Between Charity and Protest. The Politicisation of Refugee Support Volunteers

This article examines how refugee support volunteers based in Britain and in France negotiate the boundaries between charity (or humanitarian) action and social activism since the 2015 ‘refugee crisis’. Scholarly literature has often separated charity and humanitarian action from social activism, as the former is seen as lacking the goal of social and political change that characterises the latter. The set of 147 in-depth interviews we conducted in different British and French refugee support charities and networks reveals the complex relationship between charity and protest. Through the focus on the moral dilemmas that participants encounter throughout their experience in the field, this article aims to highlight the ambivalences of their engagement as well as its transformative potential. Our analysis shows how participants develop new cognitive frames, emotions and interpersonal relations that transform their engagement and lead them to link charity/humanitarian action with broader objectives of social and political change. More generally, our analysis highlights the processes through which participants construct political narratives that aim to challenge state-driven policies and discourses of “migration management”. This article aims to contribute to the reflection about the informal character of the forms of participation analysed in this special issue, through the focus on the moral dilemmas and the “quiet” and “unexceptional” politics of volunteering.


Introduction
As an activity guided by distinct-often conflicting-values and interests, charity and humanitarian aid 1 is filled with dilemmas, ambivalences, contradictions and "impossible situations" (Malkki, 2015) that volunteers and professionals have to navigate to sustain their engagement (Eliasoph, 2016). This is the "messy middleground" in which charity and humanitarian work operates: it is situated at the intersection of distinct registers of action-emancipatory and punitive; critical and reproductive of existing power relations; autonomous and co-opted by public authorities; guided by ideals of solidarity and philanthropy (May & Cloke, 2014). In this article, through the example of participants engaged in the British and French refugee support sectors since the so-called 'refugee crisis' in 2015, we examine how volunteers deal with the moral dilemmas that they encounter in the course of their experience. We explore how they make sense of their own engagement by focusing on the political dimension of their actions and by analysing the (dilemmatic and ambivalent) processes of meaning-making happening throughout their daily practice in the field. The set of interviews we conducted in different British and French charities and networks reveal the complexity of the relationship between charity (or humanitarian) engagement and politics. Although participants evoke a distant relation to politics and other forms of engagement, such as social activism, the meaning that they give to their action can shift, leading to processes of politicisation whereby individual lifestyle changes based on the compassionate practice of charity action are perceived as closely connected to broader structural changes.
To analyse these politicisation processes, we take insight from studies that have looked at the moral contradictions, paradoxes, dilemmas and "impossible situations" of volunteering and humanitarianism (Eliasoph, 2016;Finnemore, 2008;Holden, 1997;Malkki, 2015), as well as studies that aim to capture the "eventful" and transformative dimension of collective action (Della Porta, 2008). Although the sociology of collective action has paid attention to moral values-for example, through the concept of moral shock (Jasper & Poulsen, 1995)-only few studies have analysed how these values are challenged. We argue that the focus on the moral "dilemmas" (Eliasoph, 2016), "paradoxes" (Montambault, 2016), "puzzles" (Lee, 2014), "tensions" and "contradictions" (Ganuza et al., 2016) of volunteers is a good vantage point to understand participants' (situated) processes of meaning-making. As they navigate these dilemmas and try to solve the moral puzzles that they face, they need to re-evaluate the values that motivated their engagement in the first place, and they have to make sense of the different-often contradictory-logics and interests that guide their action. We maintain that it is through this process of trying to find coherence and logic in their practice of volunteering that politicisation occurs. As we will develop below, this article aims to contribute to recent studies that have analysed new forms of engagement emerging in the context of the 2015 'refugee crisis' in Europe. More generally, in line 1 3 Between Charity and Protest. The Politicisation of Refugee… with this special issue, we aim to contribute to a broader reflection on informal forms of political participations, through the focus on the "quiet" and "unexceptional" politics of volunteering (Askins, 2015;Apter, 2017).
Our analysis is based on 147 in-depth interviews with British and French volunteers in the refugee support sector. 2 Our respondents are involved in professionalised and centralised organisations that were active before 2015, as well as more decentralised organisations and networks (especially those that emerged since the 'refugee crisis'). They dedicate to different forms of volunteering, for example, giving legal advice, providing emotional and therapeutic support, organising donations of food and clothes in the Calais 'jungle', hosting refugees and offering language courses. Our sample is composed of volunteers with different profiles (in terms of age, gender and socio-economic backgrounds). However, reflecting the composition of the sector, a large part of our respondents were women, retired and self-identified as white and middle-class. The interviews were conducted between May 2017 and November 2019. The moral dilemmas and the politicisation processes that we analyse were identified inductively, in particular through the responses to questions on respondents' personal trajectories, the values that motivate their engagement, their relationship with refugees and their description of difficult situations that they encountered. We will show how the majority of our participants (86 out of 147) undergo processes of politicisation throughout their experience in the field. Beyond the magnitude of the phenomenon (which is difficult to quantify), we are especially interested in the processes through which politicisation occurs, in particular the situations in which they emerge and what changes they entail. 3 These processes could be observed across all individual characteristics, places and organisations in which we conducted our fieldwork. We nevertheless observed that politicisation was more frequent and prominent among participants involved in more informal forms of volunteering. Indeed, in contrast with professionalised and routinised activities (often taking place in the charities' offices), the more informal experience of volunteeringfor example, hosting refugees at home or helping in the camps in Calais-lead participants to encounter dilemmas more regularly as they often face unexpected situations for which they are not fully prepared. Also, as we have shown elsewhere (Monforte et al., 2019), politicisation processes are more discernible in the French 2 The interviews were conducted in London, Birmingham, Sheffield and the Midlands in the British case, and in Paris, Nantes and Calais in the French case. In Calais, we interviewed both French and British volunteers. 3 We used the NVIVO software to code our data systematically and perform a thematic analysis of the interviews. For this paper, we focus on the different types of dilemmas that we identified, and we relate them to the different ways they give meaning to their engagement, looking especially at the perceived distinction between humanitarian and politicised engagements. The politicisation processes that we analyse relate to how our participants make sense of the way their engagement has evolved over time, therefore underlining temporal changes. Although our data consist of one-off interviews conducted at a specific point in time (we did not conduct follow-up interviews), we asked them to reflect upon whether and how the meaning that they attach to their engagement has shifted over time. Moreover, we were able to trace temporal shifts by constrasting their responses to questions about their initial motivations for joining refugee support charities with their narratives about their experience of volunteering. We provide a more detailed presentation of our research methods in our project's final report (Monforte et al., 2019). case, where the 'refugee crisis' has been more visible in the public space. Although we acknowledge these differences, this article aims to analyse, at a more general level, the mechanisms through which politicisation occurs rather than to compare different settings and national contexts.
In the first section of this paper, we link the literature on charity and humanitarian action with the scholarship on politicisation, referring to studies that have analysed the 2015 wave of refugee support in Europe. Then, turning to our empirical findings, we highlight the moral dilemmas that our respondents faced throughout their practice of volunteering, as well as the cognitive, emotional and relational processes of transformation that emerge from their narratives, showing how they politicise their engagement.

Charity Action, Politicisation Processes and the Refugee Welcome Movement
Scholarly literature has often separated charity and humanitarian action from social activism, as the former is seen as lacking the goal of social and political change that characterises the latter (Anheier & Scherer, 2017). The central objective of charities and humanitarian organisations is not to critically challenge political institutions and public policies, and they do not necessarily aim for broad social change (Wilson, 2012). Consequently, in contrast with social activism, volunteering and humanitarian action are often defined as a-political or non-visibly politicised forms of engagement (Ekman & Amna, 2012;Verba et al., 1995). This discussion is closely linked to the analysis of politicisation processes. According to Rancière (2001), politicisation occurs when claims challenge state authorities and established social relations (the police) in the name of equality. Politicisation does not occur spontaneously though: to become political, collective actors need to identify inequalities and the contradiction of the police. This implies "to step out of the existing order of things and judge situations against standards and by values suppressed or inconceivable in the immediacy of the situation" (Uitermark & Nicholls, 2014: 974). In particular, politicisation is based on "public-spirited" forms of discussion and engagement which open up the possibility for deliberation and disagreement (Eliasoph & Lichterman, 2010;Hamidi, 2010).
From this perspective, critical studies on charity action and humanitarianism argue that these forms of engagement obstruct politicisation processes (Cloke et al., 2017;Eliasoph, 1998;Fassin, 2012). In contrast to more visibly politicised engagements, charities and humanitarian organisations tend to avoid "making things political" (Eliasoph & Lichterman, 2010) as they focus on concrete immediate solutions to individual problems through service provision. As they are based on a "privatization of political responsibility" (Williams et al., 2016(Williams et al., : 2293, charity and humanitarian actions do not address the system that produces the issues they aim to address, and so deligitimise broader criticisms of systemic injustice (Poppendieck, 1999). For instance, Blau (1992: 5) argues that charities are guided by "an ideology of individualism, self-reliance, and minimal government" which distracts individuals from the objective of social change and therefore maintain the status-quo (see also Theodossopoulos, 2016). From a similar perspective, Ticktin (2017) describes humanitarian action in the field of migration as an "antipolitics of care": it tends to reproduce inequalities rather than aiming at transforming the broader system on which they are based. This relates to what Rosanvallon (2008: 22) defines as the unpolitical [l'impolitique]: "a failure to develop a comprehensive understanding of problems associated with the organization of a shared world". More generally, as they are based on charitable acts of giving and emotions of compassion and pity, volunteering and humanitarian action tend to reproduce asymmetrical power-relations between aid-providers and recipients (Berlant, 2004;Vitellone, 2011).
As Fassin (2012: 87) argues, the ideologies and modes of intervention of charity and humanitarian actors are embedded in a more general transformation of governance through a "humanitarian reason", a "new moral economy that values suffering over labour and compassion more than rights". On the migration issue, this is reflected in policies that aim to assess migrants' suffering through their vulnerability and trauma (Fassin & D'Halluin, 2005). Processes of depoliticisation of migration issues are also at play when governments address population movements and border controls through technical considerations and the general idea of management (Cuttitta, 2018;Pécoud, 2015). More generally, these framings in terms of humanitarianism and management reify restrictive migration policies that mix compassion and securitisation (Huysmans & Buonfino, 2008). As shown by Aradau (2004) and Cuttitta (2018), this shows how concerns about "pity" and "risk" complement each other and create a depoliticised framework that "presents policy-making as a neutral, necessary and indisputable process, in which the possibility to choose between different political (not merely technical) alternatives, as well as that for disagreement and contestation, is limited or denied" (Cuttitta, 2018: 634).
Recently, scholars have maintained that this argument about the depolitcising nature of charity and humanitarian action should be challenged (Cloke et al., 2017;Monforte, 2020). In particular, the political dimension of hybrid forms of engagementfocusing, for example, on lifestyle politics, forms of everyday resistance, solidarity and political bottom-up experimentations-should be investigated (see the introduction to this special issue). Exploring refugee support charities, recent studies have shown how parts of the voluntary sector are transforming in the context of the 'refugee crisis', and how the perceived boundary between volunteering (or humanitarianism) and more politicised forms of engagement such as protest is blurred and can shift over time. Most of the civil society organisations, networks and individuals that support refugees arriving in Europe since 2015 do not address state institutions directly through claims-making. Instead, they aim to embody a welcoming culture and perform ideas of hospitality and feelings of compassion (Sirriyeh, 2018;Maestri & Monforte, 2020). However, this does not mean that this movement should be seen as apolitical (Fleischmann & Steinhilper, 2017;Fontanari & Borri, 2017;Sandri, 2018). For instance, as shown by Sandri (2018) and Stierl (2018), volunteers active in liminal spaces such as Calais and the Mediterranean disrupt border control policies through their presence and construct forms of engagement that are "alternative to formal humanitarian aid" (Sandri, 2018: 65). Looking at the case of Greece, Rozakou (2017) shows that refugee support initiatives bridge logics of humanitarianism and solidarity.
Similarly, Vandevoordt and Verschraegen (2019) and Schwiertz and Steinhilper (2021) define refugee support initiatives as "subversive" or "strategic" humanitarianism as they are based on the opposition to the anti-immigrant political climate and aim to achieve a broad consensus for more progressive policies. Focusing on the Italian case, Sinatti (2019) and Zamponi (2017) show how some participants have become increasingly critical of actions that were exclusively oriented towards providing emergency support to refugees. Also, as developed by Steinhilper and Karakayali (2018), refugee support volunteering opens up new spaces of encounter through which participants develop critical arguments against restrictive migration policies. More generally, Della Porta (2018: 344) argues that the wave of refugee support in 2015 aims to "challenge the very definition of borders and citizenship".
Overall, as Agustín and Jorgensen (2019) and Schwiertz and Schwenken (2020) maintain, the refugee solidarity movement is a good vantage point to examine new forms of solidarity and their transformative potential, in particular through the focus on the types of political subjectivities that it creates. Moreover, as shown by Tazzioli and Walters (2019), this movement allows us to highlight the multiplicity of practices (some more disruptive than others) that the notion of solidarity encompasses, as well as their contradictions. This article contributes to this scholarship by focusing on the moral dilemmas that participants encounter in their daily practice of volunteering, looking in particular at the narratives through which they give meaning to their engagement.

"The Hands-On Stuff Is Something Else": Volunteers' Distant Relation to Politics
The analysis of participants' interviews reveals that they often distantiate their engagement from political motivations. Most participants did not consider their practice of volunteering as an explicit political act, in the sense that it was not directly related to broader critiques of social inequalities and demands for social or political change. Despite being critical of current immigration policies, they argued that the main purpose of their activities was to make concrete changes to the lives of refugees rather than to address public authorities and governments. In their own words, they are "here to help".
In their narratives, participants refer to four main values to present the reasons why they engaged in the refugee support sector: empathy, pragmatism, neutrality and good citizenship. They emphasise the needs of refugees and the necessity to understand their pain; they stress their willingness to make concrete and immediate changes to alleviate their suffering; they argue that their engagement goes beyond partisan opinions and that they are respectful of public authorities. These moral values are central in the narratives of participants (and of charity and humanitarian actors more generally -Eliasoph, 1998;Malkki, 2015), and they are presented as being closely linked to the choice of volunteering as a form of collective action. Indeed, it is through their "acts of compassion" (Wuthnow, 1991) that they aim to assert and demonstrate their beliefs and moral values. From this perspective, more politicised forms of engagement are generally perceived as connected to different-sometimes opposite-values. This is illustrated by Jean, a participant who has been volunteering in a hosting network in Paris for 3 years. When asked about whether he sees a political dimension to his action, he makes a distinction between the act of "helping" and political activism, arguing that the latter goes against his values of neutrality and good citizenship: When we say "helping", there are things that should be natural, which I don't really consider as activist actions. (…) I think that political activism is not very… I don't want to sound disillusioned, but they are either naïve people who believe in utopian societies or people who have personal interests, not interests for the common good. (Jean,68,Paris) Similarly, Jane who is active in a hosting network in London, argues that volunteering allows her to have an immediate impact that other-more public-forms of actions such as lobbying or campaigning cannot achieve (she explains later in the interview that she sees her engagement as "private"): We decided that we weren't going to be lobbyists and campaigners. We are starfish flingers. You might not have come across this term so I will tell you the story. So, there's this boy on a beach and the tide washes in and it leaves millions of starfish stranded on the beach, flapping around and dying. And he flings them back. And a man who wonders alone says "This is completely futile, there are millions of starfish here, what can you do, you're making no difference". And the boy flings a starfish back and says "I'm making a difference to that one", he flings back another one, "I'm making a difference to that one" etc. (Jane,42,London) It should be noted that, in contrast with these positions, some participants could sometimes be involved in what they present as more visible political activities-for example, attending demonstrations or meetings with local political representatives. These participants described their engagement as being motivated by more explicit political reasons (referring, e.g. to social justice and anti-racism). However, like the other respondents, they also distinguished between these forms of action and their more "concrete" charity actions. For example, this is the case of Denise, who has been engaged in a hosting network in London for two years. When asked about her previous experience in charities and/or social movements, she explained that she has been involved in activism, signing petitions and going to demonstrations ("It's a way of making your voice heard"), but that her engagement in her charity was distinct: "the hands-on stuff is something else" (Denise, 78, London).
As these examples suggest, participants justify their engagement through values that can be linked to processes of "evaporation of politics" (Eliasoph, 1998). Values of empathy, pragmatism, neutrality and good citizenship are apolitical (or depolitcising) because they often reproduce a "humanitarian reason" (Fassin, 2012) through which those who are supported are seen as vulnerable and in need of help. They are also apolitical because they evoke discourses that view migration through the perspective of management and through technical considerations (Cuttitta, 2018;Pécoud, 2015). In particular, the value of pragmatism can lead them to present the situation through the lens of a 'refugee crisis' and a perspective that focuses on the tangible and visible results of their actions (Malkki, 2015). 4 More generally, these values are apolitical because they often lead participants to avoid opening up the possibility for disagreement (Eliasoph & Lichterman, 2010). As Redfield (2011: 53) notes in his reflection on humanitarianism, neutrality (as a value and an ethos of action) is "an 'impossible' or negative form of politics".

The Moral Dilemmas of Volunteers: How Values of Neutrality, Good Citizenship, Pragmatism, and Empathy Are Challenged
Further analysis shows that the way participants give meaning to their own engagement is destabilised throughout their experience of volunteering. Once they become fully involved in the field, volunteers often face "impossible situations" (Malkki, 2015) in which moral dilemmas emerge. These situations are unexpected and experienced as "impossible" because they cannot be easily apprehended through the moral values that motivated their engagement in the first place. They are lived as personal tests 5 and they are often emotionally difficult, leading sometimes to episodes of burnout (Maestri & Monforte, 2020). Four moral dilemmas-each related to different situations-can be evoked. They resonate with what Finnemore (2008) describes as the "typical dilemmas" of humanitarian action, as well as with the dilemmas that emerge in empowerment projects (Eliasoph, 2016). They can be presented in turn, through general questions: -How to remain neutral in the face of adversity?
The first dilemma relates to situations in which participants encounter adversity and hostility towards their engagement. These situations can happen in their immediate social circles (i.e. in their family or at work). For example, this is the case of John, a volunteer who has been actively engaged in Calais for a year, doing regular trips to donate food and clothes and to help with the organisation of the camp. When asked how he presents his engagement to his friends and family, he explains that his actions were met with overt condemnation: My parents, especially my father, he has basically decided that I'm a traitor, I think. I think in his kind of Daily Mail view of the world I've become a terrorist sympathiser (…) By helping refugees that means you're a terrorist sympathiser. (John,45,Midlands) For many participants, the experience of adversity and hostility is related to a more general observation about rising racism and intolerance in society, for example, expressed through the Brexit vote in Britain and the Front National in France. As Abdi explained, this context makes these issues impossible to avoid "on a personal level" (Abdi, 69, London). This is illustrated by Marie, a participant who has been volunteering in Calais for 2 years. When we asked her a general question about how her engagement is perceive by people around her, she explains the difficulties of talking about it: It's difficult to convince. You know that in [location], this is where more than 50% of people vote for Marine Le Pen, so it is anti-immigrants. (…) In my group, we try to organise meetings, discussions, etc. It is very difficult. Politically, we can't see any solution with them. (Marie, 75, Calais) These expressions of hostility towards what participants represent are experienced as a moral dilemma because they challenge their neutral stance. Thus, in many interviews, our respondents explain how they feel conflicted about whether they should be vocal about their engagement and confront directly adverse opinions (Monforte & Maestri, 2021). This is, for example, reported by Helen, who volunteers in Calais and who was criticised by a friend for helping people who "should go back to where they come from". When asked about how she responded to this hostile opinion, she explains that her reaction was to question her own "polite" position (she explained earlier in the interview that she likes to be "quiet" about her engagement): "I thought oh my God, how am I gonna sit here and have lunch and be polite for the next hour?" (Helen, 77, London).
-How to remain neutral and a good citizen when governments' policies are experienced as illegitimate?
For many of the volunteers we interviewed, the value of neutrality is constructed in relation to state policies and, more generally, the legitimacy of public authorities. Although they are often critical of current immigration policies, they argue that it is not their responsibility to challenge the government and that they do not have the expertise or the capacity to do so. Often, the value of neutrality is connected to values of good citizenship: in the way they present their engagement, participants are keen to show that they are respectful and law-abiding citizens, whose actions are not disruptive. However, in the course of their experience in the field, they are often confronted with situations in which these values are tested. This is the case when they witness the arbitrariness and violence of immigration policies, through their encounters with home office representatives and with police forces. These situations raise a significant moral dilemma: they lead participants to ask themselves whether they can remain neutral, good citizens, in the face of injustice. In some cases, these situations raise an even more difficult dilemma, which is implied in the second example below: does being neutral mean that they have become indulgent (maybe complicit) towards policies they disapprove?
These situations and related dilemmas are exposed by Vincent, a participant who has been active in Calais. When asked whether he experienced difficult situations, he evokes a volunteer who was convicted for a tweet about police brutality. He explains how this event raised questions about his neutrality and attitude of respect towards authorities, leading to his frustration and anger: It was revolting actually. Knowing that it was a volunteer who was very active in [organisation], I thought that this was a political decision really. (…) I thought that what he did wasn't a militant action, it was (…) about freedom of expression. It [the conviction] was a bit excessive, it is difficult to comprehend actually. (Vincent,28,Strasbourg) In another interview, Jackie, a participant engaged in a charity that aims to provide moral support to refugees in detention centres described a different situation that raised a similar dilemma. As she evoked her reaction when she visited a detention centre for the first time, she evoked how she was surprised by the "friendly" attitude that the guards displayed towards visitors: "the staff are very friendly to us as external people. Like, chattier than I'm comfortable being with someone who's made those decisions about their life". She explained how their attitude made her feel "really, really uncomfortable" as it made her question her own role: how can she have friendly interactions with people who are "desensitised" about working for the migration detention system? (Jackie, 37, Midlands). As these examples suggest, participants experience moral dilemmas when they face situations that are unexpected or, as the last respondent explains later in the interview, they "hadn't really thought about before". These situations raise feelings of "revolt", "incomprehension" and "discomfort" and, ultimately, lead them to question their own motivations. As the next examples will show, dilemmas can also emerge when volunteers encounter situations that are different from what they were prepared for.
-How to be pragmatic when things cannot be managed?
Throughout the interviews, participants refer to values of pragmatism and modesty to justify their engagement. They argue that they want to make concrete changes to the life of refugees and that they do not aim for broader structural changes. Accordingly, they often present their engagement through the tangible outcomes that they want to achieve, in particular in the context of what they perceive as a humanitarian 'refugee crisis'. The literature shows that these aspects are fundamental in the discourses of volunteers and humanitarian actors (Malkki, 2015). They often present their actions as short-term projects that need to be managed and in which their personal gratifications take the form of visible and often quantifiable outputs (Krause, 2014). To refer to the image used by one of our participants above, this is about "making a difference" to the "starfish stranded on the beach".
Similarly to the values of neutrality and good citizenship, this pragmatic stance is often challenged in the field. In particular, many respondents question their role because their action is not having the impact they expected: they do not feel that they are "making a difference". Their moral dilemma can be summed up through the following question: what am I achieving when things seem to be impossible to manage? This dilemma is often related to the idea that they are acting in the context of a crisis that is getting worse, which raises feelings of hopelessness and frustration. This is, for example, the case of Lucy, a participant who has volunteered in Calais and Dunkirk for two years. When asked about what she found most challenging in her experience, she evokes immediately her feelings of being "mentally and emotionally drained": And you run out of everything and you can't give them a pair of shoes because you've ran out and they're disappointed, not angry. (…) It's incredibly difficult and mentally and emotionally draining. (…) And there's not a lot of positives in this field at all. It's always bad news, it's always rubbish and it's never ending. And the sheer scale of the task is sickening. (Lucy,31,Midlands) As this example suggests, feelings of hopelessness raise a difficult dilemma for volunteers as they raise questions about the usefulness of their action, especially in a situation that does not seem manageable ("the sheer scale of the task"). This is echoed by Luke, who has been volunteering in a London-based charity for more than a year. Although his experience is very different from that of Lucy (he volunteers in an office, providing legal support to asylum-seekers), he describes similar feelings, for example, when he evokes the thousands of people "not getting the help": I just feel very like hopeless. 'Cause I can try and help this one person, however I know that there is out there, like out there, there is thousands of young people in the exact same situation. And those 99,999 people not getting the help. So yeah, I think it's just very, it is really frustrating. (Luke,26 London) These examples show how feelings of hopelessness and frustration (related to the general negativity in the field) push them to question their own role and motivations. As Claire explained when talking about her experience in Calais, this made her question why she is volunteering. When responding to a question about the emotional difficulties linked to her engagement, she argued that she found it difficult not to feel like a "volunteer tourist" and that she often asked herself: "Why do I go? What do I do when I'm [there]?" (Claire, 27, Calais).
-How to be compassionate when we make distinctions between different groups?
Finally, as we have shown elsewhere (Maestri & Monforte, 2020), moral dilemmas emerge when values of compassion and empathy are tested, especially in situations in which participants interact with refugees whose character is different from what they expected. These situations lead them to question the fundaments of their engagement as they raise difficult questions about the limits of their compassion. This moral dilemma can be summed up through the following question: who am I to judge who deserves my compassion? This can be illustrated, for instance, by Jackie, the volunteer we presented above who provides emotional support to refugees in detention centres. During the interview, as she describes her relation to refugees, she evokes her feeling of "guilt" when she instinctively distinguished between deserving refugees (she evokes later in the interview the image of a young man "who was designed to make you feel good about yourself") and undeserving others (referring to refugees who committed crimes): The other thing that I've had to think about a bit in myself is when the person that you're working with is somebody that in another situation you wouldn't necessarily choose to be friends with. Somebody who has done things, or has views or whatever that wouldn't usually be the things that you would be comfortable with. (…) I think it's just made me question myself a little bit about… I suppose guilt about having an instinctive reaction and making judgements that are not fair judgements to make. (Jackie, 37, Midlands) As this participant highlights, this moral dilemma emerges when volunteers are in situations in which they question their position and the power that is given to them as the bearers of compassion. This shows that values (and feelings) of compassion and empathy are often linked to processes of distinction, inclusion/exclusion and hierarchisation (Nussbaum, 1996). Similar dilemmas can emerge when volunteers realise that their organisation excludes some groups from their services. For example, this is the case of the organisations that provide services to asylum seekers and refugees but not to destitute asylum seekers or undocumented migrants. Participants who volunteer in these organisations can face situations in which they are not able to help people in need because they are not in the "adequate" legal category. This raises the following question: How can I act according to values of compassion if my engagement is selective? This is the case of Marianne, a volunteer active in a charity providing legal support to refugees in Nantes. When we asked her whether she is critical of some aspects of the work done by her organisation, she explained that she did not understand why her charity could not help homeless refugees (it only supported refugees that were referred by the local council): In the beginning, the organisation was telling us 'this is not the core of our action'. And I would tell them: 'it doesn't bother you that some people live in the street, just across the road, and they don't answer to you because they are completely exhausted?' (Marianne, 45, Nantes) As these examples illustrate, participants' experience of volunteering is often made of difficult and unexpected situations that challenge the moral foundations of their engagement and raise significant emotional strains. They often evoke feelings of hopelessness, guilt and despair, and they talk about episodes of burnout that they have experienced personally or that they have witnessed. To sustain their engagement, participants must learn to navigate these moral dilemmas and deal with the negative emotions that they generate: they must find ways to make their experience "meaningful" again (Florian et al., 2019). In other words, they must find ways to respond to the question asked by one of the participants above: "Why do I go?". As we will develop now, this can be analysed through the transformative dimension of their engagement.

3
Between Charity and Protest. The Politicisation of Refugee…

"Making a Statement": the (Quiet) Politicisation of Volunteers
In the interviews, participants present the moral dilemmas that they encounter as transformative. They can be turning points that quickly change the bases of their engagement, or they can be triggers for slower incremental shifts. As we have shown elsewhere (Maestri & Monforte, 2020), participants can deal with these dilemmas by readjusting their practice of volunteering. For example, some decide to focus on more routinised and physical tasks in order to avoid difficult situations. Also, they can decide to re-orient their activity within a more managerial framework, in order to shift the responsibility of their action to their organisation. Finally, some respondents decide to leave their charity and dedicate to more clearly politicised actions. 6 In what follows, we want to focus on the meaning that they give to their engagement (rather than on their practice). Overall, the analysis shows that volunteering is "eventful" (Della Porta, 2008): throughout their experience, participants go through cognitive, emotional and relational changes that give a new-more politicised-meaning to their engagement. 7 To overcome the dilemmas we just presented and to make their experience meaningful again, they justify their engagement through references to new emotions (outrage), frames (linked to social change) and (more diversified) networks which bring them closer to politicised engagements based on notions of social justice and equality (Rancière, 2001).
The emergence of emotions of outrage can be illustrated by the case of Mary, a London-based participant who visits refugees in detention centres. Although she explained earlier in the interview that the value of neutrality was central to her, she argues that her experience raised feelings of indignation and outrage over time, especially due to the increased awareness of the climate of hostility towards refugees and the injustice of migration policies (the first and second dilemmas above): So, there was the feeling that if you were part of [name of the organisation], you were being made aware of what was going on, and it wasn't easy. The difficulty was, you know, getting very indignant about what was happening. So, it's the difficulties that, if you're part of [name of the organisation], you can't ignore these things. It's the politics of this country. (Mary,75,London).
Similar processes can be observed in relation to the cognitive dimension of their engagement. As shown by Della Porta (2008: 41), activists "appropriate, transform and transmit knowledge" in the course of protest. In the same way, we observe that volunteers adopt new knowledge in the course of their experience. For example, this is the case of Jeanne, who has been volunteering for more than 15 years in a charity that provides legal guidance to refugees in Paris. When asked how her engagement has evolved over time, she evokes how the growing perception of immigration policies as illegitimate ("the failure of migration policies") challenged her neutral and pragmatic stance and opened "another perspective" in her organisation as well as in her personal engagement. In particular, this new perspective brought her closer to ideas that she used to consider "idealist": This idea [opening the borders] came up about fifteen years ago. And we felt things were changing. (…) We are in another perspective today, this discourse becomes more accepted at the [name of the organisation]. In the beginning, opening the borders was like: 'Okay you are really an idealist!' But today, the failure of migration policies is so visible that we tell ourselves: 'But it's true actually'. (Jeanne, 60, Paris).
Finally, participants engage in relational transformations as they construct linkages with new actors, thereby changing the nature of their networks. For example, this is the case of Marianne, the participant who evoked a dilemma about the selective nature of her action because her charity could not help homeless refugees. When we asked whether her practice of volunteering had changed over time, she explained how she created new networks outside her charity, leading her to re-define her own engagement (she started an informal hosting network in Nantes): I designed some tools so that we could monitor things. I started to get in touch with lawyers (…) I also became a member of the Facebook groups used by the [name of the organisation]. So, when some situations were a complete disaster -for example single mothers being homeless with a baby -I used Facebook to find them a host family. And after a while, some new people came. And we even designed a new training with these new people for the rest of the organisation. (Marianne,45,Nantes).
These examples show that volunteers re-evaluate and transform the meaning of their engagement throughout their experience in the field. These transformations are the result of learning processes (e.g., when participants learn about specific aspects of migration policies), of self-criticisms (e.g., when they challenge exclusionary processes in their actions) and of specific encounters (e.g., when they construct new networks). Also, these transformations can be performed hesitantly or embraced more enthusiastically by volunteers depending on the situation in which they find themselves. Finally, they can be enabled by the culture of their organisations, or they can be the result of a disagreement between volunteers and their charity.
Although they still distantiate themselves from more visibly politicised forms of engagement such as social activism, these changes lead participants to present their actions as a form of "quiet" (Askins, 2015) or "unexceptional" (Apter, 2017) politics that often passes unnoticed, yet is meant to challenge the established social order (Bayat, 2010). This politics is based on everyday acts that aim to embody values of social justice and equality and, in doing so, concretely and symbolically challenge restrictive migration policies. Thus, rather than through radical acts of rupture and demands for social change (in line with Rancière's reflection), the quiet politics of volunteers emerges through more mundane, everyday, acts that take a symbolic and expressive meaning (Yates, 2015;Monforte, 2020). This is, for example, illustrated by Claire, who has been volunteering in Calais for 3 years, and who argues that it is through her actions that she "makes a statement". During the interview, when discussing the political dimension of her engagement she explains that, although her actions are not "dramatic", the meaning attached to them suggests that she is "symbolically tak[ing] a stand": I think due to Europe's response, due to Britain and France's State responses to it, you know, you make a stand by going. And, and that's unavoidable because actually that is given to you. You know you could go for completely benign reasons. You could be going just because you have always done this sort of thing, you have always fed the hungry, you have always worked with the homeless and you just see this as a continuation of that. But nobody else would. (…) In some ways you make a statement. I mean you book a ferry ticket, you go through customs, you travel to another country, I mean it's twenty miles across a bit of water, it's not that dramatic. But kinda it's like you know, symbolically you take a stand by going. (Claire, 27, Calais).
The quiet politics of volunteers is often based on personalised lifestyle changes (Lichterman, 1996) that allow for the construction of alternative political subjectivities, in a context in which these changes are seen as contentious and subversive. As Claire puts it in this last extract, these changes are often felt as "unavoidable", as something "that is given to you", not something that our respondents have strategically chosen. In the context of increasingly restrictive immigration policies, widespreading of racism and the construction of an exclusionary citizenship, daily acts of compassion and kindness towards refugees are given a new meaning: they become the expression of a political subjectivity that challenges dominant politics around migration.

Conclusion
The analysis of volunteers' narratives about their motivations for-and practice ofgiving support to refugees shows how the meaning given to their own engagement can transform as a result of their experience. Although most participants present the motivations for their engagement through depoliticised narratives, the analysis of their trajectories and everyday practice of volunteering shows how they can shift towards critical political subjectivities. Part of what makes collective actions "eventful" (Della Porta, 2008) is the set of unexpected-often ambivalent-situations that participants encounter throughout their concrete experience. These difficult situations raise dilemmas that challenge the moral values on which they initially based their engagement. In particular, values of empathy, pragmatism, neutrality and good citizenship are troubled by situations in which they cease to be meaningful. To make their experience meaningful again, many volunteers engage into cognitive, emotional and relational transformations through which they re-define their actions as a matter of social justice and equality.
The focus on the transformative dimension of collective action reveals the significance of hybrid engagements that blur the distinction between charity and humanitarian action on the one hand and more visibly political engagements such as protest on the other (Bosi & Zamponi, 2020;Monforte, 2020). This shows that the "messy middle ground" (May & Cloke, 2014) of charity and humanitarian action does not necessarily constrain the construction of critical political subjectivities. Moreover, the focus on the moral dilemmas faced by participants throughout their practice of volunteering underlines the specific situations, spaces and encounters through which these hybrid engagements emerge. We have stressed that volunteers politicise their engagement as a result of unexpected situations which often emerge in liminal spaces and through unpredictable encounters (Monforte & Maestri, 2022), highlighting how the fuzzy and often ambivalent nature of acts of compassion is an essential dimension of the processes through which volunteers make sense of-and justify-their engagement (Eliasoph, 2016).
Overall, these findings resonate with studies that stress the fluid and hybrid nature of solidarity actions in the Refugees Welcome movement (Fleischmann & Steinhilper, 2017;Schwiertz & Schwenken, 2020). More generally, our analysis contributes to the discussion about the significance of "quiet" or "unexceptional" forms of politics in contemporary societies (Askins, 2015;Apter, 2017). We have shown that our participants construct a non-demonstrative, yet expressive, engagement which falls into the realm of the "microphenomenology of political life" (Apter, 2017: 4), in particular when their daily acts of compassion and kindness are presented as concrete embodiments of social change. Recent studies have shown how alternative political subjectivities can emerge not only through visible, exceptional actions that aim for a rupture of the established social order, but also through more mundane, non-spectacular, actions that fall under the radar of contentious politics (Bayat, 2010). As our focus on dilemmatic situations shows, the emergence of these alternative political subjectivities is not necessarily the result of conscious, strategic, decisions. It can be the outcome of complex cognitive, emotional and relational processes through which individuals try to find solutions to impossible situations and reassess the meaning given to their engagement. Thus, we echo the call in this special issue for an in-depth analysis of the transformative processes which are rooted in the ambivalent and eventful practice of collective action.