solidarity intentions when entered first, but lost predictive power once emotions were entered. This pattern was consistent across models, indicating that these beliefs may primarily predict solidarity indirectly via emotions. Perceived power imbalance was non-significant when first entered in the model, but became significant once emotions were added, across all models. This pattern suggests a potential suppression effect, whereby the inclusion of emotions allowed the unique contribution of perceived power imbalance to be detected. Perceived collective ownership of the land by Palestinians was not a significant predictor, however, when removing the highly correlated predictor perceived injustice, it became a significant predictor before emotions were added (OR = 1.44 [1.16, 1.79], p = .001).
In the second step, emotions explained additional variance to beliefs, χ2step(5, N = 412) = 95.03, p < .001, R2 = .48. Emotions toward Palestinians—sympathy, guilt, and admiration—remained consistently significant predictors across all models, regardless of the order of entry. On the other hand, emotions toward Israelis—moral outrage and hate—remained mostly non-significant throughout the models and entry order; except when highly correlated emotions were excluded (sympathy and admiration), then moral outrage became significant (OR = 1.61 [1.29, 2.02], p < .001). Thus, moral outrage might be a weaker predictor of low-cost solidarity intentions.
In the third step, attitudes did not significantly explain further variances for low-cost solidarity, χ2step(1, N = 412) = .70, p = .402, R2 = .48. While antisemitism did not significantly predict solidarity intentions in the main model or in the alternative model 1, in the alternative model 2, where antisemitism was entered first, it was initially significant (Model 2: OR = 1.34 [1.11, 1.63], p = .003) but lost significance after adding beliefs and emotions. This suggests that antisemitism may function as a predictor for some individuals, but did not have a unique influence on low-cost solidarity intentions when more direct psychological constructs (e.g., specific beliefs, emotions) are considered.
Predictors of High-Cost Solidarity for Palestinians
The results of our main hierarchical regression for high-cost solidarity can be found in Table 6, and the results of our alternative models in Tables S5 and S6 on OSF. Control variables explained significantly 22% of the variance of high-cost solidarity, χ2step(5, N = 412) = 101.65, p < .001; R2 = .22. Being younger, Muslim, and having higher self-reported knowledge were related to more high-cost solidarity intentions. Political orientation and gender did not predict high-cost solidarity.
Hierarchical Logistic Regression for Predicting the Occurrence of High-Cost Solidarity Intentions for Palestinians in Germany.
Note. OR = Odds ratio.
In the first step, beliefs about the conflict significantly predicted more variance of high-cost solidarity, χ2step(4, N = 412) = 49.08, p < .001; R2 = .31. Perceived collective ownership of the land by Palestinians, perceived peaceful intention and power imbalance significantly predicted high-cost solidarity, but became non-significant once emotions were entered, across all models. This might suggest that these beliefs are not as proximal to high-cost solidarity, and may be mediated by emotions. Perceived injustice did not significantly predict high-cost solidarity across all models; however, perceived injustice was initially a predictor for high-cost solidarity when highly correlated predictors (i.e., collective ownership perceptions) were removed (OR = 1.40 [1.07, 1.83], p = .014). Thus, perceived injustice might be a weaker predictor of high-cost solidarity in issue-specific repressive contexts.
In the second step, emotions explained significantly more variance, χ2step(5, N = 412) = 68.19, p < .001; R2 = .42. Guilt toward Palestinians was a robust predictor across models, regardless of entry orders. Sympathy and admiration toward Palestinians remained non-significant for all models and entry orders. Moral outrage and hate toward Israelis also did not significantly predict high-cost solidarity intentions, but hate was significant when emotions were entered first into the model (alternative model 1; OR = 1.35 [1.04, 1.75], p = .023), thus being a weaker predictor of high-cost solidarity under issue-specific repression.
In the third step, antisemitism did not explain further variance for high-cost solidarity for Palestinians, χ2step(1, N = 412) = 1.99, p = .158; R2 = .42. However, when antisemitism was entered first, it was significant (OR = 1.76 [1.39, 2.24], p < .001), remained significant when beliefs were entered (OR = 1.45 [1.10, 1.90], p = .008), and only lost significance once emotions were introduced. Thus, for some individuals, hate toward Israelis and antisemitism might share some variance based on a common belief system or ideology.
Exploratory Analysis
We ran additional analyses including our control variables and additional predictors that might be important and relevant for third-party solidarity; we did not include these predictors in our main model for two reasons: (a) we did not have enough statistical power to add more predictors to the model based on sample size, and (b) these predictors were not relevant to answer our main research questions about which conflict-related beliefs, emotions, and attitudes predict third-party solidarity intentions in issue-specific repressive contexts. First, although perceived risk of solidarity might potentially explain why third-party members differ in their motivations for solidarity compared to individuals in less repressive contexts, perceived risk/repression and objective repression do not always align (e.g., objective repression might be higher than perceived repression; Honari, 2018). Thus, objective repression in democracies may create unique barriers that cannot be explained by perceived risk but still complicate solidarity in these countries, such as adopting certain beliefs that are not part of the dominant narrative (e.g., ownership perceptions of Palestinians). Nevertheless, it is interesting to examine whether perceived risk has a (de)mobilizing effect on third-party solidarity. Our analysis revealed that perceived risk (e.g., stigmatization, losing one’s job, police violence) was not related to both low-cost (OR = 1.18 [0.89, 1.57], p = .241) and high-cost solidarity intentions among Germans (OR = 1.17 [0.82, 1.65], p = .385). Additionally, we examined whether past solidarity engagement for Palestinians predicted third-party solidarity intentions, and found it to be a non-significant predictor, again for both low-cost (OR = 1.33 [0.79, 2.24], p = .283) and high-cost solidarity intentions (OR = 1.25 [0.82, 1.92], p = .299). 1
Discussion
Study 2 replicated the findings of Study 1: perceived peaceful intentions and guilt consistently predicted more low-cost and high-cost solidarity for Palestinians. As in Study 1, hate and antisemitism were no or only weak predictors of third-party solidarity. Different from Study 1, after distinguishing between low-cost and high-cost solidarity, we found that perceived injustice consistently predicted low-cost solidarity intentions but was a weaker predictor for high-cost solidarity intentions. In contrast, collective ownership perceptions were only a strong predictor for high-cost solidarity, while a weaker predictor for low-cost solidarity. Hence, stronger moral beliefs that deviate from the dominant narrative and highlight the severity of the situation seem to be particularly relevant in encouraging people to engage in more costly actions. Perceived power imbalance predicted both types of solidarity intentions. Positive emotions (i.e., sympathy and admiration) only predicted low-cost solidarity and were weak predictors of high-cost solidarity when being the only predictor in the model. Also, moral outrage seems to be a relatively weak predictor for both low-cost and high-cost third-party solidarity. This finding might indicate that under issue-specific repression, these predictors are not powerful enough to motivate high-cost solidarity, but rather only motivate less costly actions. Interestingly, although not part of our research question, perceived risk did not predict people’s intentions for low-cost or high-cost solidarity. This might be potentially explained by the repression not seeming to be relevant to people who anyway were not willing to participate in action, while people who passed this threshold to care about Palestinians and show intentions to act already might be more immune to repression tactics, such as being labeled as antisemitic.
Furthermore, Study 2 showed the importance of differentiating between various solidarity actions regarding cost since the motivators between low-cost and high-cost actions might differ (Odağ et al., 2023). However, we do not know from these two studies whether our results would apply to other democratic, yet issue-specific repressive contexts.
Study 3: The United Kingdom in 2024
We decided to conduct another study to examine whether we can replicate our results in another context that can be considered a democracy, yet issue-specific repressive when it comes to solidarity for Palestine in 2024. We chose the United Kingdom because of its milder form of issue-specific repression compared to Germany (Amnesty International, 2024; Fekete, 2024).
Methods
Participants and Procedure
We again aimed to recruit 500 participants based on our a priori power analysis. We collected data at the end of December 2024 from 493 participants who were representative of the U.K. population in terms of age, gender, and education, through the panel provider Bilendi. We excluded 80 participants due to low quality, and four participants since they had lived in the United Kingdom for less than 5 years (see Study 2 for exclusion criteria).
The final sample had 409 participants (189 women, 220 men; age: M = 48.62, SD = 13.91, range = 18–99; education: no/low-level: n = 9, mid-level (GCSE): n = 127, high-level (A-Levels or higher): n = 271, other: n = 2). Participants were born (n = 377) or lived in the United Kingdom for a long time (n = 32, range: 7–60 years); thus, they were familiar with the (political and historical) context of the United Kingdom. Many were Christian (n = 193) or had no religion (n = 183), followed by Muslims (n = 15), Jews (n = 4), or others (n = 15), with low levels of religiosity (M = 2.57, SD = 1.83). Our sample was politically in the center (1 = left, 7 = right; M = 4.01, SD = 1.28). Most people were neutral toward Israel/Palestine (52.1%; scale 5–7), followed by sympathizers with Palestine (27.6%, scale 8–11) or with Israel (20.3%, scale 1–4); therefore, different from our German samples since more people in the United Kingdom took a side and were overall more sympathetic toward Palestine. Participants indicated having moderate knowledge about the Israel-Palestine conflict (M = 4.28, SD = 1.63; 2 items; r = .74, p < .001). A sensitivity test indicated that we had a power of 0.8 to detect small effects (OR >= 1.36).
Measures
We used the same measures as in Study 2. Participants rated items on a 7-point Likert scale: 1 = completely disagree; 7 = completely agree. We measured perceived injustice toward Palestinians (α = .91), perceived peaceful intentions of Palestinians (α = .74), perceived collective ownership of the land by Palestinians (α = .96), and perceived power imbalance (α = .92). For emotions toward Palestinians, we measured sympathy (α = .88), admiration (α = .90), and guilt (α = .94). For emotions toward Israelis, we measured moral outrage (α = .79) and hate (α = .85). Antisemitism was measured with eight items, similar to Study 2 but excluding the item “The Jews are irreconcilable and harm Germany when they repeatedly point the finger at the Holocaust” since it was not applicable to the U.K. context (Kempf, 2013; α = .88). For solidarity intentions toward Palestinians, we again measured low-cost solidarity and high-cost solidarity with the same items, with a small change from merging the items “write an article under my real name” and “writing an article under a pseudonym” from Study 2 into one item “write an article” since participants seemed to not differentiate between these two actions (i.e., both loaded really highly on one factor in Study 2). An EFA yielded overall similar factor loadings to those in Study 2 for our items (see additional analysis on OSF, Table S12). In contrast to Study 2, the item “convince others to be an ally” loaded on low-cost solidarity for the U.K. sample. Furthermore, in contrast to Study 2, in which some items had cross-loadings, the items “write a letter to the U.K. government to request to stop the support for Israel as long as Palestinians are oppressed” loaded on low-cost solidarity, and “demonstrate,” and “contact the media to request coverage of the situation of Palestinians” loaded on high-cost solidarity. These differences may be explained by cultural and socio-political differences in risk and effort (i.e., their overall cost) associated with those actions. However, for measurement consistency and for better interpretability, we only included the items that also loaded clearly on one factor in Study 2 for our main analysis. The final measure for low-cost solidarity contained four items (α = .94) and high-cost solidarity included 10 items (α = .97).
Nevertheless, we believe that it is relevant and interesting to observe the different factor loadings between samples since they might shed light on the sociopolitical differences for solidarity intentions between countries. Thus, we, again, ran analyses that predicted solidarity intentions based on the specific EFA of the sample (i.e., including items in the solidarity measure if they loaded clearly on one factor, even if that was not the case in Study 2; low-cost: 7 items, α = .95; high-cost: 12 items, α = .98), and reported these analyses in the Supplemental Materials on OSF (Tables S19–S24).
Results
Descriptive Statistics and Statistical Procedure
Solidarity intentions for low-cost solidarity were only slightly skewed for our U.K. sample (skewness 0.89, kurtosis −0.67); however, they were still extremely skewed for the high-cost solidarity intentions (skewness 1.75, kurtosis 2.04). For consistency between the studies, we again created a binary variable for each indicating no to little intention (=0; scores =< 2; nlow-cost = 235, nhigh-cost = 300) or some intention for solidarity (=1; scores > 2; nlow-cost = 174, nhigh-cost = 109). Again, we ran two hierarchical logistic regressions as our main models, one for low-cost solidarity and one for high-cost solidarity. In the first step, we entered beliefs, followed by emotions, and lastly, attitudes. We also ran two alternative models, which can be found on OSF. Similar to the previous studies, while some predictors showed moderate to strong intercorrelations (r > .6; see Table 7), there was no problem with multicollinearity (VIF < 5). However, given the high correlations among some predictors, we ran additional models in which we removed the highly correlating predictors (see Study 1) and reported in text when there were any significant differences with this approach.
Descriptives of and Correlations Between Key Constructs of Study 3.
p < .05. ***p < .001.
Predictors of Low-Cost Solidarity for Palestinians
The results of our main hierarchical regression for low-cost solidarity can be found in Table 8, and the results of our alternative models in Tables S13 and S14 on OSF. Control variables significantly explained 18% of the variance of low-cost solidarity, χ2step(5, N = 409) = 82.81, p < .001; R2 = .18. Being younger, a woman, left-wing, and having higher self-reported knowledge were associated with higher low-cost solidarity intentions. Being Muslim did not predict low-cost solidarity intentions.
Hierarchical Logistic Regression for Predicting the Occurrence of Low-Cost Solidarity Intentions for Palestinians in the United Kingdom.
Note. OR = Odds ratio.
In the first step, beliefs significantly increased the explained variance for low-cost solidarity, χ2step(4, N = 409) = 115.99, p < .001; R2 = .39. Perceived injustice was a significant predictor across models and order entry. Perceived peaceful intentions and collective ownership perceptions were significant predictors when first entered into the model, but became non-significant once emotions were added. In contrast, perceived power imbalance did not predict low-cost solidarity intentions across models.
In the second step, emotions explained additional variance to beliefs, χ2step(5, N = 409) = 93.91, p < .001; R2 = .51. Guilt toward Palestinians, moral outrage and hate toward Israelis consistently predicted low-cost solidarity intentions, across models. In contrast, sympathy was not a significant predictor across models. Similarly, in most models, admiration was not a significant predictor, however became significant when emotions were entered first into the model (OR = 1.48 [1.05, 2.10], p = .027), or once highly correlated emotions (sympathy, guilt, moral outrage) were removed (OR = 2.30 [1.62, 3.27], p < .001). This suggests that its effect is weaker and is likely to be suppressed in our main model.
In the last step, antisemitism did not explain significantly more variance of low-cost solidarity intentions, χ2step(1, N = 409) = 1.43, p = .231; R2 = .51. However, in our alternative model 2, when antisemitism was added first, it was a significant predictor but became non-significant once emotions were added. Again, this might potentially be caused by shared variance for some participants between hate toward Israelis and antisemitism.
Predictors of High-Cost Solidarity for Palestinians
The results of our main hierarchical regression for high-cost solidarity can be found in Table 9, and the results of our alternative models in Tables S15 and S16 on OSF. Control variables explained significantly 19% of the variance of high-cost solidarity, χ2step(5, N = 409) = 88.22, p < .001, R2 = .19. Being younger, left-wing, and having higher self-reported knowledge were associated with higher solidarity intentions. Being Muslim and gender did not predict high-cost solidarity.
Hierarchical Logistic Regression for Predicting the Occurrence of High-Cost Solidarity Intentions for Palestinians in the United Kingdom.
Note. OR = Odds ratio.
In the first step, beliefs significantly explained additional variance for high-cost solidarity, χ2step(4, N = 409) = 71.84, p < .001, R2 = .32. Perceived peaceful intention was the only robust significant predictor of high-cost solidarity among beliefs, across models and regardless of entry order. Perceived collective ownership was a significant predictor until emotions were added; again, this pattern was observable across all models. Thus, emotions might mediate the relationship between ownership perceptions and solidarity. Perceived injustice remained non-significant across all models and entry orders; however, perceived injustice became a significant predictor when highly correlating predictors were removed (i.e., collective ownership perceptions), suggesting that it might be a weaker predictor. Power imbalance, although almost significant in our main model, remained non-significant across all models.
In step 2, emotions significantly explained more variance, χ2step(5, N = 409) = 68.13, p < .001, R2 = .43. Only guilt toward Palestinians and hate toward Israelis consistently predicted high-cost solidarity across all models and regardless of entry order. Hate toward Israelis was a significant predictor when initially entered into the model, but became non-significant once antisemitism was added. Thus, for some people, there might be shared variance between these two constructs. Sympathy and admiration toward Palestinians, and moral outrage toward Israelis, did not predict high-cost solidarity in the main model; however, when we accounted for potential suppression effects by removing highly correlated emotions, admiration and moral outrage were significant predictors. Thus, they might be relevant for high-cost solidarity, but only to a minor extent.
In step 3, antisemitism did not significantly explain further variance of high-cost solidarity for Palestinians, χ2step(1, N = 409) = .05, p = .826, R2 = .43. However, when antisemitism was included first and as the only predictor in the model, it was a significant predictor for high-cost solidarity until emotions were added. Again, pointing to some overlap of hate toward Israelis and antisemitism for some people.
Exploratory Analysis
Again, we conducted two exploratory analyses, including our control variables, and perceived risk or past solidarity, respectively, to predict third-party solidarity intentions in democracies facing issue-specific repression. In contrast to Study 2, logistic regression showed that perceived risk was associated with a higher likelihood to indicate low-cost (OR = 1.45 [1.23, 1.70], p < .001) and high-cost solidarity intentions (OR = 1.50 [1.25, 1.81], p < .001). Thus, perceiving risk, and potentially repression from official institutions and society, seems to be related to third-party members’ mobilizing. Again different to Study 2, past solidarity was also a positive predictor of third-party low-cost solidarity (OR = 2.95 [2.06, 4.23], p < .001) and high-cost solidarity (OR = 2.61 [1.99, 3.44], p < .001).
General Discussion
In this contribution, we examine which beliefs, emotions, and attitudes predict third-party solidarity toward oppressed groups involved in intergroup conflicts, within democracies characterized by issue-specific repression. Overall, our studies showed that some previous predictors for solidarity—perceived peaceful intention of the oppressed, perceived injustice, moral outrage, guilt, sympathy, admiration—are also (weak) predictors for solidarity actions in democracies with different levels of issue-specific repressions (i.e., Germany, United Kingdom). Furthermore, we expanded the solidarity literature by demonstrating that beliefs about the conflict that deviate from the dominant narrative, such as collective ownership of land, can be relevant predictors of third-party solidarity. Perceived power imbalance was also a predictor of third-party solidarity; however, we could not identify a clear pattern across countries. Furthermore, hate might be a weak to strong predictor of both low- and high-cost solidarity intentions depending on the context. We also highlight that antisemitism, which is often discussed as a central motivator for Palestine solidarity, was not a significant predictor in our full models, and only became significant when it was the only predictor in the model. We summarize the results of our three studies in Table 10.
Overview of Results From Our Main Model of All Studies for All Predictors of Different Solidarity Intentions.
Note. GER = Germany; UK = United Kingdom;— = not measured;
Our first aim was to examine whether established predictors of solidarity in relatively less repressive contexts can predict third-party solidarity in democracies where solidarity is more repressed. Similar to less repressive contexts (Bruneau et al., 2017), perceptions about the oppressed group are relevant for third-party solidarity in issue-specific repressive contexts. Specifically, perceiving the oppressed to have peaceful intentions was a consistent predictor of low-cost and high-cost solidarity. Having peaceful intentions may indicate that a group is a victim, as they are involuntarily participating in the conflict (Dahlum et al., 2022). This belief deviates from dominant narratives that might portray the oppressed group as an aggressor (e.g., terrorist; Abu-Laban & Bakan, 2021). People who reject these dominant narratives may be better equipped to recognize and resist repressive tactics (e.g., legitimizing oppression) and may find it easier to allocate support when a clear victim is perceived.
Although moral outrage as well as perceived injustice predicted solidarity, we could not consistently replicate the usually strong relationships between these constructs across different contexts (Agostini & van Zomeren, 2021). Thus, in democratic contexts with issue-specific repression, other conflict-related beliefs and moral emotions, such as guilt, seem to be stronger predictors of solidarity. Injustice perceptions and moral outrage were overall high toward Israelis and Palestinians. People might not view groups in conflict as pure victims or perpetrators, but rather classify them as holding both aspects (Demirel, 2023), which may blunt the motivational force of outrage or injustice alone. Therefore, under repression, where dominant narratives justify injustice toward the oppressed, these perceptions may not be sufficient to foster solidarity. Thus, although injustice perceptions undoubtedly are necessary to become active, beliefs that deviate from dominant narratives seem to be more determinant for motivating third parties’ high-cost solidarity.
Compared to previous research, which yielded mixed results on the role of guilt in solidarity (Iyer et al., 2007), our results showed that guilt was one of the strongest and most consistent predictors for different solidarity intentions of third parties in contexts with issue-specific repressions. Guilt may reflect a desire to avoid complicity in perceived injustices, especially when one’s own government is involved, or stem from awareness of historical responsibilities. Future work should explore which types of guilt (e.g., guilt of inaction or historical guilt) most strongly motivate third-party solidarity.
In line with recent developments in the solidarity literature (Thomas et al., 2021), we also tested positive emotions as predictors for third-party solidarity. Contrasting previous research (Ditrich & Sassenberg, 2024; Landmann et al., 2025), we found that positive emotions, such as sympathy and admiration, played only a minor role in third-party solidarity when considering other beliefs and emotions; except for low-cost solidarity in Germany, for which they were strong predictors. Therefore, in the face of issue-specific repression, positive emotions might not be strong enough to motivate solidarity among third parties.
Our second aim was to extend the solidarity literature by investigating novel predictors derived from the intergroup conflict literature, including collective ownership perceptions, perceived power imbalance, and hate. Our results underscore the importance of conflict beliefs not only for conflicting groups’ political actions but also for third-party members’ solidarity. For instance, we show that collective ownership perceptions play a crucial role for third parties. Psychological research has only recently considered the territorial aspects of conflict (Storz et al., 2022; Verkuyten & Martinovic, 2017), although many intergroup conflicts are rooted in territorial disputes and claims about indigeneity (HIIK, 2023). Thus, it makes sense that the perception of the rightful owner of the land has an impact on determining who to support. As pointed out, particularly under issue-specific repression, beliefs that deviate from the dominant narrative may be required to motivate third parties to show solidarity. Moreover, in some contexts, the perceived power imbalance between conflicting groups determined third-party solidarity intentions, which aligns with research that shows increased support when a group is perceived as the underdog (Bruneau et al., 2017). Interestingly, in the milder repressive context for Palestine solidarity (i.e., the United Kingdom), this perception was not powerful enough to motivate high-cost solidarity, whereas in the more repressive context (i.e., Germany), it was a predictor of both low-cost and high-cost solidarity. Future research will need to investigate what might be the underlying explanations for these differences in patterns.
In contrast, hate toward Israelis played mostly a negligible role, and only in the United Kingdom a stronger role. Research showed that over time and under low control, moral outrage can turn into hate (Miller & Jasper, 2022), possibly explaining the U.K. sample’s response after more than a year of genocide. Greater access to information about the treatment of Palestinians, thus higher knowledge of the severity of injustice, in the United Kingdom might have amplified higher levels of moral outrage and hate. However, when issue-specific repression increases (i.e., Germany), it seems that this link becomes weaker.
Our third aim was to address the repression of solidarity in democratic contexts. While issue-specific repressions of solidarity in democracies occurred in the past (e.g., indigenous rights in Canada, or anti-apartheid movements in the United States; Lambertson, 2004; Minter & Hill, 2008), and became more apparent in recent years (e.g., for Kurdish rights; Kaufer, 2019), psychological research about this phenomenon remains scarce. We observed in our data that people are generally less willing to be active in these contexts. Overall, people are more willing to engage in solidarity when it is low-cost than when it is high-cost. Thus, under issue-specific repression, people might find ways to be supportive, yet work around the costs associated with the repression of solidarity. Future research will have to investigate whether and why participants strategically avoid certain actions (e.g., personal cost, sustainability of solidarity engagement). Interestingly, positive emotions toward Palestinians seemed more relevant in Germany, while negative emotions toward Israelis were more important in the United Kingdom in predicting particularly low-cost solidarity actions. Again, these differences might be caused by differences in access to information about the conflict. In Germany, the suffering of Palestinians is partly covered, however justified (e.g., to fight terrorism; Kempf, 2014), while in the United Kingdom, there might be more information available that does not justify the actions of Israel. Thus, the differences in exposure to information and narratives might also shape which emotions we (perceive to be allowed to) feel, and what motivates us to act. Additionally, exploratory analysis revealed that when people perceive risk, and thus potentially issue-specific repression, it is either not or a positive predictor of third-party solidarity; similar to the mixed findings of research that explored whether perceived risk and repression increase collective action (Ayanian et al., 2025; Uysal et al., 2022). Research will need to investigate whether third-party members might be mobilized by perceived repression, or if people who are willing to engage in solidarity are just more aware of the risks.
More generally, our data underscores previous work that points out that motivations are shaped by the cost associated with an action (Corcoran et al., 2015). Moreover, context matters when examining solidarity, since in our data some of the same actions loaded on different factors in Germany and the United Kingdom, representing the different realities of being active in terms of common political practices and levels of repression. When we considered these differences in factor loadings in our analyses, we found a slightly different, clearer pattern of which constructs predicted low-cost versus high-cost solidarity (see OSF, Table S25). As researchers, we should thus reflect whether accounting for differences in our outcome measures based on varying socio-political climates is better suited to examine political phenomena, compared to approaches that use the same items to measure a construct across countries (cf. Kusano et al., 2025 for critique on country comparison methods).
Lastly, antisemitism was not a significant predictor of solidarity when including other predictors across our contexts and (representative) samples; however, it was a weak predictor when entered first and as the only predictor in the model. This finding contrasts with the misperception and misrepresentation of Palestine solidarity within the media and society, which highlight antisemitism as the core of the movement and as the main reason for its repression. The weak relationship between antisemitism and solidarity aligns with previous research showing that while antisemitic individuals may engage in Palestine solidarity, the vast majority have other reasons, such as human rights concerns (Beattie, 2017; Kempf, 2015). Furthermore, potentially caused by the constant conflation of these terms in public spaces, some people might struggle to understand the distinction between Israelis and Jews, which might increase “carry over” effects, where witnessing the injustice faced by Palestinians impacts the perceptions of Jews more generally (Ozer et al., 2025). Thus, these findings alert us to be cautious with conflating antisemitism with anti-Zionism to minimize potential harms. However, overall, based on the small associations in our data, our findings align with theoretical and empirical papers that disentangle the critique of Israel from antisemitism (Gordon, 2024; Kempf, 2015).
Limitations and Future Directions
We would like to acknowledge that our studies come with some limitations. First, some of our measurements have some limitations. We acknowledge the complexity of belief systems that often include perceptions not only for one group in conflict but for both groups. For instance, people might see one, both, or neither of the groups in conflict as rightful owners of the land (Nooitgedagt et al., 2023). These belief patterns might be more apt in capturing the complexity of our mental models toward conflict, and future research should address these nuances. In addition, our measure only focused on solidarity intentions. Research indicates that intentions and behavior do not always align, and intention measures typically yield higher scores than actual behavior measures (Sheeran & Webb, 2016; Uluğ et al., 2022).
We also acknowledge that the repression of solidarity with Palestine in countries in the Global North creates a unique context that we have not witnessed to a similar extent for other issues in recent years; thus, there might be a limit in the generalizability of our results for solidarity toward other conflict contexts (Simons et al., 2017). However, this unique case has far-reaching implications on understanding how and why people resist shrinking civic spaces that threaten democratic foundations and should apply to similar issue-specific repressive contexts across the globe (e.g., United States). We also expect that our findings might apply to solidarity toward other conflicts, such as the fight for liberation of Kurdish people, which also faces issue-specific repression in democratic countries (Sentas, 2016). We, furthermore, believe it contributes to our understanding of what might have motivated people in past struggles, such as the movement against Apartheid South Africa, which was also severely repressed (e.g., in the United States; Minter & Hill, 2008).
Conclusion
With the emerging field of solidarity, the question arises: how do individuals in various contexts become motivated to take action by different factors? We currently know most about the motivations of solidarity movements targeting social injustices within their own societies or third-party members showing solidarity in democratic, less repressive contexts. Our studies address this gap in the literature by expanding the scope of solidarity to other, less frequently focused, types of actors: third-party members showing solidarity in issue-specific repressive contexts. Our studies show that some of the motivators for these individuals overlap with the motivators that are traditionally found in the solidarity literature (perceived peaceful intentions, perceived injustice, moral outrage). However, we extend the literature and highlight that specific conflict beliefs (collective ownership) and emotions (guilt) are relevant to consider since they are stronger and more consistent predictors compared to traditionally focused predictors. Furthermore, we highlight that perceived power imbalance, positive emotions (sympathy, admiration), hate, and antisemitism play only a minor role in third-party solidarity for Palestinians. Our results underscore the distinctiveness of solidarity under repression from traditionally researched solidarity and highlight the importance of examining these forms of solidarity, and their associated costs, separately. In light of emerging conflicts around the world rooted in historical and structural injustices, research that understands the processes behind solidarity for these contexts seems more relevant than ever to create an avenue for sustainable peacebuilding.
Footnotes
Acknowledgements
We sincerely thank Prof. Dr. Jürgen Maes (University of the Bundeswehr Munich) and Andreas Würth for sharing their dataset with us for Study 1, and Dr. Rim Saab for her invaluable feedback on previous versions of this article.
Ethical Considerations
For all our studies, we received ethical approval from the University of Sussex (Study 1: ER/JS2146/1, Study 2: ER/JS2146/7, Study 3: ER/JS2146/11).
Consent to Participate
All participants provided written informed consent prior to participating.
Funding
The authors received no financ
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
