Separating Families, Recuperating the “Nation-as-Family”: Migrant Youth and the Cultural Politics of Shame

This study investigates the intersections of policy, affect, and the lives of migrant youth. We approach the Trump Administration’s contingent reversal of a “zero tolerance” family separation policy as an illustrative case for understanding how affect mediates policy-making processes. Combining Critical Policy Analysis (CPA) and affect studies, we analyze 184 print media texts between the declaration of zero tolerance (May 2018) and President Trump’s repeal of his executive order (June 2018). We argue that mainstream media invited publics to sympathize with migrant youth and shame zero tolerance policy and its defenders. While shame catalyzed nationwide #KeepFamiliesTogether protests, it also animated political actions that recuperated “America” as a tolerant nation (e.g., “Love, not hate, makes America great”). In doing so, Education Policy Analysis Archives Vol. 28 No. 84 2 shame suppressed structural critiques of U.S. state violence toward migrant as well as Black, Indigenous, and minoritized families and youth. We conclude by discussing how a “pedagogy of discomfort” offers one way to build toward more historically responsive and intersectional coalitions for migrant and education justice.


Introduction: Whose Families Belong Together?
It's about keeping families together, while at the same time, being sure that we have a very powerful, very strong border. (U.S. President Donald Trump, June 20, 2018) On June 20, 2018, President of the United States, Donald Trump, issued an executive order that temporarily repealed a "zero tolerance" immigration policy. The policy criminally prosecuted all undocumented migrants and asylum seekers crossing into the U.S. and detained adult migrants in criminal facilities separated from their children (Kandel, 2019). The Trump Administration's surprising reversal of federal policy coincided with an uncommon showing of public unity opposing zero tolerance. Republicans, Democrats, Evangelicals, Methodists, the Business Roundtable, humanitarian organizers, and immigrant rights and advocacy organizations collectively condemned the Trump Administration and its policy of separating families. This national outcry culminated in the largest single fundraiser in Facebook history, raising over $20 million for Texas refugee nonprofit RAICES (Williamson & Nixon, 2018). Amid an historic moment of deep social and political divisions, how might we explain this demonstration of national unity and the decision of an uncompromising administration to shift their policy stance on immigration?
This study approaches the Trump Administration's repeal of zero tolerance as an illustrative case for exploring the role of emotion in policy-making processes, particularly amid "post-truth" contexts in which affective appeals, at times, outweigh scientific evidence (McIntyre, 2018). We build on recent efforts to trouble the narrow boundaries concerning what "counts" as education policy (Anyon, 2005, p. 66) and consider how cultural-political debates about borders and migration have direct, material consequences for 1.2 million refugee youth attending U.S. public schools and roughly three million refugees that have resettled in the US since 1975 (Koyama & Chang, 2019). Combining Critical Policy Analysis (CPA) and scholarship on emotion and affect, we interpret policy makers and policy texts broadly to include journalists and the textual artifacts they author and authorize (Diem, Young, & Sampson, 2019). We wanted to understand how media texts invited "publics"-diverse cultural political groups within a digitally-mediated public sphere (Varnelis, 2012)-to form "impressions" of migrant families and youth (Ahmed, 2004, p. 6). We asked: How did major news media outlets invite publics to feel in relation to migrant youth and zero tolerance policy? What emotions animated widespread political actions, such as charitable donations and nationwide protests?
We argue that mainstream media moved publics to sympathize with migrant youth through impressions of zero tolerance as a policy of incarcerating crying children in cages. Such images catalyzed feelings of shame for America and mobilized nationwide protests in opposition to zero tolerance. Yet, shame simultaneously recuperated ahistorical narratives of America as a land of hope and opportunity (El-Haj, 2010). Protest cries such as "Immigrants Make America Great" decoupled movements against zero tolerance from past and ongoing forms of state-sanctioned violence enacted toward Black, Indigenous, and minoritized communities. These findings motivate our interests in better understanding affect as a social resource for mobilizing intersectional coalitions for migrant and education justice. We conclude by discussing how a "pedagogy of discomfort"teaching and learning practices that critically and reflexively interrogate our emotional investments in narratives of self and nation (Zembylas & Boler, 2002)-might interrupt assumptions about tolerance as a sufficient response to zero tolerance. Skattebol and Hayes (2016) remind that affective relations are "underpinned by a myriad of collective legacies and social relations" (p. 9). Accordingly, we situate our analyses of affect and zero tolerance within broader legacies of U.S. family separation policies. 1 Wary of blurring disparate histories, lived experiences of, and ongoing resistance to state violence, we see the task of mapping a history of state violence and family separation as essential for understanding present affective responses to zero tolerance policies.

Building a Nation on Family Separation
To organize this review, we turn to Lakoff's (2006) notion of frames, or cognitive and cultural maps of meaning-making. According to Lakoff, the "nation-as-family" frame represents an enduring cultural framework and is evident in taken-for-granted terms such as founding fathers, motherland, big brother, or Uncle Sam. This "nation-as-family" frame informs two contrasting moral systems: a "nurturant parent" frame rooted in norms of empathy, tolerance, and diversity; and a "strict father" frame that emphasizes discipline, the rule of law, and obedience to political and moral authorities (p. 52). U.S. immigration and education policies have historically reflected a strict father frame that elevates values of discipline and assimilation of minoritized groups into U.S. nationbuilding efforts (El-Haj, 2010). But such frames are also inadequate. Family separation policies exceed norms of tough love or obedience; they dehumanize.
Elsewhere, Lakoff (2016) argues that a strict father frame does not only separate "them" and "us," but institutes social and moral hierarchies that seek to make "them" more like "us." This social hierarchy privileges "Our Country above other countries" and extends to social structures that position "Men above women, Whites above Nonwhites, Christians above non-Christians, Straights above Gays" (Lakoff, 2016). The strict father frame is thus about forced assimilation and dehumanization.
Examining the U.S. federal government's role in legally separating Native, Black, and Asian families offers a glimpse into the land, bodies, and labor upon which the U.S. was built. We briefly distill these histories of state violence to situate zero tolerance policy in founding nation-building narratives. Native and Indigenous scholars chronicle U.S. attempts to separate Native youth from their families in order to achieve the civilizing mission: "Kill the Indian. Save the Man" (Lomawaima & McCarty, 2006, p. 110). While such oft-used phrases flatten disparate Indigenous experiences of boarding schools beginning in the late-18 th century (Child, 2018), it concisely conveys the dehumanizing intent of state family separation policies. Native American boarding utilized austere dress codes, muted cultural expression, and hierarchical pedagogies to erase Indigenous cultures and extend American beliefs in cultural and linguistic superiority (Brave Heart, 2000). Schools were thus one means of separating Native families couched in broader projects of conquest that called for "the annihilation of Indigenous peoples and other-than-human kin" (Estes, 2019, p. 16) Dimensions of a strict father framework are also evident in nation-building efforts predicated on the normalization of Black suffering with origins in the 17 th -century African slave trade (Williams, 2012;Wynter, 1979). Tadman (1989) estimates that between 1820 and 1860, roughly 200,000 Black people were sold each decade resulting in the forced separation of approximately one in every three children born into slavery. Past systems of slavery shape present institutionalized 1 Other scholars have contributed more comprehensive accounts of family separation in the US (Brave Heart, 2000;Dunaway, 2003;Uchida, 2015). In distilling this brief history, we acknowledge generations of resistance among Indigenous and minoritized groups engaged in locally situated struggles for survival, dignity, and justice (Tuck, 2009). Our focus on state actions is intended only to surface patterns of continuity in U.S. state policies and to trouble ahistorical, utopic notions of "America." scripts (Dunaway, 2003), such as modes of state-sanctioned 21 st -century policing that have incarcerated Black people at a rate five-times higher than Whites (Alexander, 2012). Racialized and gendered patterns of imprisonment represent a modern form of family separation that disproportionately impact Black transgender and queer folks, Black men, and Black female-headed households (Sharp & Marcus-Mendoza, 2001;Taylor, 2016).
A strict father approach is also evident in the illegal internment of Japanese Americans and Japanese immigrant families during World War II. Between 1942Between -1946, the U.S. government forcibly relocated 120,000 Japanese families to remote concentration camps across the U.S. (Tateishi, 2012). Although family separations were not an explicit part of internment policies, Nisei (second generation) Japanese Americans struggled to maintain family relations and cultural practices amid federally authorized polices of legalized incarceration (Fujino, 2005;Uchida, 2015). Patterns of trauma contributed to patterns of drug related deaths in Japanese American communities and efforts to hold families together throughout ensuing generations (Fu, 2008).
As this brief and incomplete historical analysis indicates, Native and Black histories represent the ground upon which America was founded and set in motion later applications of a dehumanizing strict father approach to the incarceration of Asian families. From this historical vantage point, zero tolerance represents a continuation, not an aberration, of U.S. policy toward Indigenous, Black, and minoritized groups. These patterns of continuity are evident in Attorney General Jeff Sessions's warning to migrant families: If you smuggle a child, we're going to prosecute you. And that child will be separated from you, probably, as required by law. If you don't want your child separated, then don't bring him across the border illegally. It's not our fault that somebody does that.

(PBS NewsHour, 2018)
For Sessions, separating families is "not our fault," but, rather, a reflection of law-breaking migrant parents. Implicit in his account is the construction of a specific child; one who is minoritized, illegal, parented by somebody else, less than human (Meiners, 2016). Zero tolerance represents a policy response that promises order, discipline, and stability. As we have sought to illustrate, Sessions's rationale animates foundational, dehumanizing ideas upon which "America" was made and is continually remade.
Despite the explanatory strength of Lakoff's nation-as-family metaphors, a language for making sense of how cognitive frames become codified in state policies and change over time remains limited. A more explicit engagement with the intersections of knowledge, culture, and power are needed to complement Lakoff's psychological framework. As we later elaborate, the absence of policy action in response to empirical reports (Shepherd & Obser, 2017) motivated our specific interests in the role of affect as a policy-making resource. Studying the cultural politics of emotion offered one way to link enduring sociocultural frames with present material and policy conditions of family separation. We turn to Critical Policy Analysis and affect studies to address these conceptual limitations.

Critical Policy Analysis, Affect, and Emotion
Critical Policy Analysis (CPA) offers a generative approach for analyzing the relationships among power, policy, and daily lived experience (Diem et al., 2019;Dumas & Anderson, 2014). It approaches policy as a sociocultural "practice of power": a process in which policies circulate in multiple directions and influence local talk and practices of interpreting actors, who "make" new norms, activities, and practices (Levinson, Sutton, & Winstead, 2009, p. 767). From this perspective, policies do not pertain solely to official government texts, but include a constellation of white papers, media reports, websites, and administrative guidelines (Diem et al., 2019). Informed by CPA, we explored the policy-"making" role of journalists as key agents who evaluate, generate, and disseminate knowledge. We sought to understand how media articles functioned as policy texts, which author and authorize information about formal state policies. We also wanted to understand how everyday organizers-in our case, participants in the #KeepFamiliesTogether movement-(re)made media narratives through local forms of talk, practice, and activism.
To sharpen our application of CPA, we turned to feminist and poststructural conceptions of affect-the unnamable feelings, dispositions, and sensations-and emotion-the nameable, cognizable feelings (Ahmed, 2004;McKenzie, 2018;Petersen, 2011). We assumed that individuals do not only rationally weigh the pros and cons of a proposed policy but are emotionally drawn to and find truth in specific kinds of policies (Anderson, 2014;McIntyre, 2018). We drew primarily on Ahmed's (2004) notion of "impressions," which conceptualizes emotions as forged through the contact of "objects of emotion" (p. 8). According to Ahmed, individuals actively form impressions about objects, but these objects also "press" upon actors. Emotions are relational and involve "(re)actions of towardness or awayness in relation to such objects" (p. 8). Petersen (2011) makes a similar point when she describes how emotions create "felt proximities, distances, desires, disgust and disconnections" (p. 12). We used the language of "impressions" and spatial metaphors of "proximities" to study how media invited publics toward and/or away from particular objects of emotion, such as migrant families and the nation.
Using this combined conceptual framework, we approached the media as a "shared textual location" (Petersen, 2011, p. 64) through which publics felt and made meaning of migrant families and youth. 2 We wanted to understand how the media forged affective proximities and distances that may have moved publics to take political action and donate material resources to migrant and refugee nonprofits. Such an approach builds on the work of recent affective analyses in education studies (McKenzie, 2017;Skattebol & Hayes, 2016;Zembylas, 2018) and extends CPA interests in troubling rational, goal-oriented approaches to policy analysis.

Researcher Positionality, Data Collection, and Data Analysis
Our interests in accompanying migrant, Black, Indigenous, and minoritized families and youth to realize justice on their own terms motivated this joint inquiry. Chang's commitments to equity-oriented, collaborative, community-based research (Glass et al., 2018) motivated his interests in partnering with Koyama and Kasper, whose scholarly praxis directly engages migrant and refugee families through accountable and reciprocal research relations. Together, we approached this project as a way to pose new questions and develop new conceptual tools for education and migrant scholars, educators, and activists.
To acquire a broad national sense of how media invited publics to feel, we prioritized three of the most widely circulated U.S. newspapers: USA Today, The New York Times, and The Wall Street Journal. Using ProQuest and EBSCOhost, we collected media texts that contained the terms, ("Immigra*") AND ("Child" or "Youth" or "Kids") AND ("Family separat*" OR "Zero tolerance"). We bound this search by a concentrated timeline of inquiry beginning with the Trump Administration's instatement of zero tolerance policy (May 2018) and their later, contingent repeal (June 2018). We assumed this condensed timeline would allow greater depth of analysis and potential insights into how objects of emotion emerged and may have animated political action in the June 30, 2018, #KeepFamiliesTogether protests. In total, we collected 178 texts: The New York Times (94), The Wall Street Journal (65), and USA Today (25).
We imported all texts into qualitative data analysis software, NVivo, and began by developing "InVivo" codes-themes developed using verbatim text (Saldaña, 2009, p. 91). We specifically coded for how media framed migrant families, zero tolerance policy, and the nation. Each of these primary themes included four to five sub-themes. Examples of sub-themes for migrants included "(il)legality" and "child/dependent/kids"; for zero tolerance policy, "tearing," "theft," and "camps/cages"; and for the nation, "pride/patriotism" and "a/shamed." To discern a sense of public perceptions, we focused our analyses on opinions, editorials, interviews, and visual images of rally and protest signs from #KeepFamiliesTogether protests. Such data are limited in that they are filtered through the interpretive frameworks of journalists and editorial decision-making teams. Wherever possible, we sought links to direct images, tweets, or video coverage to gather a sense of public thoughts, perceptions, and feelings.
Additionally, we coded for our primary interests in affect and emotion. We did not assume emotions are "in" texts (Ahmed, 2004, p. 13). As we illustrate, the naming of emotions like "shame" was less important than how such terms invited publics to form "impressions" of particular objects of emotion. For instance, when journalists cited politicians who argued zero tolerance "is a shame to our country" (Rogers & Stolberg, 2018), we interpreted these moments as opportunities to ask: What does shame do in this situation? What kinds of boundaries between "our country" and migrant families do such utterances "impress" upon publics? Using writing as a process of inquiry (Richardson, 1994), we drafted analytic memos at the intersection of emergent subthemes (e.g., crying children and cages). We then discussed how these memos addressed our overarching interests concerning the role of media in crafting "(re)actions of towardness or awayness" between readers and specific objects of emotion (Ahmed, 2004, p. 8).
In the following section, we discuss how the media constructed impressions of migrant youth as "crying children" and zero tolerance policy as a strategy of imprisoning youth in "cages." We then illustrate how media dissemination of these objects of emotion forged feelings of shame that brought publics "toward" migrant families and "away" from Trump and zero tolerance. In a final section, we trace how shame catalyzed nationwide #KeepFamiliesTogether but problematically, in ways that recuperated mythic, liberal notions of America as a tolerant nation.

Circulating Objects of Emotion: "Crying Children" and "Cages"
Media coverage of family separations focused public attention on two objects of emotion: zero tolerance policy and migrant children. Journalists often forged impressions zero tolerance as a policy of caging children. Examples include: "children in chainlink cages" (Fernandez & Benner, 2018), "makeshift jails" (Diaz, 2018), and "Border Patrol processing cages" (Henninger, 2018). Goldberg (2018) described the policy as "bureaucratic sadism"; a case where "hundreds of children wait in a series of cages." Even Republican pollster Whit Ayres, observed, "Somehow, I don't think that putting kids in cages is likely to go over very well with suburban moms" (Goldberg, 2018). From pro-immigration journalists to anti-immigrant sympathizers, media circulated impressions of zero tolerance as a policy of "caging" children.
A second object of emotion included the targets of zero tolerance policy, "crying children." Rather than focus on migrant families, journalists focused specifically on children, utilizing what Meiners (2016) describes as "the tactic of forefronting children" to invoke claims of purity and innocence (p. 12). Gershman (2018) described a mother separated from her 6-year old by noting, "her child was taken away screaming and crying to a shelter in Chicago while her mother sought asylum in California." Other journalists portrayed migrant children as "anxious, depressed, crying" (McKinley, Robbins, & Correal, 2018). Such examples invited readers toward migrant children and away from the Trump Administration.
Notably, even defenders of zero tolerance reified images of crying children. Corroborating McIntyre's (2018) suggestion that, in post-truth contexts, all empirical evidence is considered suspect, conservative pundits alleged that Democrats trained children to cry. Conservative commentator Ann Coulter charged that liberals "coached" children and derogatorily referred to migrant children as "child actors weeping and crying on all the other networks 24/7 right now" (Koblin & Hsu, 2018). New York Representative Michael Grimm did not appear to contest the authenticity of crying children, but seemed to minimize the significance of such events when he remarked, "I can take you to any nursery, and you're going to hear the exact same things as a mother who leaves to go to work and leaves her child at day-care" (Epstein & Hook, 2018). These examples illustrate what Boler (2004) terms "feeling power": the ways in which the feelings of dominant interests and values are normalized, whereas those of subordinate groups are deemed inauthentic. Yet, even as Coulter and Grimm challenged the authenticity of crying children, impressions of family separation "stuck" (Ahmed, 2004, p. 8) and may have constituted an emotional basis for mass political action.
Taken together, media circulated stories that brought two objects of emotion-crying children and cages-into contact. By focusing media attention on crying children and not migrant families or parents, the media unmoored debates about zero tolerance from deficit narratives of migrants as "'swarms' or 'marauders' who threaten to 'flood' Western countries" (Esses, Hamilton, & Gaucher, 2017, p. 87). Attention to children, and the pain they endured, invited publics to form an impression of migrants as a part of the national family: as one of "us" (Lakoff, 2016). Such rationales are evident in the deliberate strategizing of Paolo Mendoza, an artist and organizer of "I am a child" posters. Mendoza invoked Civil Rights protest signs that claimed, "I am a man," as one way to mobilize immigrant activism. She explained: I wanted to humanize these children because the only way this unconscionable treatment continues to happen is when these children are dehumanized [ . . .] So, it was important to me to remind all adults, no matter what your political ideology is that these are children. That's it. They're children. And they have a human right to stay with their parents. (our emphasis, Lang, 2018) Elevating attention to children rather than migrants writ large drew publics toward the targets of zero tolerance policy. These strategies recruited "reasonable" publics to negotiate a social contract on behalf of a migrant youth deemed in need of paternalistic rescue (Meiners, 2016, p. 52;c.f., Kirshner, 2015). As we later detail, #KeepFamiliesTogether organizers adeptly extended these media impressions. But before examining how publics took up media narratives, we shift toward an analysis of how the contact of crying children and cages congealed feelings of shame and introduced a third object of emotion: the nation.

Forging Shame and Making the Nation
Shame translated the central objects of emotion from a narrative of migrant families to a story that centered American cultural values of tolerance, liberty, and mercy. We offer two interpretations of the cultural and political work of shame: an optimistic reading that suggests shame called on ordinary Americans to "do better" (Hulse, 2018); and a critical reading that suggests shame offered a way for speakers to signal their own moral standing "in" a tolerant nation (Ahmed, 2004, p. 109). Both interpretations reveal how shame diminished attention to migrant families as well as past and ongoing forms of state-sanctioned violence.
An optimistic reading considers how shame represented a call to action and a demand for the US to live up to its democratic ideals. Former first lady, Rosalyn Carter, stated, "the practice and policy today of removing children from their parents' care at our border with Mexico is disgraceful and a shame to our country" (Rogers & Stolberg, 2018). Similarly, Republican Senator Orrin Hatch of Utah, declared "America is better than this inhumane, anti-family zero tolerance policy" (Hulse, 2018). In other cases, shame was not explicitly named but reinforced impressions of zero tolerance as a stain on America, as in statements such as: "We should be a better country than one that tears families apart" (Stevens & Mervosh, 2018). Across these instances, shame forged a collective "impression" of the US as a tolerant, democratic nation and called on readers to oppose a "disgraceful," "inhumane," and purportedly aberrant federal policy.
A more critical interpretation of shame considers the self-serving interests of policy makers, who may have expressed shame as a way to signal their own moral worthiness. Ahmed (2004) argues that expressions of shame offer speakers a way to "expel the badness" of emotional energies directed toward the self (p. 103). She elaborates, "If you feel shame, you are 'in' the nation, a nation that means well" (Ahmed, 2004, p. 109). Resonant with popularized phrases like-"Not my president"-publicly shaming zero tolerance may have provided speakers and journalists a way to distinguish themselves as feeling citizens distinct from an unfeeling president and his morally reprehensible policy of caging children.
This critical reading offers potential insight into the ways speakers may have utilized expressions of shame to create greater emotional distance between themselves and zero tolerance policy, and on occasion, even expressing shame as an invitation for others to join them "in" the nation (Ahmed, 2004, p. 109). Goldberg (2018) described the Trump Administration this way: "Apparently there are some people close to Donald Trump with the capacity for shame. Not decency or courage, of course, but at least furtive recognition that they're complicit in something vile." Goldberg forges an impression of herself as situated "in" the nation; one that is not morally complicit and is far removed from the architects of zero tolerance. Similarly, Bruni's (2018) indictment of the Trump Administration animates the emotional distancing work of shame: Deny him and his government will stay its heartless course, no matter how much trauma is inflicted on these kids, no matter how much shame is heaped on America, no matter how profound the betrayal of its promise, no matter how deep the interment of its soul. (Bruni, 2018) For Bruni, to defend zero tolerance is to betray America, a nation comprised of "people of generosity and mercy." By framing zero tolerance as a "heartless course," Bruni minimizes legacies of U.S. family separation. He also separates himself, and those who agree with him, from moral complicity in past and ongoing projects of U.S. state-sanctioned violence. Taken together, optimistic and critical interpretations of shame forge an impression of zero tolerance as bad, immoral, disgraceful, and shameful. The cultural work of shame elevates an image of "America" as "a nation that means well" (Ahmed, 2004, p. 109) and positions migrant families as objects of American mercy. In doing so, shame works to bring publics "in" the nation. Yet, as we illustrate in the following section, the emotional bases of political action minimized relations of political solidarity with migrant families and a broader structural and historically responsive critique of U.S. family separation policies.

Tracing Shame in the #KeepFamiliesTogether Protests
On June 30, 2018, more than 600 #KeepFamiliesTogether marches took place across the US. Protests took place "from liberal, immigrant-friendly cities like New York and Los Angeles to more conservative regions like Appalachia and Wyoming" (McCausland, Guadalupe, & Rosenblatt, 2018). Although we cannot account for the multiple reasons people participated in protest marches, we argue that shame represented one powerful affective resource underlying political action. We trace how shame emerged as an emotionally-laden, meaning-making resource among op-ed writers, parents, and protesters, and ultimately, moved ordinary people to "make" policy by demanding an end to family separation policies (Levinson, Sutton, & Winstead, 2009, p. 767).
Early expressions of shame were evident in op-ed stories among concerned citizens who, in addition to expressing feelings of anger, indignation, and worry, expressed shame for their President. The words of one retired Methodist pastor offer a representative line of moral argumentation. He expressed: [I]f you can hear the voices of those little children on that audiotape and not be moved to tears-and then not be moved to moral outrage, righteous indignation, deeply and strongly felt anger at the horrifically horrible treatment of children done in the name of our nation-then I worryingly wonder about the condition of your humanity, your soul, your spirit, your heart, your caring and compassion. (Saunders, 2018) The Methodist pastor fuses media impressions of cages and crying children to describe the horrific treatment of children "done in the name of our nation." His phrasing is instructive. By tying zero tolerance and the nation, the pastor creates an impression of family separation as one shameful expression of present America but not representative of U.S. nation-building projects rooted in slavery and settler-colonialism. To express shame is to join in a collective feeling of moral outrage set in contradistinction to those who are un-American; that is, those who are inhumane, dispirited, heartless, and uncaring. The June 30, 2018, #KeepFamiliesTogether protests unfolded against this cultural backdrop of national shame. On several occasions, shame constituted the very language of protest, such as Chicago activists exclaiming, "Shame!" at the Trump hotel (Raice, Barrett, & Duehren, 2018), or organizers chanting, "Shame, shame, shame!" at U.S. homeland security secretary, Kirstjen Nielsen, as she ate dinner at a restaurant in Washington D.C. (Williamson & Nixon, 2018). But on other occasions, participants did not explicitly name shame. Instead, they spoke of inchoate, affective sensations that mobilized their political actions. An attorney who attended a march expressed that hearing stories about family separation was difficult and added, "It hits you in a real gutpunch, emotional way" (Raice et al., 2018). Likewise, one mother participated in protests to set an example for her two children. She explained: "It's really hard to sit at home with all this crazy stuff going on" (Raice et al., 2018). Participating in #KeepFamiliesTogether protests may have served as a way to alleviate personal shame, understood as "an intense and painful sensation that is bound up in how the self feels about itself" (Ahmed, 2004, p. 103). Notably absent in these accounts are references to migrant families, whose lives are either not mentioned or reduced through abstract appeals such as "all this crazy stuff going on." Media coverage of protesters' motivations thus expressed passing concerns about migrant children and tended to elaborate how political action set an example for American children.
Conversely, shame was also evident in its emotional opposite: pride. Popular protest signs proclaimed, "Immigrants Make America Great" and "Love, not hate, makes America great." These statements expressed what El-Haj (2010) describes as "The Beauty of America": liberal ideals of tolerance and freedom that cloak racialized and gendered discourses of nationalism. But theories of change that hinge on tolerance have a way of "analytically disappearing the political and historical constitution of conflicts and subjects" (Brown, 2006, p. 18); in this case, de-coupling zero tolerance from past and present structures of settler-colonialism and anti-Black racism (Clay, 2019;Estes, 2019). In this way, pride ameliorated bad feelings about the US, and in the process, minimized attention to legacies of state-sanctioned family separation.
One nonprofit consultant and #KeepFamiliesTogether participant admitted "All Americans are immigrants, except Native Americans" (Raice et al., 2018). She explained her motivations for participating by adding, "We need to show the alternate face of the U.S." (Raice et al., 2018). By "alternate face" here, the speaker animated legacies of immigrant struggle, but minimized distinctive legacies of coloniality and racialized forms of state-sanctioned violence. Likewise, President Bill Clinton, who, despite his tough stance on immigration during his presidential tenure, remarked, "These children should not be a negotiating tool [ . . .] reuniting them with their families would reaffirm America's belief in and support for all parents who love their children" (Baker, 2018). Expressions of pride thus invoked ahistorical views of U.S. state formation and replaced statesanctioned forms of violence with naïve narratives of democratic inclusion and tolerance. So far, we have argued that the media circulated objects of emotion (cages and crying children) that congealed feelings of shame for zero tolerance and pride in America. As the opening quote from Jess Morales Rocketto makes clear, shame can lead to unprecedented displays of national unity, encouraging people from diverse backgrounds to "show up for immigrants" (McCausland et al., 2018). Our findings do not diminish the political possibilities of #KeepFamiliesTogether protests but seek to raise critical questions about the terms upon which struggles for justice are waged. Failing to interrogate the emotional and political bases of resistant political action may contribute to activist responses that reify the very structures of inequality that researchers, organizers, and educators seek to transform. In the remainder of this section, we discuss theoretical and pedagogical implications of our analysis and identify possibilities for advancing migrant justice amid post-truth, liberal democratic contexts.

Beyond "Nurturant Parent": Toward a "Pedagogy of Discomfort"
One implication stemming from our analysis concerns the limits of a liberal "nurturant parent" frame as a sufficient contrast to a "strict father" frame (Lakoff, 2006, p. 52). Values of empathy, diversity, and tolerance might contrast with those of a strict father frame but participate in social processes that delay justice. Like tolerance, shame "disguises power" (Brown, 2006, p. 26); it minimizes public reckoning with structural injustices and favors individual and interpersonal solutions. When mediated by emotions like shame, the nurturant parent frame sanctions modes of engagement that allow individuals to feel better about themselves, regardless of whether such actions materialize actual symbolic or material gains for migrant families and other Indigenous or historically marginalized groups.
Given the limits of a nurturant parent frame, further research might explore ways of troubling the assumed goodness of tolerance and our own feelings of moral worth bound up in narratives of the nation. Ahmed (2004) argues that one of the key objectives of emotional struggles is to make explicit "the norms that we wish to contest" and "the wounds we wish to heal" (p. 201). Confronting violences naturalized through policies of zero tolerance and through liberal norms of American tolerance opens conceptual space for a politics of solidarity; what Taylor (2016) defines as "standing in unity with people even when you have not personally experienced their particular oppression" (p. 215). This political understanding of solidarity is built on humanizing and dignifying understandings of migrant and refugee families as constrained, but not determined by unjust structures and policies. Through this asset-based lens, migrants are not reducible to images of crying children, but agentic actors who scholars, educators, and activists might accompany in struggles for dignity and justice (Tomlinson & Lipsitz, 2019).
If past nation-building legacies underpin affective relations (Skattebol & Hayes, 2016), then troubling past and ongoing forms of state violence offers one way to advance a politics of migrant solidarity. Zembylas and Boler's (2002) notion of a "pedagogy of discomfort" is helpful in this regard. A pedagogy of discomfort explores "how emotions define how and what one chooses to see, and conversely, not to see" (Zembylas & Boler, 2002, p. 6). Such a pedagogy "discomforts" emotional investments individuals might have in particular constructions of self, society, and nation.
In the contexts of zero tolerance policy, a pedagogy of discomfort might encourage students to question their emotional responses to media narratives of "crying children." It might also raise questions about who does and does not qualify as a child (Meiners, 2016), and who instead, resides on the opposing side of "a very powerful, very strong border" designed to keep American families together (Trump, 2018). Such an approach resists "the trap of sentimentality" (Zembylas, 2018, p. 3) and invites young people to explore individual and collective approaches to problematizing structures that drive "migrant caravans" to seek safety and security, such as the U.S. role in backing anti-communist and authoritarian leaders in Central and Latin America (Sklaw, 2019). 3 Additionally, a pedagogy of discomfort might encourage educators to utilize pedagogies of participatory codesign, which can springboard collective, self-determined practices of imagining and sustaining educational and political projects that are driven by and accountable to migrant families and migrant youth (Ishimaru et al., 2019;Kirshner, 2015).

Implications for Critical Policy Analysis and Affect Studies
Our findings also have consequences for CPA and affect studies. The absence of policy action in response to mounting social scientific evidence elevates the significance of affect and emotion in mediating policy-making processes. In December 2017, six months prior to intensive media coverage of family separations, a coalition of immigrant and refugee advocacy organizations appealed to the U.S. Office of Homeland Security and Office for Civil Rights and Civil Liberties. Writing on behalf of immigrant advocacy organizations, Shepherd and Obser (2017) documented over 155 children in Arizona separated from their families. Political inaction in response to Shepherd and Obser's appeal speaks to the limits of rational debate and scientific appeals in a posttruth moment. As De Sousa (1990) argues, "emotions are among the mechanisms that control the crucial factor of salience among what would otherwise be an unmanageable plethora of objects of attention, interpretations, and strategies of inference and conduct" (original emphasis, p. xvi). Absent a national outcry, federal policy makers ignored Shepherd and Obser's demands. Our findings illustrate how emotions, like shame, elevated attention to zero tolerance and mediated policy-making processes with mixed and contradictory political implications.
Conversely, our findings also inform studies of affect by linking cultural critique with policy and material realities. Affect studies tend to focus on analyses of culture as a critical "arena of consent and resistance" (Hall, 1981, p. 192). While attention to the politics of popular culture is indeed warranted, such studies, at times, lack direct connections to matters of material or policy significance (Boler & Zembylas, 2016). Our analysis began with material contexts-the construction of detention facilities, nationwide political actions, $20 million in donations to RAICES (Williamson & Nixon, 2018, June 20)-and traced the cultural political work of shame in association with these material realities. We hope our findings motivate further refinements and applications of affective policy studies to trouble the narrow boundaries of what "counts" as educational policy (Anyon, 2005, p. 66) and offer insight into new modes and mechanisms of policy-making.

Further Research
One important consequence of this study concerns the role of affect in mediating critical pedagogical teachings of national histories and fomenting sustainable political actions. Ahmed (2004) writes that the terms "passion" and "passive" share the same Latin root "passio" meaning "suffering" (p. 2). In this sense, affect studies represent a complementary field of scholarship for conceptualizing how we might transform passive publics inured to the suffering of others into passionate publics energized by a collective sense of injustice. Teachers Against Child Detention (TACD) offers one example of these possibilities (teachersagainstchilddetention.org). In partnership with Educators for Migrant Justice and Black Lives Matter at Schools (among other grassroots and justice-oriented organizations), TACD organizes teach-ins and political action campaigns to demand justice for immigrant and refugee youth. Further studies might explore on-the-ground efforts to transform affective moments into sustainable movements for education justice (Chang, 2019;Warren, 2018), particularly in ways that build on the practical, pedagogical and community-building practices of organizations like TACD, Educators for Migrant Justice, and Black Lives Matter at Schools.
Other studies might explore tensions between the momentary windows of "newsworthy" events and the ongoing, mundane violences enacted toward migrant families. An eight-day window of intensive media coverage (from June 18, 2018 -June 26, 2018) facilitated a national outcry, yet coverage of migrant families has since winnowed. Further research might explore how grassroots organizers sustain public pressure even as mainstream media entities no longer deem issues facing migrant families newsworthy. We agree with Petersen (2011), who writes, "Rather than advocate a less passionate politics, we should better understand the politics of our passions" (p. 163). Better understanding the cultural politics of emotions can enhance our collective understandings of the messy contradictions encoded in activist struggles for migrant and education justice.

Conclusion
We want "heart" and security in America! (President Donald Trump, as cited Williamson & Nixon, 2018) Why are you doing this? Do you have a heart? (Anonymous audience member's question to former U.S. Attorney General, Jeff Sessions, as cited in Meckler, 2018) Since the summer of 2018, the Trump Administration has continued its assault on migrant and refugee families. At the time of our writing, the Trump Administration issued a "final rule" that legalized indefinite migrant family detentions (Swanson, 2019). Understanding how affect and emotion mediate policy-making processes and social movements is needed now more than ever. As many media scholars have argued, states and corporations not only know, but target emotions to advance their political agendas (Anderson, 2014). This project seeks to build on ongoing efforts to grasp the role of emotion in catalyzing and sustaining activist and scholarly movements for education and migrant justice. In this sense, we aim to build toward collective political and educational projects that seek to reclaim a politicized notion of "heart": one that troubles traditional narratives of family and nation and elevates possibilities for more historically responsive, intersectional movements for justice.

About the Authors Ethan Chang
University of Hawai'i at Mānoa ethanchang@ucsb.edu orcid.org/0000-0002-8242-3984 Ethan Chang is a sociologist of education and assistant professor of educational leadership at the University of Hawai'i at Mānoa. He is a former middle and high school public teacher with prior teaching and research experiences in Hawai'i, South Africa, and Palestine. His research broadly explores the intersections of policy, place, and race in struggles to redress inequities of educational opportunity.

Jill Koyama
University of Arizona jkoyama@email.arizona.edu orcid.org/0000-0003-0990-4930 Jill Koyama is a sociocultural anthropologist and Associate Professor in Educational Leadership and Educational Policy Studies and Practice (EPSP) and Director of the Institute for LGBT Studies at the University of Arizona. Her work is situated across three integrated strands of inquiry: the productive social assemblage of policy, the controversies of globalizing educational policy, and the politics of education policy and newcomer education.

Julie Kasper
University of Arizona juliekasper@email.arizona.edu orcid.org/0000-0002-5952-9350 Julie Kasper is a Graduate Fellow at the University of Arizona pursuing her doctorate in Educational Leadership while leading the Refugee Educator Academy at the Carey Institute for Global Good. A National Board Certified Teacher in English as a New Language and classroom teacher for over 16 years, her current work focuses on curriculum and program design, teacher professional learning, and teacher and youth leadership development, especially in relation to refugee education and work with culturally and linguistically diverse learners. Research interests include culturally responsive pedagogies and school leadership, learning design both within and outside formal education spaces, and theoretical imaginings around-and practical efforts toward-educational equity and social justice.
education policy analysis archives Readers are free to copy, display, distribute, and adapt this article, as long as the work is attributed to the author(s) and Education Policy Analysis Archives, the changes are identified, and the same license applies to the derivative work. More details of this Creative Commons license are available at https://creativecommons.