Evaluating the impact of the EU’s normative message of the “EU as a model” on external public perceptions: an experimental study in Israel

ABSTRACT

This article analyses whether the EU’s normative idea, as embedded in the idea of the “EU as a model”, influences external public perceptions of its normative message; in other words, does exposure to the story of how the EU emerged from the ashes of World War II to become a community of peace and security impact public views of the EU as a normative power? To examine these questions, a population-based survey experiment was conducted in Israel – a country with contradictory relations with the EU. The findings are clear and straightforward: exposure to the “EU as model” message increased recognition of the normative actorness of the EU. This impact holds when controlling and interacting with the exposure to various cultural filters which are key factors in shaping views of the EU like political attitudes. The implications of these findings for EU studies and public diplomacy as well as the potential contribution of the approach proposed in this article are discussed.

KEYWORDS:

• External public perceptions
• public diplomacy
• Israel
• survey experiment
• EU

Introduction

The nature of EU power in the international arena has been subject to a vibrant debate and research over the last two decades (Beck and Grande 2007, Hettne and Söderbaum 2005, Manners 2002, Tonra and Christiansen 2018). As part of these research efforts, external perceptions of the EU have been studied to discover whether the EU is perceived from the outside as an efficient global actor, a model to be imitated, or a self-interested hypocrite (Speyer et al. 2021). These studies developed into a subfield of external perceptions of the EU which combines insights from the fields of international relations and EU studies (Chaban et al. 2015, Chaban et al. 2021, Chaban and Holland 2019, Chaban and Pardo 2018, Chaban, and Elgström 2021a, Isani and Schlipphak 2017, Lucarelli 2014). Most studies focused on the elites, but a few also explored how the EU is perceived by the public as part of the growing attention to EU public diplomacy (Fanoulis and Revelas 2023). Much of the analysis of public diplomacy has analysed the processes of public diplomacy by focusing on elites (Mai’a and La Porte 2017): for example, the Chinese students in the Jean Monnet Programme (Yang 2015) or decision makers in Ukraine (Chaban and Elgström 2021). The impact of public diplomacy on the public has rarely been studied (Song and Fanoulis 2023); nor, to the best of our knowledge, the idea of the “EU as a model”, influences laypeople’s external perceptions of the EU.

The European Commission has, nonetheless, invested efforts in analysing how the EU and EU’s policies are perceived in key countries and by its neighbours via cross-sectional surveys (EU Neighbours South 2021, European Commission 2015). These reports and other studies of EU external perceptions (Chaban et al. 2015, Chaban et al. 2021, Chaban and Holland 2019, Chaban, and Elgström, 2021a, Isani and Schlipphak 2017, Keuleers 2015, Lucarelli 2014) provide rich descriptions of the ways the EU is viewed in various parts of the world while offering various explanations for these perspectives. Thus far, all studies of external public perceptions of the EU have been cross-sectional surveys measuring public perceptions at one point in time and without any experimental element. As such, they can describe how the EU is viewed from the outside and the variables related to these views; they cannot, however, determine the impact of the EU’s normative message on the ways the EU is perceived.

This article uses an experimental approach to examine whether the EU’s normative message, as embedded in the development of the EU and its stated values, influences external public perceptions of EU actorness. Its theoretical point of departure is the notion that the normative component of the EU is a model that others can follow (Manners 2006). The EU was constructed as a successful and prosperous peace community from the ashes of World War II (Diez 2005) – a model which can be followed by other conflictual regions. However, it is not clear from previous studies that used cross-sectional surveys whether the external perceptions of the public are influenced by this underlying normative message of the “EU as a model”; therefore, an alternative research design that inspects influence is required. This article’s methodological contribution is in its adoption of the experimental approach that was recently used to study the influence of public diplomacy on laypeople (Hellmann and Oppermann 2022).

The article uses the case of Israel – a country with ambivalent perceptions of Europe and the EU (Pardo 2015, Pardo and Zahavi 2022). Despite the ongoing involvement of the EU in various aspects of the Israeli–Palestinian conflict, there is a gap between the EU’s normative ambitions and its actual success in moderating this conflict (Harpaz and Shamis 2010, Müller 2019). In fact, there is a contradiction between the EU’s self-representation as a normative power and the way it is perceived by Israelis. Harpaz and Shamis (2010) suggested that the normative message of the “EU as a model” might be counter-productive and prompt a backlash against its involvement (Harpaz and Shamis 2010). This is especially relevant as the dominant political right in Israel opposes the view of the EU as a paradigm for peacebuilding (Müller 2019) and promotes alliances with Eurosceptic forces (Filc and Pardo 2021). These conditions make Israel a suitable case for exploring the influence of the “EU as a model” on the public while considering the differences between distinct segments of Israeli society.

To inspect the influence of the normative message of the “EU as a model” on Israeli perceptions, we conducted a population survey experiment (N = 1286). Participants were exposed to the message that the EU is a model for peace, as recognised by winning the Nobel Prize in 2012 and by EU core values. We then analysed whether exposure to this message influenced public perceptions of the EU as a normative power. We also explored the role of cultural filters in response to the normative message by examining the possible influence of sociodemographic characteristics, levels of religiosity, political orientation, chauvinism and EU knowledge. This allowed us to assess the influence of the “EU as a model” on different sectors of society as the EU is a politicised issue (Chaban and Elgström 2020, 2021a), like other international issues which are politicised as part of internal debates (Bearce and Jolliff Scott 2019, De Vries et al. 2021).

The article proceeds as follows. First, we discuss the idea of the “EU as a model” as part of the Normative Power Europe (NPE) framework and then review the subfield of external perceptions of the EU and how it is related to NPE. Next, we emphasise the role of cultural filters and the “power of the local” in shaping these perceptions, briefly presenting our use of the Israeli case. Based on this review, we set out our expectations from the survey experiment and the study rationale before going on to explain the findings. We conclude the paper with a discussion of the theoretical and practical implications of our findings as well as the potential and the limitations of the approach adopted here for the research fields of EU studies and public diplomacy.

NPE, public diplomacy and the EU as a model

Scholars have long debated whether the EU is an emerging global actor with an imperialist agenda, a normative power that uses common European values such as human rights, or just an economic power lacking other components of hard power such as the military (Beck and Grande 2007, Hettne and Söderbaum 2005, Manners 2002, Tonra and Christiansen 2018, Zielonka 2008). Of the various perspectives on the nature of the EU in the international arena, we focus on the NPE theory. Normative power is a concept designed by Manners in the 1990s (2002, 235-236) to consider historical developments. It is based on ideas of ideological power that were used as early as the 1960s and 1970s.

Normative power is, in Manners’ words, “the ability of an actor to shape the perception of what is normal and proper, without using force but by attractive existence of the actor which makes others cooperate with it and adapt it as a model” (2006a, p. 168). Normative power can be identified by the effect it has on what is considered appropriate behaviour by other actors (Manners, 2006a). Manners changed the traditional discussion framework of international relations by using a realist approach which offered a conceptual alternative to researchers in the field of European integration and the power of the EU beyond the discussion of military power, civilian power, economic power and soft power (Bull 1982; Carr 1962; Duchêne 1972). The theoretical development of normative power responded to changes in the distribution of global power and fundamental changes in the structure of the international arena that took place in the last few decades, which necessitated a redefinition of the roles of the actors and a theoretical conceptualisation of how they operate in the new political reality.

It is interesting to note the concept of Ethical Power Europe, which emphasises what it does (rather than what it is) and is, in this regard, a more proactive term (Aggestam 2008). The concept of the Ethical Power Europe and the EU is guided by an assumption regarding the role of Europe, which stems from its self-image as a cosmopolitan power characterised by the global influence and the moral quality of the values, norms and principles that guide its policies (Nunes, 2011). Ethical power scholars claim that the EU has become a significant force in the international arena thanks to the leadership of international/multilateral institutions within which it outlines ethical and legitimate political action (Aggestam 2008), especially given Europe’s perception of itself as a “Force for Good” (Pace, 2008). Thus, ethical power is based on international cosmo-political norms, while normative power is the product of all the interests of the member states of the EU and the supranational interests together with the global and cosmo-political norms which bring joint political action (Whitman 2013).

Despite its centrality in the field of European Studies and international relations, normative power theory has many critics. Diez (2013) argued that the EU’s policy in various areas affects policies external to it due not to its normative power but to it to its hegemony in certain areas. In doing so, Diez criticises the paucity of both empirical proofs of the theory of normative power and empirical research in the field, which leads to the excessive centrality of the theory of normative power (Diez 2013). Others have joined Diez, criticising the universal norms of the normative power theory as hidden Eurocentrism, soft imperialism or hegemony (Janusch 2016).

Some have also claimed that the normative power argument is apologetic and uncritical (Sjursen 2006). Realist scholars tend to criticise it as being liberal, idealistic and reductionist (Hyde-Price 2006). In addition, there are claims that the literature on the EU as a normative power is ambiguous about the role the EU is considered to play. This critique sometimes seems to suggest that the EU not only interprets and/or writes itself into a given normative context or order but also challenges this order and its norms. The EU’s “normative” dimension is also linked to the idea of the EU as a “novel” or “different” actor, which would suggest that it breaks with the established normative order (Sjursen 2006). Post-colonial researchers have often described normative power as a practice by which the EU portrays itself as a force for good while depicting other actors as inferior, thereby disempowering them rhetorically. Other critics have seen the EU’s embrace of normative power theory as a process of self-identity building by the EU itself (Scheipers and Sicurelli 2008), thus, an internal and not purely external policy. Some scholars have also challenged the theory using empirical cases, demonstrating the many aspects and cases in which the theory does not meet reality (Gordon and Pardo 2015, Wagnsson and Hellman 2018).

Examining the EU’s actions as a global actor through the lenses of norms and the theoretical framework of normative power remains relevant due to current developments in the European Neighbourhood and the global arena (Aydın-Düzgit and Noutcheva 2022, Romanova 2016). It is especially relevant as many EU practitioners use the normative component as part of the EU Neighbourhood Policy (Keukeleire et al. 2021), particularly in Israel (Pardo and Zahavi 2022). This question has become all the more significant since the outbreak of the war in Ukraine (Kögler 2023, Vajsa et al. 2022) which has fundamentally changed the international environment of the EU (Bosse 2022). Although the question of whether the EU is indeed a normative power has been discussed vigorously for more than 20 years (Diez 2005, Larsen 2020, Whitman 2013), our focus here is not on whether NPE reflects empirical reality or not. Instead, this article explores whether the “EU as a model” assumption, which is part of the EU’s normative message, is accepted by the public in a third country. We follow Diez’s (2005) suggestion and focus on how the EU was discursively constructed as a normative power and the implications of this discourse. The constructed discourse of the normative component reflects an overall narrative or strategic narrative about the EU integration project. Strategic narratives construct a shared meaning that helps to shape the behaviour of domestic and international actors, which is a growing research topic in the field of international relations (Miskimmon et al. 2013). Shared meaning relates not only to the present time but also to the past. In other words, shared meaning is rooted in the past and explains development from the past to the future. While the EU is connected to various strategic narratives (Manners and Murray 2016), this article focuses on just one narrative: the appealing story of how the EU emerged from the ashes of World War II to become a community of peace and security.

A fundamental aspect of NPE is that the story of EU integration, alongside its founding principles, such as liberty, democracy and respect for human rights, is central to justifications for the EU both within its borders and without. The EU’s experience of institutional cooperation and the peaceful resolution of conflicts that ruled Europe for centuries are a source of inspiration for peace, democracy and socioeconomic development in many parts of the world (Chaban et al. 2017). From this point of view, NPE is part of the discourse that sees the EU as a model for a peaceful community against a war-torn past that can inspire other conflictual regions. An example of this message can be found in the 2020 speech of Josep Borrell, the high representative of the Union for Foreign Affairs and Security Policy, marking 75 years since the end of the war:

The two World Wars left Europe devastated and hungry. But out of the rubbles, the European Union was born. The European Union’s founders imagined a continent that instead of exporting war would promote peace, democracy and human development within its borders and around the world. (EEAS 2020)

A symbolic representation of this message can be found in the EU winning the Nobel Prize in 2012 for the promotion of peaceful relations between its member states. This is, of course, not the only narrative that is related to this achievement (Manners and Murray, 2016).

The notion of being a model is a fundamental aspect of EU public diplomacy. Public diplomacy refers to the ability to win through various means, “the hearts and minds of people around the world” (Gilboa 2008, p. 55). It is aimed at influencing foreign audiences to, in turn, influence their own governments. EU public diplomacy has expanded in recent years to advance EU goals among member states as well as third countries and other international actors (Fanoulis and Revelas 2023, Song and Fanoulis 2023). While EU public diplomacy promotes, overall, normatively positive images of the EU, there is a special role for the “EU as a model” as part of this message. The EU is

both a unique historical peace experiment as well as a major power in a multipolar international order, a prosperous Union bound to promote peace and security worldwide in the medium and long terms … [It] might inspire similar moves in other regions of the world.¹ (Fanoulis and Revelas 2023, p.7)

The normative power of the EU is one of the assets used by the EU when intervening and mediating in conflict situations, modelling itself as a “force for good”. This self-description is used as a rhetorical and symbolic tool for influencing conformation with the values, norms and ideals promoted by EU actors (Diez and Pace 2011).

Detailed analysis of the actual influence of the EU’s normative components in conflict resolutions across various cases is mixed (Diez and Pace 2011, Vasilyan 2014). In the specific case of Israel/Palestine, despite the EU’s lasting involvement in various aspects of conflict resolution, there is a gap between its normative ambitions and its actual success (Harpaz and Shamis 2010, Müller 2019, Pardo 2015). While numerous studies have analysed the EU’s normative component in international relations and in conflict resolution (e.g. Beck and Grande 2007, Diez and Pace 2011, Hettne and Söderbaum 2005, Manners, 2002, Zielonka, 2008), the current article looks only at external perceptions of the EU and the extent to which the “EU as a model” can actually change public views of the EU.

NPE and external perceptions of the EU

Various studies have examined the EU’s message and how it is perceived by the receiver (especially the elites) outside the EU (Chaban et al. 2013, Pardo 2015, Chaban and Kelly 2017, Miskimmon and O’Loughlin 2019, Speyer et al. 2021). The aspect of normative power is connected to the study of external perceptions since the EU’s ability to play a normative power role depends on how it is perceived by other actors (Lucarelli and Fioramonti 2009, Sicurelli 2015). External perceptions enable the exploration of whether the EU is indeed recognised as a “better power” that promotes positive values and collective norms in the global arena and highlights interactions with the outside as part of the formation of an EU/European identity. According to Speyer et al. (2021), this line of research facilitates a prescriptive view that has typically been overlooked by EU-centric scholarship. There is, in addition to growing attention in the field of international relations to public opinion and to the ways in which international issues are part of domestic debates (Bearce and Jolliff Scott 2019, De Vries et al. 2021).

External perceptions of the EU have been studied from a global and comparative perspective via different methodological approaches, sites and levels of analysis (Speyer et al. 2021). Most studies focused on elites (e.g. Bengtsson and Elgström 2012, Chaban et al. 2013, Chaban and Kelly 2017), while a few explored external perceptions of the EU beyond the elites and looked at the public (Chaban et al. 2015, Chaban et al. 2021, Chaban and Holland 2019, Isani and Schlipphak 2017). Public opinions about the EU have become more important as the EU developed the means and tools for public diplomacy (Speyer et al. 2021). Several large-N studies of public opinion conducted in recent years have offered an overview of how the public in different places across the globe view the EU (Chaban et al. 2015; Chaban and Holland 2019). Studies have also offered various explanations of the factors that shape attitudes towards the EU, such as individual characteristics, religiosity or the ties between the EU and one’s country (Isani and Schlipphak 2017, Keuleers, 2015, Schlipphak 2013). Despite these rich descriptions of EU external views, the question of how the “EU as a model” message changes public opinion has not yet been analysed.

External perceptions of the EU, cultural filter and the “power of the local”

Studies of external perceptions of the EU have emphasised the role of cultural filters, namely, the local norms and values that should be considered as part of interactions with the EU as a normative actor (e.g. Gordon and Pardo 2015). Cultural filters reflect how local norms and values should be considered when analysing the interactions between the EU as an international actor and different actors from the world community. Manners explained: “The cultural filter is based on the interplay between the construction of knowledge and the creation of social and political identity by the subjects of norm diffusion” (2002, p. 245). Many studies have used the concept of cultural filters to explore how the EU is viewed and how EU norms are diffused (e.g. Chaban and Pardo, 2018; Mišík 2019). These studies point to the power of local cultural filters as a response to EU attempts to promote norms. Gordon and Pardo claimed that the local “is an extremely complex arena often informed by contradictory ideologies, cultural filters and identities and by ongoing power struggles among different domestic actors” (2015, p. 424).

Studies of external perceptions of the EU provide support for the function of cultural filters in explaining different views of the EU among different segments of the public. Findings have shown that the cultural filter is a key factor in shaping public perceptions of the EU as a normative power. A study in Turkey that used focus groups found historically based scepticism towards EU normative power in the form of the continuation of the Sèvres syndrome – a fear of territorial partition by foreign powers (Aydin-Düzgit 2018). The study also found perceptions of the EU to be divided between supporters of the Turkish government and the opposition. Other studies found that external perceptions of the EU reflect that the EU is not just an economic process but also a political process which blurs boundaries between distinct national communities and redefines national sovereignty. Schlipphak’s (2013) study across 47 countries found that holding a supranationalist position is related to more favourable attitudes towards the EU. Isani and Schlipphak’s (2017) study across Arab countries examined ideational factors in the form of what was defined as a desire for sovereignty – a concept reflecting the refusal to bow to external demands to reform. Their findings showed that people who view Western demands for reform as harmful to their own national interests and sense of dignity have more negative feelings towards the EU. Another factor that is related to identity is religion, and there is evidence that a higher level of religiosity is negatively related to the view of the EU (Isani and Schlipphaks 2017).

Overall, these findings imply that when analysing the public acceptance of the EU’s normative message, the power of the local should be considered as power struggles among different domestic actors and their shaping of the image of the EU (Gordon and Pardo 2015). This is highly relevant given that public information about international issues is limited. Attitudes towards international bodies, like the EU, are dependent on the knowledge and political cues that the public receives from political elites in their country (De Vries et al. 2021). Public contestation about international cooperation is politicised by these political actors, which then shapes public opinion (De Vries et al. 2021). The interactions between the EU as an international actor and domestic actors are particularly complex when there are gaps between EU self-perceptions and external perceptions or when EU policy is highly politicised, as in the case of Ukraine (Chaban and Elgström 2020, 2021b). This study, therefore, examines not just the overall influence of the EU normative message but also whether the public’s social and political identities and knowledge about the EU are reflected in a dissimilar influence. For that, the analysis accounts for religiosity, national chauvinism, political orientation and EU knowledge.

The Israeli Case

The EU faces many challenges in its ambition to be a global actor and a normative power. One particularly challenging arena is its Southern Neighbourhood (Keukeleire et al. 2021). After the 2004 enlargement, the EU issued its largest foreign policy plan yet, the European Neighbourhood Policy (ENP), as a joint initiative among 16 countries that were not part of the enlargement programme. The purpose of the ENP was to establish a second circle of European influence between the EU and its neighbours, and it is one of the key tools of EU foreign policy. Israel, which is part of ENP-South, thus provides a unique neighbouring country setting for an exploration of external perceptions given the conflicting trends of Israel’s relationship with Europe and Israeli–EU relations (Pardo and Zahavi 2022).

Despite Israel’s strong connection with the United States, Israelis view Europe as their cultural and economic neighbourhood (Du Plessix 2012). Historically, Zionism was developed in Europe and the founders of the State of Israel wanted to establish a European stronghold in the Middle East. To date, many Israelis have taken dual citizenship with European countries to secure a European passport with all its benefits (Harpaz 2013). At the same time, Europe also represents the dark history of the Holocaust and is seen as a region where Jews can never be safe. Both anti-Semitism and the emergence of a Muslim (and anti-Zionist) population in Europe are popular subjects in the Israeli media and among Israelis. Given its longstanding involvement in the Israeli–Palestinian conflict, the EU is also regarded by many Israelis as pro-Palestinian and, hence, less friendly towards Israel (Harpaz and Shamis 2010, Pardo 2015). In recent years, under the leadership of the right, Israel has become a soft Eurosceptic country, cooperating with Eurosceptic parties in Europe to weaken core liberal EU norms (Filc and Pardo 2021). Furthermore, the political right opposes core aspects of the EU model and undermines the belief in the EU as a model for peacebuilding (Müller 2019). This trend reflects the fact that many Israelis are sceptical of the cosmopolitan norms of Europe, seeing them as naive and unrealistic (Pardo 2015).

In their detailed analysis, Harpaz and Shamis suggested that there is an inconsistency between the “EU’s self-representation as a Normative Power” (2010, p. 599) and the way NPE is perceived by Israelis. They claimed that the external use of NPE might, therefore, be ineffectual in boosting the EU’s normative political agenda in the region: “This dissonance and this gap may compromise Europe’s normative status, widen its external illegitimacy and prejudice its ability to pursue effective involvement in the Middle East” (p. 608).

Harpaz and Shamis (2010) examined this gap by focusing on the elites and on policy. A previous analysis of the Israeli public’s views of the EU also emphasised the gap between the Israeli public and the elite’s views of the EU (Pardo 2015). The gaps between the public’s and the political and social elite’s perceptions of the EU are because the latter have a higher image of the EU (Pardo and Peters 2010). At the same time, of the eight Southern Neighbourhood partner countries, the Israeli public has the highest level of knowledge and understanding of the EU while also displaying conflicting views of the EU and its relations with Israel (EU Neighbours South 2021). Indeed, a recent report (Friedman and Ruff 2022) has shown that these conflicting views have been recognised by EU diplomatic bodies and that their activity in the field of cultural diplomacy in Israel is motivated by a cohesive and clear set of European values which are accepted by only parts of Israeli society.

The current study

While external public perceptions of the EU can be used to describe how the EU is viewed and how these views can be explained, they cannot determine whether they are influenced by the normative message. This article focuses on the question of whether the EU normative message, as embedded in the idea of the “EU as a model”, influences external public perceptions of the EU’s normative actorness. Due to this focus on influence, the article adopted an experimental rather than cross-sectional design. It used a population-based survey experiment to inspect whether Israelis’ exposure to the idea of the “EU as a model” strengthens their views of the EU’s normative actorness or creates a backlash due to a possible gap between their views and EU self-representation as a normative power (Harpaz and Shamis 2010). In other words, the experiment examined whether people exposed to the “EU as a model” treatment were more likely to change their views than those who were not exposed to the treatment. It also enabled an accounting for cultural filters.

Method

Participants and data collection

Participants were recruited via an online register panel and completed an online questionnaire in exchange for a sum approximating $2. Online panels conduct web-based surveys of registered respondents. We used the Midgam Project, a leading online survey research company with a pool of over 60,000 potential respondents. Participants were informed that participation was anonymous, given contact details in case of queries, and required to complete a consent form. Some participants were excluded due to an instructional manipulation check (IMC) failure, indicating a lack of sufficient attention to the survey questions. We also excluded respondents who completed the survey in the upper 5% of the required time.

Data collection took place January 8–9, 2023. While the sample is heterogeneous and includes quotas for age and gender to ensure a representative sample, it is not a probability sample of the entire population of Jewish Israelis and is thus not a representative sample of the target population. Such panels have, nonetheless, been used in numerous studies in Israel (e.g. Feinstein and Bonikowski 2021 Sulitzeanu-Kenan and Halperin 2013) and are seen as an approximation of the target population. The sample (N = 1286) was heterogeneous: 50.6% women; mean age 40; 8% born outside Israel; 51% secular; 29% high school education or less; 44% below average family income.

The experimental manipulation

The survey proceeded in three parts. First, respondents were asked questions about their knowledge of the EU, their national identity and their political orientation. Second, respondents were randomly assigned to the condition group or the control group. Third, respondents were asked about their evaluation of EU actorness in the context of Israel and these questions were used to construct the dependent variable.

The treatment condition was based on a short text that presented in a simple manner the development of the EU from past conflict to current peace while providing evidence that supported this message and also referring to EU values. The text was as follows:

The EU developed after World War II in order to settle the basic sources of conflicts and crisis throughout European history. Despite various difficulties, the EU gained increasing influence and was a successful model for the ways in which countries that had been enemies for centuries, like France and Germany, can cooperate. After the fall of the Soviet Union, the EU expanded to the East and successfully included countries with a communist past, like Poland and Hungary. In 2012, the EU won the Nobel Peace Prize for contributing for “over six decades … to the advancement of peace and reconciliation, democracy and human rights in Europe,” as determined by the prize committee. The EU basic values are human dignity, freedom, democracy, the rule of law, equality, and human rights.

After reading the text, respondents were asked two factual questions about the text (which year and which prize were mentioned in the text) as a manipulation check and an open question about their feelings. This open question provided an overall estimation of respondents’ reactions to the condition, and we received various responses. For some, this was new information that made them feel optimistic. Others, however, were more sceptical of this description and used more negative expressions like hypocrisy or referred explicitly to the EU as hostile towards Israel. Between these opposing responses, there were also respondents who claimed indifference to this information.

The sample was split between those who did not read the text (n = 657) (the control group) and those who read the text (n = 627) (the condition group). Respondents were randomly assigned to the groups. Randomisation checks with one-way ANOVA analysis showed no significant differences between the two groups in age (F = .24, p = .15), gender (F = .00, p = .96), education (F = .00, p = .96), SES (F = .60, p = .43), religiosity (F = 1.36, p = .24), political orientation (F = 2.04, p = .15) or knowledge about the EU (F = .02, p = .87). We, therefore, assumed that the respondents in both groups share the same characteristics.

Measures

The dependent variable, “perceptions of the EU as a normative power” (PEUNP), was constructed to examine views of the normative component of the EU’s normative actorness in the context of Israel. The variable was constructed from items such as “the EU is a model for peaceful conflict resolution”. Five items were found loaded on one factor in the exploratory factor analysis (65%; Eigenvalue  = 3.61) and the loadings of the factors were above .64. The scale internal reliability was α = .86.² The control variables were based on standard measures of sociodemographic characteristics (age, gender, education, socioeconomic status [SES]), religiosity, political orientation, national identification and chauvinism as well as a test of EU knowledge by using a set of five questions about the EU. The items’ wording and descriptive statistics of the scales are presented in Table 1.

Table 1. Measures and descriptive statistics.

Variables	Items	Range	Mean(S.D)
PEUNP	The EU is a model for peaceful conflict resolution The EU is a model for a way in which former enemies could cooperate EU actions in our region promote peace EU actions in our region contribute to democracy and securing human rights EU actions in our region are dictated by interests alone (reverse item)	1–7	3.3 (1.25)
Gender (female)		0/1	50.5%
Age			40.25(12.64)
Education		1–13	7.21(2.37)
Religiosity	0 = Secular; 1 =  Traditional, Religious, Ultra-orthodox	0/1	51%
Income	0–5		2.69 (1.26)
Political stance	1 = Right 11 = Left	1–11	4.18 (2.49)
Test of EU knowledge	Does the EU have a united army? Does the EU have a constitution? Which city is the EU capital: Berlin, Paris, Brussels, The Hague? Which country is in the process of leaving the EU: Greece, the UK, Spain, Hungary? Which of the following is not an EU member state: Estonia, Switzerland, Denmark or Lithuania?	0–1	.53 (.25)
National identification	identify with Israel Israel is an important part of my identity Israeli identity is more important to me than other types of identity It is not important for me to see myself as Israeli (reverse item)	1–7	5.69 (1.34)
Chauvinism	The world would be a better place if people from other countries were more like Israelis Generally speaking, Israel is a better country than most other countries In comparison with other nations, Israel is a very moral nation Other countries can learn a lot from Israel	1–7	4.73 (1.48)

Findings

As the randomisation test indicated no differences between the groups, we simply compared the PEUNP mean between the groups. The results indicated that PEUNP was higher in the condition group (M = 3.68; SD = 1.30; n = 629) than in the control group (M = 3.29; SD = 1.25; n = 657) and that this difference is significant t(1284) = 5.465, p = .000, d = .07. The differences between the groups are statistically significant, indicating that the observed effect did not happen by chance. But does it have any practical significance for the real world? Assessments of significance commonly use effect size, which is medium in this case (Cohens’ D = .3). Another indication of effect size is that the group exposed to the text perceived the EU to be normative power approximately 6.5%.

Yet, beyond the numbers, the assessment of the experiment results requires additional studies that will conduct similar experiments to obtain an overall estimation. In addition, there are no sufficiently longitudinal surveys of external perceptions of the EU – or even of other international bodies – that can be used to help estimate the extent to which these attitudes are immovable. We can, however, gauge from other issues (Adida et al. 2018) that the significance of the change is in line with the standards of studies that use similar approaches.

The finding of a mean difference does not rule out the possibility that differences among each group and not between the groups might influence the response to the condition. Therefore, the potential influence of confounding variables should be accounted for. Table 2 presents OLS regression models in which the sociodemographic characteristics (age, gender, education and SES), religiosity, political orientation, national identification and chauvinism as well as EU knowledge are related to PEUNP in Model 1. It is evident that religiosity, chauvinism and people with greater EU knowledge are related to lower levels of PEUNP, while affiliation with the political left is related to higher levels of PEUNP. Model 2 included the condition as a dummy variable. The results indicated that despite controlling for various potential confounding variables, those exposed to the condition have higher levels of PEUNP. The contribution of the condition to reducing the unexplained variance is 2%.

Table 2. Predicting the influence of the treatment on perceptions of the EU as a normative power.

	Model 1	Model 2
Gender (female)	.10***	.10***
Age	−.27	−.27
Education	.03	.03
Income	.04	.04
Religiosity	−.13***	−.13***
Political affiliation (left)	.31***	.31***
EU knowledge	.16***	.16***
National identification	.08**	.08**
Chauvinism	−.14***	−.14***
Condition		.12***
R²	21%	23%

*p < .05; **p < .01; ***p < .001.

Notes: N = 1286. Standardised coefficients.

To further inspect whether the condition influenced the entire sample, we ran another analysis interacting the stronger predictor, political affiliation, with the treatment while controlling for all covariates (unreported in Table 2). While the interaction indicated the reduction of the effect (Δt = 2.54), it remained significant. In other words, even among those affiliated with the right, the condition increased their PEUNP. Figure 1 further shows the influence of the condition among those affiliated with the left and the right.³ Despite the much higher levels of PEUNP among the former (Figure 1(a)) than the latter (Figure 1(b)), it is evident that there is a significant gap between those who were exposed to the condition and those who were not. The figures also illustrate that the magnitude of the effect is higher among those affiliated with the right than the left.⁴ Put differently, the influence of exposure to the notion of the “EU as a model” was stronger among those with, a priori, more negative views of the EU.

Figure 1. a PEUNP difference between the groups among those affiliated with the left. Figure 1. b PEUNP difference between the groups among those affiliated with the right.

Conclusions

While the idea that the EU, as a normative actor, is contested among scholars (Tonra and Christiansen 2018), NPE is a central aspect of the ENP (Keukeleire et al. 2021). It is, specifically, part of the EU’s involvement in conflict resolution and peacebuilding attempts among its neighbouring countries, particularly Israel/Palestine (Diez and Pace 2011, Harpaz and Shamis 2010, Müller 2019, Pardo 2015). In addition, NPE is part of the EU’s increasing public diplomacy efforts which involve various EU actors trying to influence the public in third countries as a way of promoting various goals. Studies inspecting external perceptions of the EU have indicated public awareness of the EU, recognition of EU activities within different spheres and the EU’s positive image across most countries (Chaban et al. 2015, Chaban and Holland 2019, Speyer et al. 2021). However, the question of whether the normative component of the EU by itself influences external perceptions has not yet been explored.

By using the case of Israel, a country with contradictory relations with the EU (Pardo 2015), this article has tested whether the EU’s normative message, as embedded in the idea of the “EU as a model”, influences external public perceptions of the EU’s normative actorness. Conceptually, we argued that the “EU as a model” underlines both the elite’s and the public’s views. Empirically, a population-based survey experiment was used to explore whether exposure to the idea of the “EU as a model” impacts views of the EU in terms of either public recognition of its normative actorness or a backlash against its self-representation as a normative power. The findings are clear and straightforward: exposure to the “EU as model” message increased PEUNP. This impact held when we controlled and interacted with the exposure with various cultural filters which are key factors in shaping views of the EU like political affiliation.

Prior to discussing the findings’ theoretical and practical implications, the study limitations should be recognised. Of great relevance here is the argument by Manners (2021) in the introduction of a recent collection about emotional norms in EU foreign policy that the choice of methodology is critical. While a population-based survey experiment enables us to investigate the impact, surveys are a limited tool for grasping complex concepts, such as attitudes towards the EU or normative components. Furthermore, establishing the internal and external validity of the experiment requires additional studies that will replicate the design with different manipulations and across dissimilar contexts. The generalisation of the findings requires further investigation in other countries with similar contradictory relations with Europe, such as Turkey (Aydin-Düzgit 2018). Surveys are, in addition, a limited tool for constructivist or critical interpretations of NPE, and only studies that adopt additional methods, such as Chaban and Holland’s (2019) analysis of the media coverage alongside surveys, could strengthen the theoretical arguments proposed here. Another limitation of the survey that should be acknowledged is the gap between the abstract view of the EU normative message in abstract and the actual political reality. The EU’s role in the Middle East and, especially, its relations with Israel, require further contextualisation of views of the EU. A generalisation of the findings requires not only the study of other countries but also a more focused examination of how the EU message is related to views about EU involvement in the relationship between Israel and the Palestinians.

Regardless of these limitations and the need for further studies, our findings contribute to the understanding of NPE and to the field of public diplomacy. They demonstrate that the strategic narrative of peace and cooperation, which is embedded in the creation of the EU, is not just a narrative constructed by scholars or as part of a discourse among elites. To the extent that our empirical analysis is valid, it is clear that exposure to the “EU as model” message is a strategic narrative that is recognised among the general public of a third country and is not only the EU’s self-representation; indeed, the framing of the “EU as a model” impacts views of the EU’s normative actorness. Our study focused on a key EU narrative: the captivating story of how the EU emerged from the ashes of World War II to become a community of peace and security. It should be remembered that the EU is connected to various strategic narratives, such as economic or environmental narratives (Manners and Murray 2016). Future studies might examine to what extent these narratives also impact public perceptions.

Our findings showed that exposure to the “EU as a model” message holds even beyond cultural filters. It is well known that the recognition of the EU as a normative actor must consider cultural filters and, especially, the power of the locals (Gordon and Pardo 2015). The interactions between the EU as an international actor and domestic actors are extremely complex, with perception gaps between EU self-perceptions and external perceptions or cases of EU policy being highly politicised as in the cases of Ukraine (Chaban and Elgström 2020, 2021b) and Israel (Harpaz and Shamis 2010, Müller 2019, Pardo 2015). Yet, despite the importance of cultural filters, the findings imply that the “EU as model” message is not limited to those who hold an a priori more positive image of the EU. In fact, we found that among those who hold more negative views of the EU (right-wing affiliation) the influence of the exposure to the “EU as a model” message was a little higher than among those who tend to hold more favourable views of the EU (left-wing affiliation). These findings also have implications for public diplomacy.

Despite growing interest in the field of public diplomacy and the ways in which the EU uses public diplomacy (Fanoulis and Revelas 2023, Song and Fanoulis 2023) we we still lack knowledge about the impact of public diplomacy on the wider public. Most analyses of public diplomacy have examined actors, tools and messages but not their impact. This article did not inspect the influence of a specific act which is part of the public diplomacy toolbox. Instead, we focused on the underlying message of the actor story itself and found that exposing people to this story led to higher estimations of the EU’s normative model. Unlike previous studies that described public views of Europe (Friedman and Ruff 2022), our approach showed the direct impact of the “EU as a model” message on views of the EU.

In a recent special issue of the Journal of Contemporary European Studies on “Global Perspectives on European Union Public Diplomacy”, Song and Fanoulis concluded:

Although the contributors to this Special Issue touch upon aspects of the effectiveness of EU public diplomacy in their articles, a systematic and comprehensive methodological roadmap for investigating the effectiveness of EU public diplomacy around the world is lacking. We hope that future research can shed light on this unexplored territory. (2023, p. 6)

While this article does not provide this “comprehensive methodological roadmap”, it nonetheless indicates a possible line of research.

Besides the theoretical implications of our findings, they also have practical implications. The EU project’s success story can be used, if framed correctly, to inspire higher public estimation of the EU and, perhaps, to better accept its role in conflict resolution. If exposure to the “EU as a model” message was found to be influential in Israel – a country with ambivalence and even scepticism of the EU – it can probably be found in other cases. While there is strong evidence that European normative values are only accepted by parts of Israeli society (Pardo 2015), the current findings demonstrated the potential influence of the “EU as model” message on all parts of Israeli society. Public diplomacy practitioners can, therefore, use these findings to sharpen their message in Israel and elsewhere. However, the practical implications of the findings should also consider the question of the magnitude of the exposure to only one paragraph. Survey experiments, like the one used here, which are embedded in cross-sectional surveys cannot tell us much about the duration of the effects (Gaines et al 2007). We recommend that future studies ask to what extent the change in attitudes is sustainable. This question is central in the case of Israel, whose political leadership is constantly bashing the EU and, especially, the EU normative message as Israel increasingly becomes a soft Eurosceptic country (Filc and Pardo 2021).

Acknowledgement

The authors would like to thank The Simone Veil Research Centre for Contemporary European Studies, Ben Gurion University of the Negev, for supporting the project.

Disclosure statement

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

Notes

1 The emphasis is ours.

2 For a robustness check, we also included a scale for overall sympathy to the EU and not just its normative component. Yet, we could not establish distinctive validity from PEUNP as their correlation was .76. The robustness check was, therefore, not implemented.

3 As political affiliation is a continuous variable, we created a dummy variable according to the median.

4 The mean differences among those affiliated with the right (.92) were higher than those affiliated with the left (.67).