Introduction

On February 24, 2022, Russia launched a full-scale invasion of Ukraine, causing millions of people to seek safety internally or in other countries. The European Union and its Member States have responded in a supportive way to the arrival of displaced people from Ukraine, mobilising legal and infrastructural frameworks of reception that vary from country to country (Gerlach & Ryndzak, 2022; Trauner & Valodskaite, 2022). Hungary, with a shared border with Ukraine, became both a destination and a transit country for masses of displaced people, primarily women and children. The host society started to experience extraordinary times in the spring and summer months of 2022. However, it is far from obvious if and in which ways these events were idiosyncratic compared to other recent extraordinary times, such as the 2015 refugee wave in Europe and the pandemic in 2020–21.

The experience of actively reacting to forced human mobility events since 2015 due to the interlinked effects of regional and global ‘crises’ led to the multiplication and diversification of civic solidarities with the displaced Ukrainians in 2022 across Europe. This phenomenon has inspired scholars interested in the longer-term trajectories of civic solidarityFootnote 1 to map and explain how support for displaced people from Ukraine is organised, shared, and justified by different societal actors by contrasting various forms of solidarity with the diversity of refugees over time (Dollmann et al., 2022; Baszczak et al., 2022). In a similar vein, our research aims to explore the reception of displaced people from Ukraine and understand civilian expressions and acts of solidarity by relating the latter to phenomena experienced during ‘the long summer of migration’ of 2015 and to solidarities generated by the COVID-19 pandemic.Footnote 2

Beyond highlighting historical links and (dis)continuities that have emerged in the last decade or so, we aim to contextualise civic solidarities through their relationship with governance structures—more precisely, the political discourses and state institutions that shape, and in many instances, weaken or erase civic solidarities. The first and more frequent perspective on the governance of refugee reception comes from political science and sociology, with their long tradition of analysing population attitudes (prejudices, xenophobia) and various practical forms of civic engagement (associational life, advocacy, political mobilising) as influenced and shaped by political and public discourses and measured through political attitudes and party preferences. On the other hand, we intend to analyse civic solidarities as embedded in the societal distribution of care duties which reflect limited public welfare provisions, feminisation, and the devaluation and volunteerisation of care work in most polities of Europe. This latter aspect of refugee reception is mainly explored in feminist sociology and anthropology and in the literature on refugee solidarity, civil society, social work, and volunteering (Muehlebach, 2012; De Jong, 2019).

By combining attention to the role of political discourse in shaping civic solidarities and the distribution of solidary acts, we intend to canvass a complex picture of refugee reception through civic forms of solidarity in Hungary. We reveal that while solidary attitudes are primarily interlinked with ambivalent political discourses, actual support practices mirror broader societal configurations. While attitudes indicate an unprecedented level of solidarity and an almost society-wide consensus, practical solidarity acts are primarily implemented by a devalued, downsized, and overburdened care sector staffed by a feminised labour force and backed by an enormous amount of voluntary civic support. Throughout this endeavour, we highlight tensions and ambivalences between attitudes and practices: what people reflexively think and declare as their standpoints might have non-evident and complex relations with how they act on a practical level—a relationship rarely explored by research in separate disciplinary perspectives and fields of inquiry.

Regarding the structure of the paper, first, we highlight two outstanding aspects of governing refugee reception: the importance of political discourses in forming attitudes of solidarity and the nexus of solidarity acts and the distribution of care duties in society. Next, we depict Hungary’s ‘refugee-welcoming’ context, the research agenda, and the method. The analytical section discusses typical forms of solidarity acts and dominant attitudes to accepting Ukrainian refugees. By highlighting tensions between attitudes and actions, we also unveil the effects of the political climate and the role of caring positions and institutions in solidarity with displaced people from Ukraine. A discussion and conclusions follow in the final section.

Civic Solidarities Shaped by Political Discourses and the Distribution of Care

The relationship between broader political discourses and the solidarity attitudes of people is of prime interest to us in the context of authoritarian-populist governance. Populist politics is eager to produce, magnify, and capitalise on xenophobic attitudes and everyday racialisation, thus often finds easy targets among minorities and migrants. In line with the literature, we argue that the willingness to help refugees tends to be weaker in countries where xenophobic and anti-immigration-promoting parties are strong. These political forces and their media apparatus effectively frame refugees and immigration as a threat to the stability and integrity of host societies. This often affects the positions of moderate parties and generates suspicion and fear in the wider public (Koos & Seibel, 2019; Wodak, 2019). The current Hungarian regime represents a textbook case of how the ‘refugee and migrant threat’ may be instrumentalised to enact a firm populist political and policy order. Political polarisations also intervene in the formation of attitudes and motivations regarding refugee solidarity. Pro-government voters seem to be more liable to accept securitisation narratives, whereas opposition voters have been more open to humanitarian narratives (Janky, 2016; Barna & Koltai, 2019; Gerő & Sik, 2020; Tilles, 2021).

The literature also widely discusses how the dominant political narratives related to refugees and migrants and the host societies in Europe construct and instrumentalise ‘deservingness’ hierarchies and classifications. Legitimating the process of the redistribution of resources to specific segments of society by assigning moral value to individuals and various social groups is increasingly perceived as a central challenge to managing social inclusion (Anderson, 2013; van Oorschot et al., 2017; Holzberg et al., 2018; Streinzer & Tosic, 2022). The reasoning behind different perceptions of deservingness is pertinent to broader European publics and even beyond, yet authoritarian-populist regimes often use it in intensively polarising ways. The political and media discourses that rapidly emerged in Central and Eastern Europe after the start of the current war in Ukraine also involved expressing selective symbolic support for everyday solidarity attitudes and practices concerning the displaced from Ukraine and elsewhere (Cantat, 2022). Accordingly, our inquiry intends to explore the intricacies of how the political climate and dominant narratives deployed during the Ukrainian refugee crisis offered ambiguous guidance and invitations to solidarians in Hungary to step in to actively welcome different groups of the displaced.

Solidarity values and attitudes are also tinkered with due to the growing interest of neoliberal governance regimes in volunteering and not-for-profit actors’ services associated with managing the consequences of crises, such as the economic crisis of 2008 and its impacts (Musick & Wilson, 2008; Muehlebach, 2012). Authors have described how the ideal of moral neoliberal citizenship encourages and exploits voluntary work by those in precarious positions in the labour market (Muehlebach, 2012). Critical sociology of care also provides warning accounts of the withdrawal of the state from social services and the radical commodification of care (Fraser, 2016; Melegh & Katona, 2020; Fodor, 2022). The reproduction of society through care is supported by and solidifies inequality relations of gender, paid and unpaid work, formal and informal employment schemes, and the distribution of wealth. These nexuses are also demonstrated and intensified in various domains of refugee solidarity work (De Jong, 2019; Szczepanikova, 2010; van der Veer, 2022).

Notwithstanding the massive neoliberalisation of solidarity work, researchers are keeping their eyes open to the potentialities of reclaiming or protecting spaces in which bottom-up and civic actions may resist exploitation (Woodly et al., 2021). Solidarity research extensively discusses how the drive to assist and care may become widespread at times of crisis. The rich literature that emerged in relation to the ‘refugee crisis’ in 2015 in Europe widely discusses the politics of care by zooming in on the relations between solidarians and displaced people in terms of horizontality, equality, and non-charity, often labelled ‘subversive humanitarianism’ (Feischmidt & Zakariás, 2019; Vandevoordt, 2019). The newest solidarity literature reflects upon the cross-cutting, complex, and global pandemic experiences of solidarity and civic interventions during the economic- (2008), the ‘refugee-’ (2015), and the pandemic-related crises (2020). Accordingly, care may be seen as a means of prefiguring and enacting ‘alternative ways of being together in a fundamentally non-exclusionary, non-sentimental manner’ (Ticktin, 2021: 916).

Sources and Spaces of Welcoming in Hungary in 2022

According to the official statistics, by June 2022, 1,312,550 people had arrived from Ukraine to Hungary, either directly crossing the joint border or through Moldova and Romania. Hungary became the first stop for these displaced people (UNHCR, 2022). Most of them, like the refugees and asylum seekers from the Middle East in 2015, transited through Hungary, staying only for a couple of days or weeks. It is difficult to assess the number of displaced Ukrainians staying in Hungary at any moment since February 2022. Converging information from UNHCR (UNHCR, 2022) and official domestic sourcesFootnote 3 suggests that by the middle of the summer, 24,231 people had applied for temporary protection, and a further 50,000 people with dual Hungarian-Ukrainian citizenship were present. Members of this latter group, mainly from the Transcarpathian region,Footnote 4 are not subject to registration or visa obligations. The rapid inflow of more than a million people placed considerable pressure on the immigration system and, similarly to the arrival of Middle Eastern and African refugees in 2015, generated massive attention from civil solidarity actors.

The domestic political climate for the reception of displaced people in Hungary has been complex and controversial since the start of the current war against Ukraine. First, the government’s central narrative has embraced solidarity with displaced people in line with the current pan-European political consensus on actively welcoming Ukrainian refugees but in contrast with the official xenophobic communication deployed in 2015 (Bernáth & Messing, 2015; Messing et al., 2022). Second, such governmental support for Ukrainian refugees reinforced the rejection of Middle Eastern and African refugees and asylum seekers by contrasting the former as ‘real’ vs. the latter as ‘bogus’ refugees.Footnote 5 Third, the Hungarian government has taken a unique position on the war in the European Union by articulating anti-Ukraine political messages: it has opposed arms transfers and economic sanctions against Russia and emphasised the power ambitions of Western political actors and NATO. It has also stressed the responsibility for the war of the USA and President Zelensky and highlighted the Ukrainian government’s ‘oppressive policies’ towards ethnic minorities in general and the ethnic Hungarians of Ukraine in particular.

In the spring and summer of 2022, displaced people from Ukraine arrived at the ruins of refugee and immigration assistance infrastructure in Hungary. In congruence with its militant anti-immigrant discourses pursued since 2015, the government has deconstructed the institutional system of asylum: it has closed down almost all refugee accommodation facilities, ended almost all types of integration assistance and care services, and made the legal-administrative procedure for claiming asylum almost entirely inaccessible to applicants (Soltész & Vadai, 2022; Kováts & Soltész, 2022; Szabó, 2020; Tóth, 2022). After February 24, 2022, although the legal framework of granting temporary protection status was quickly established and some financial resources were mobilised, rebuilding the refugee assistance apparatus was slow. In the first weeks of the war, this placed an enormous burden on municipalities, grassroots civil initiatives, and international aid agencies.Footnote 6 However, an unparalleled upsurge in solidarity activism activated people from different walks of life with variegated forms of motivation.

Solidarity Actors, Acts, and Rationales

Data and Methods

To map solidarity with displaced people fleeing the war in Ukraine and actual practices aimed at helping them, we conducted a population survey of a sample of 1000 respondents, representative of the population in Hungary according to gender, age, settlement type, region, and educational background. Data collection was completed in June 2022.Footnote 7

We measured solidarity with two sets of questions. On the one hand, we asked respondents about their support for the admission of different groups of displaced people to Hungary. We defined four such groups based on the different images and figures of refugees described by the media in Hungary in the respective period: Ukrainian refugees in general, Transcarpathian Hungarians, the Roma from Ukraine, and students from Asia and Africa (Noyoo et al., 2022; Eredics et al., 2022). The question differentiated between no admission, admission for a limited period (for a couple of months or until the war ends), and indefinite admission. In this way, we hoped to uncover how differently Hungarians perceive the various groups of displaced people based on their (assumed) deservingness.Footnote 8 However, due to the generally supportive sentiment, in the analysis, we use a simple dichotomous variable differentiating only between people who support the admission of displaced people regardless of the group they belong to, at least until the end of the war, and those who would only support the admission of displaced people for a limited period or would not support their admission to Hungary at all. The questionnaire also inquired about acts of solidarity, i.e. the actual provision of help to the displaced people. These acts were identified as voluntary or paid work, financial or in-kind donations, and support finding accommodation and work.

In the section on statistical analysis, we first present descriptive results that explore the differences and similarities between the attitudes of solidarity and actual participation in help for the displaced. We use two-dimensional cross-tables to characterise the social position of those involved in assistance work by taking into account two groups of variables: (1) socio-demographic and socio-economic variables: gender, age group, type of municipality, administrative-geographical region, household size, educational attainment, financial situation, labour market sector, and experience of discrimination; and (2) associational life and political attitudes: religiosity and relation to the church, civic participation before the war, civic assistance to refugees in 2015, political party preferences, and finally, perceptions of the war in Ukraine (responsibility assigned to various political actors for starting the full-scale war).Footnote 9 We also applied binary logistic regression models to explore the multi-dimensional relationships between solidarity actions and attitudes and their potential background factors. Such models help assess the association between an outcome variable (in our case, solidary actions and attitudes) and specific background variables after filtering out the potentially confounding effects of other background variables. The statistical package SPSS 20 was used for analysis.

Non-selective Acceptance and Acts of Solidarity: Levels, Forms, Institutional Contexts

We approached attitudes to refugee acceptance using ‘deservingness’ measures: besides employing the broadly labelled category ‘refugees from Ukraine’, we aimed to explore how the willingness to admit differed for specific subgroups of refugees.

As Fig. 1 shows, as of June 2022, and regardless of the category, at least three-quarters of respondents would welcome specific groups for the duration of the war or longer. In addition to the dominant viewpoint that all displaced groups should be admitted to Hungary for (at least) the duration of the war, differences in perceptions of the various categories are apparent. Hungarians from Transcarpathia are the only category most respondents (59%) would accept for a period longer than the war. In the case of Ukrainian refugees in general, Roma from Ukraine and those originally from African-Asian countries, the proportion is much smaller (37%, 28.5%, and 31%, respectively). In addition, while the proportion of those who would entirely reject Hungarians from Transcarpathia is almost invisible in our sample, 8–9% of respondents would entirely reject the admission of Roma from Ukraine and of persons of African or Asian origin.

Fig. 1
figure 1

Willingness to accept various categories of displaced people

Full size image

Despite these divergences, our data show that most respondents were welcoming, especially when compared to population attitudes identified in earlier studies on contexts other than the war in Ukraine (Bernát & Simonovits, 2016; Messing & Ságvári, 2016; Sik et al., 2016). Due to this general spirit of acceptance, we decided to derive a single measure from the four. Merging all the categories together, at the time of the survey, 69% of respondents with valid answers (N=972) claimed they would accept any group of displaced persons, at least for the duration of the war in Ukraine, regardless of their social background; the rest (31%) were either selective (would accept only specific groups for the duration of the war) or would accept all groups for a shorter period or not at all.

Regarding helping practices, a large proportion of respondents, 40%, had actively assisted Ukrainian refugees at least once at the time of the survey.Footnote 10 Most people had helped with material and financial donations (around 23% of respondents in both cases), while 8% had helped with voluntary work. The proportion of respondents who had helped with accommodation was relatively small among all respondents (2%), as was the proportion who had provided employment (4%) or helped in other ways (4%) (Table 1).

Table 1 Forms of assistance (N=1000)
Full size table

Three-quarters of the solidarians helped by providing financial or in-kind donations but did not volunteer or share their homes or other private resources. The remaining quarter of helpers (also) helped with volunteering, accommodation, or job placement. To measure the intensity of assistance, we also asked about its frequency. At the time of the data collection, less than a third (31%) of helpers had helped only once, a significant majority (64.5%) had helped a few times, and almost 5% reported helping regularly.

In view of the level of volunteering in recent decades in Hungary, these participation rates seem to be exceptionally high, especially as regards support for displaced people (Sik & Zakariás, 2021). Engagement in this wave of solidarity exceeds that in 2015 by ten times (40% compared to 3.5% respectively, Zakariás, 2016). Among those who provided assistance through organisations (54% of respondents), most acted through aid organisations, followed by municipality or government-related institutions (including schools, social and other institutions, as shown in Table 2). A relatively large proportion of respondents mentioned religious communities, church institutions, various market actors (mainly banks, supermarket chains), and NGOs. Ten per cent of the helpers (4% of the total sample) were partly or fully employed in helping refugees, while 90% had helped on a purely voluntary basis.

Table 2 Organisational background
Full size table

Attitudes, Acts, and the Distribution of Care

A position of non-selective acceptance is associated with specific socio-economic characteristics, such as respondents’ educational level and financial status. Educational level is a central dividing line between people with non-selective and selective attitudes. Although more than half of our respondents without a school leaving certificate would also accept everyone, the proportion of non-selective attitudes is above 70% among those who had finished at least high school.

The proportion of those who would welcome everyone is significantly smaller among those with no savings compared to the average of the total sample (58% vs 69%). The association between savings and welcoming attitudes, however, is not linear: while the second, third, and fourth financial situation categories (having 7–12, 3–6, and 1–2 months of savings) are associated with a larger proportion of non-selectively accepting respondents (74%, 73%, and 72%, respectively), those in the best financial situation (those with at least one year’s savings) have only the same rate of non-selective acceptance as the sample average. In terms of occupational sectors, our data show that the largest proportion of respondents who would welcome all displaced people from Ukraine are connected to the health and commerce sectors. In contrast, those linked to agriculture, tourism, hospitality and, surprisingly, social service occupations are represented in the smallest proportions (although the latter category is associated with a small sample size).Footnote 11

Non-selective acceptance attitudes were also explored using multi-dimensional logistic regression models (see Table 5 in the Appendix). The results show that the effects of education and financial status are corroborated in these models and similar to those identified in the two-dimensional cross-tabulations. It is also noteworthy that, just as we found no significant gender differences in the prevalence of acceptance in the two-dimensional cross-tables, we found no significant differences between the odds of women or men being accepting in the regression model (that is, after filtering out the potential confounding effects of other background variables). Also, in the multivariate model for the attitude of non-selective acceptance, occupational sector was not found to be significantly associated with willingness to accept refugees from Ukraine.

Compared to attitudes of acceptance, acts of solidarity are more strongly associated with a broader range of socio-demographic and socio-economic characteristics of respondents. First, we found that (in line with the gender inequality in the caring professions) women helpers were significantly and strongly over-represented, as were middle-aged people (50-65 years old) (Fig. 2 and Appendix Table 3).

Fig. 2
figure 2

Gender distribution and solidarity acts

Full size image

Second, the relationship between socio-economic status and offering help to displaced people is mixed. In line with the literature on charitable giving, a significantly larger proportion of helpers is identifiable among those who have graduated high school and even more among those with tertiary education than those without a high school diploma. While no significant relationship was found with financial situation (at a 0.05 level of significance), the sector associated with the respondent’s present or past job is very strongly related to providing assistance to refugees. The largest proportion of helpers (58%) are connected to the public sector (health, education, and social sectors; and within that, the health and social sectors [66% and 74%]); furthermore, business owners also report higher-than-average involvement in solidarity acts (Fig. 3).

Fig. 3
figure 3

Distribution of solidarity acts according to respondents’ occupational sector

Full size image

As with attitudes, a binary logistic regression model of helping was also constructed (See Appendix, Table 3).Footnote 12 Acts of solidarity primarily reflect two-dimensional relationships: each background variable alone (i.e., after filtering out the effects of the others) is similarly associated with helping as the raw background variables. Gender, age group, and educational attainment are similarly related to helping as in the cross-tabulations; the effect of occupation sector also remains significant, although slightly modified.

It is little surprise that our data reveal that helping refugees is unequally associated with those who are already heavily engaged in care duties— that is, women and those working in gendered and underpaid care sectors. Furthermore, while general attitudes towards accepting refugees from Ukraine in Hungary were weakly or not linked with gendered roles and occupational positions, participation in acts of solidarity was found to be much more strongly related to gender and occupation and related socio-demographic characteristics.

Associational Ties and Political Attitudes

Looking at the patterns of acceptance attitudes according to political party preferences (Fig. 4), the largest proportion of respondents who would welcome all groups are supporters of the opposition (not including earlier radical-right Jobbik party supporters). The figure is slightly lower than the sample average (65%) among supporters of Fidesz/KDNP and radically lower than the sample average (at 46–50%) for supporters of Jobbik and the radical-right Our Homeland Movement (Mi Hazánk Mozgalom).

Fig. 4
figure 4

Willingness to accept refugees and participation in acts of solidarity according to party preference, % of respondents

Full size image

Besides solidarity attitudes, participation in helping is also strongly associated with political party preferences (Fig. 4). While helpers are over-represented both among supporters of the left-liberal opposition and the governing Fidesz-KDNP (44% and 42%, respectively), helpers are strongly under-represented among supporters of Jobbik and the Our Homeland Movement and among those who would not vote (25%, 27%, and 28%, respectively).

Unsurprisingly, willingness to accept all groups seeking refuge and involvement in assistance differed in line with one’s pre-war civic activity and involvement in humanitarian aid in 2015 (see Tables 3 and 4 in the Appendix). The impact of religiosity was more ambiguous: while people connected to institutional religiosity were more involved in helping than the rest of the respondents, their attitudes towards non-selective acceptance did not differ from those of the larger sample.

Finally, we also explored differences in solidarity attitudes and practices according to perspectives on the war—specifically, according to the perceived responsibility for the ‘Current War in Ukraine’—namely (1) Vladimir Putin, (2) Volodymir Zelensky and the Ukrainian government, or (3) the USA and NATO. Our results show that those who hold Vladimir Putin primarily responsible, and those who hold Volodymyr Zelensky and the Ukrainian government and the USA and NATO only slightly or not at all responsible, are much more likely to be non-selectively accepting of refugees than the sample average. In the case of solidarity acts, this association was found for only one actor: among those who say Vladimir Putin is not at all responsible, the assistance participation rate is only 19%; among those who see the Russian President as slightly or highly responsible, assistance is close to the sample average (Fig. 5).

Fig. 5
figure 5

Solidarity practices according to perception of responsibility of Vladimir Putin for the Ukrainian War

Full size image

Looking at the effects of political and associational variables inserted into the multivariate model for acceptance attitudes, we see that the impact of participation in 2015 (or the memory of it) and civic activity prior to the war disappear—these are possibly incorporated into the other variables that are included. Attitudes to acceptance are strongly related to belonging to a right-wing extremist electorate and, to a lesser degree, to having a small amount of savings. These patterns might tally with explanations for the rise of the extreme right in Hungary and the theory of status anxiety.

Our multi-dimensional model also reveals that acts of solidarity are impacted by political party preferences, although less than attitudes of acceptance, and by participation in refugee support in 2015. Interestingly, at the time of our survey, the potential patterns of status anxiety underlying acceptance attitudes were not found for acts of solidarity, which phenomenon proved to be dominated by caring roles, habituses, and infrastructures, as explored in the previous section.

Discussion and Conclusions

In our research, we explored various forms of solidarity in Hungary with people fleeing the war in Ukraine. We examined voluntary (and, to a lesser degree, paid) forms of support and population attitudes towards the admission of and assistance to displaced people. Based on a survey of 1000 persons, representative of the adult population of Hungary, we found that at the time of the survey (June 2022), a large part of society (36–44%) had been actively involved in solidarity activities.

Assistance took various forms: the most common was donating money or in-kind help (23–23%) and volunteering (7%) while offering accommodation, support for job search and other forms of assistance were mentioned by less than 5% of respondents. At the time of the survey, only 4% of the respondents (10% of those actively involved in helping) reported that their participation was mediated through paid employment.

Besides actual support, we also looked at the attitudinal aspects of solidarity with people fleeing the war in Ukraine: first, regarding willingness to be personally involved. In addition to the 40% who had actively helped, another 28% claimed they would have helped if they could. Second, we measured willingness to admit people to Hungary, regardless of their background, at least for the duration of the war: almost 70% of our respondents claimed to support a non-selective long-term approach, while the rest rejected this or envisioned only shorter-term or selective acceptance of any refugees from Ukraine.

An important finding in relation to this (non-)selectivity is that at least three-quarters of respondents claimed they would welcome displaced people at least for the duration of the war, regardless of the population group (people fleeing the war in Ukraine in general or specific displaced groups such as Roma from Ukraine, Hungarians in Transcarpathia, or African or Asian students from Ukraine). Beyond this consensus, however, there are differences: ethnic Hungarians from Transcarpathia are the only category of persons that most respondents would accept for longer than the duration of the war, while acceptance is much weaker for Ukrainian refugees in general, for Ukrainian Roma and students from African-Asian countries fleeing Ukraine.

Compared to the humanitarian support offered to Middle Eastern and African asylum seekers by civil society in Hungary in 2015 (Zakariás, 2016)—based on the active involvement of less than 5% of the adult population—the participation rate of 40% in June 2022 attests to an enormous increase. One pillar of this widespread active citizen involvement was the broad infrastructural palette of organisations, initiatives, and networks with low threshold entry points for helping. This was backed by calls for support from aid organisations and NGOs, public and state institutions, private companies, and ubiquitous media and social media activities in the first months of the war. The fact that half of those involved in helping referred to the organisational embeddedness of their acts and that involvement was found to be independent of respondents’ material-economic position (measured by savings) points to the density and diversity of such helping infrastructure. This led to abundant opportunities for people to translate their helping intentions into action.

Regarding the uneven distribution of civic engagement among specific groups of society, we found that active helping is heavily gendered and differs according to occupational background, with the substantial over-representation of the social, educational, and health sectors. In addition to the general over-representation of women among volunteers internationally, the strong interlinkages between refugee support, gender, and care sectors may partly be fostered by the unfolding ‘carefare regime’ (Fodor, 2022) that is promoted by the current authoritarian-populist government. This contributes to the severe overburdening of women with the unpaid duties of social reproduction and, in parallel, marginalises them into heavily underpaid and undervalued labour market positions, often in the (state and NGO) care sector.

In contrast to such a disproportionate distribution of solidarity acts according to gender and occupation, we detected a noteworthy difference with solidary attitudes, with the latter showing no gender differences and respondents in social service occupations being even less accepting than the sample average. This divergence of actions and attitudes may be partly connected to the salience of institutional-organisational infrastructures as much as by non-reflected, pre-intentional embodied norms, dispositions, and praxes. However, such a contrasting distribution of attitudes and acts of solidarity calls for further inquiry.

The gendered and occupational inequalities associated with helping have not changed compared to the post-2015 period, which was also characterised by a central role for care professionals, the dominance of female activists, and almost exclusively voluntary and unpaid engagement (Feischmidt & Zakariás, 2019). What did change, however, compared to 2015, is the attitudinal response to the current large-scale mobilisations: that is, the convergence of solidary attitudes, with 68% of respondents either involved in helping or claiming their readiness to be personally involved; and almost 70% of the population embracing the general duty of Hungarian society to universally accept people fleeing the war in Ukraine.

Such prevalence of openness and solidarity attitudes among the population can be linked to the lack of xenophobic discourse targeting people fleeing due to the Russian invasion and the recourse to humanitarian discourses by most politicians and the Hungarian media (Messing et al., 2022). Preferences for parties that delivered messages of humanitarianism are closely correlated with both solidary attitudes and practical involvement: supporters of left and liberal parties, as well as of the governing Fidesz/KDNP parties, showed stronger solidarity patterns than allies of the former radical-right party Jobbik, or the radical-right Our Homeland Movement which formulated explicitly xenophobic messages against the displaced from Ukraine.

Our survey data revealed a complex picture regarding the relationship between solidarity and political discourse in Hungary. On the one hand, the current political discourse coexists alongside formerly dominant ideologies and concepts, exerting long-term effects on everyday attitudes and perspectives. More specifically, the juxtaposition of the ‘deserving Ukrainian refugee’ and the ‘undeserving African/Asian/Middle Eastern migrant’ that characterises Hungary’s current official political language interacts with the racialising and xenophobic discourses prevailing in the Hungarian public since 2015. Such an effect may partly explain the non-solidary behaviour and attitudes of supporters of the Jobbik party, which has changed its alignment with xenophobic and racist political and public discourses only in recent years.

On the other hand, attitudes of selectivity and the rejection of solidarity proved to be closely associated with identification with war narratives promulgated by the Russian government—more specifically, emphasis on the culpability of the Ukrainian government and Western political actors and the notion of Russian victimhood. This approach is endorsed by current Hungarian governmental discourse, which frequently echoes the war narratives of the Russian government and discourages openness towards displaced groups and their welcoming. Comparison of the two-dimensional and multi-dimensional models of non-selective acceptance shows that the lower levels of universal acceptance of displaced people by the ruling party Fidesz-KDNP supporters, as well as of the radical-right opposition Mi Hazánk supporters, may be closely related to the endorsement of such war narratives that hold Western military actors and Ukraine responsible and perceive the victim as Russia and the Russian government.

In conclusion, our research presents an ambivalent picture of Hungarian society in the context of the war against Ukraine and the subsequent refugee-welcoming challenges. On the one hand, it highlights the exceptional momentum and mobilising power of civil solidarity in terms of expressed attitudes and practical involvement. At the same time, the results also reveal the limits and vulnerabilities of civil solidarity. First, in linking to research on populism, xenophobia, and racism (Wodak, 2019), our study highlights the volatility of civil solidarity: its exposure to populist political discourses that cherish or condemn moral economies of assistance according to vested interests. Second, our findings deepen the knowledge of how civic solidarity is embedded in a neoliberal reliance on citizens’ individual resources (disposable time and material means), salient inequalities in sharing the burdens of humanitarian support (differences in the scope of paid and unpaid involvement, and the disproportionate responsibilities of different caring sectors), and a shrunken state infrastructure for refugee protection in the country under study. All this is embodied in how the consensus and relative evenness of solidarity attitudes at the time of our survey was unevenly translated into practical help and burdened disproportionately those who were already heavily charged with care responsibilities. With this finding, we highlight the need to include the perspectives of care in inquiries that zoom in on political and social attitudes.

Finally, our research raises questions that warrant further exploration. Our inquiry does not address the perspectives and experiences of the recipients of solidarity support. Another important query concerns the temporality of solidarity: helping attitudes and practices change over time, as revealed by various forms of qualitative empirical evidence in our context. The fact that most refugees from Ukraine have left for other countries shows that the exceptional wave of solidarity cannot make up for a poor state infrastructure for refugee assistance. The civic ecosystem of care and solidarity cannot cater to the fundamental need of displaced people for educational, health, welfare, and income-generating provisions and opportunities in the medium and long term. However, the ecosystem may assist with short-term coping, albeit associated with abrupt displacement, uncertainties, trauma, and a sudden loss of human dignity. Future research may also dwell on the experiences of displaced people from Ukraine who remain in Hungary for various reasons or engage in circular mobility and of those civic helpers who do not give up their mission to care after the first wave of compassion and active solidarity. Their evolving relationships and linked moral economies should be a subject of further scholarly endeavours.