Canadian Journal of Communication Vol 39 (2014) 235-248
©2014 Canadian Journal of Communication Corporation

Homelessness: Emotion Discourse and the Reproduction of Social Inequality

Barbara Schneider
University of Calgary

Barbara Schneider is Professor in the Department of Communication and Culture at the University of Calgary, 2500 University Dr. N.W., Calgary AB  T2N 3A7  Email: .

ABSTRACT  This article draws on data gathered in focus groups to analyze how people talk about homelessness and compares the findings to how homelessness is represented in the media, specifically newspapers. It examines how ideas about homelessness that circulate in society are taken up, used, and reproduced by people in social interaction. People “care” about homelessness and use emotion discourse in the focus group context to construct a moral identity and to manage interactional dilemmas. They express sympathy for homeless people, deflect responsibility for any negative feelings they may have, and shift responsibility for doing something about homelessness. In using emotion discourse, they reproduce conceptions of homelessness that circulate widely in the media and in society generally; this, in turn, reproduces existing social relations of inequality and exclusion.

KEYWORDS  Homelessness; Media representations; Audience research; Emotion discourse

RÉSUMÉ  Cet article a recours à des données provenant de groupes de discussion afin d’analyser comment les gens parlent d’itinérance entre eux, puis il compare les résultats de cette analyse à la représentation de l’itinérance dans les médias contemporains, particulièrement les journaux. Il examine comment les gens dans leurs interactions adoptent, utilisent et reproduisent des idées sur l’itinérance qui circulent déjà dans la société. Les gens se « soucient » de l’itinérance et ont recours à un discours émotif dans leurs groupes de discussion pour se construire une identité morale et gérer les dilemmes interactionnels. Ils expriment de la sympathie pour les sans-abris, tout en déclinant toute responsabilité pour les sentiments négatifs que ces derniers peuvent éprouver et rejetant sur autrui le devoir d’agir sur l’itinérance. En utilisant un discours émotif, ils reproduisent des conceptions sur l’itinérance qui circulent abondamment dans les médias et dans la société en général. Leurs pratiques reproduisent ainsi des rapports d’inégalité et d’exclusion.

MOTS CLÉS  Itinérance; Représentations médiatiques; Étude d’audience; Discours émotif

Social science’s conception of media audiences has undergone significant theoretical shifts in the past 30 years. From the passive cultural dopes of mass culture theory, to the active interpreters and resisters of media reception theory, to the emotional audience (e.g., Döveling, von Scheve, & Konijn, 2011), to the reportedly “dead” audience (e.g., Press & Livingstone, 2006), the concept of the audience has become ever more elusive. This has posed problems for those wishing to study audiences, particularly when, as in this study, there are no specific media texts, events, or objects that anchor the conception of audience in a specific group of people.

In this article, I draw on data gathered in focus groups to examine how people talk about homelessness. I take the position articulated by Ross and Nightingale (2003) that media allow people who may have little direct experience of particular aspects of social life to share access to knowledge about these things. Although people who have stable housing typically have casual, intermittent contact with homeless people panhandling on the street, most have little personal in-depth knowledge of homelessness. What those people know, or think they know, about the causes of and possible solutions to homelessness comes from the media. There is, in fact, no other source for this knowledge except the media (see for example, Fenton, Bryman, & Deacon, 1998; van Dijk, 2000). People may sometimes reference the same media texts in their interactions, particularly when a spectacular event related to homelessness is reported widely (“Did you hear about that homeless woman who set herself on fire?”), but they also tend to express a general shared sense of “what everybody knows” about homelessness, without any reference to specific media texts. Dahlgren (1995) notes this when he describes the mediated public sphere as “a space—a discursive, institutional, topographical space—where people in their role as citizens have access to what can be metaphorically called societal dialogues, which deal with questions of common concern” (p. 9).

Homelessness is just such an issue of common concern, one which elicits strong expressions of emotion. People “care” about homelessness; in particular, the people who came to our focus groups thought it was an important social issue, worthy of discussion and action. The focus group data drawn on in this article was gathered as part of a larger project on media representations of homelessness, in which I examined the content of media representations (Schneider, 2012; Schneider, Chamberlain, & Hodgetts, 2010; Remillard & Schneider, 2010) and the work of journalists who cover homelessness (Schneider, 2013). Here, following advice (Fenton, 2007; Kitzinger, 2007) to attend to the cycle of production, content, and reception in the study of media, I complete the circuit by attending to what people say about homelessness. I seek to understand how ideas about homelessness that circulate in society are taken up, used, and reproduced by people through social interaction. I focus particularly on expressions of emotion in relation to homelessness. When talking about homelessness in the focus group context, people use emotional discourse to construct a moral identity and to manage interactional dilemmas. In doing so, they reproduce discourses about homelessness that circulate widely in the media and in society generally, thus reproducing existing social relations of inequality and exclusion.

Who is the audience?

The very idea of an audience is, of course, a discursive construct. It is, as Hartley (1988) says, “a creation of criticism” (p. 236). It is therefore very hard to identify what exactly an audience consists of or to say something definitive about it. We can, however, say something about how audiences have been represented in scholarly work and how they have been studied. Work in this area since the middle of the twentieth century has been described as oscillating between approaches that see media texts as playing a dominant role in shaping audiences and approaches that see audience members as active users and interpreters of texts. For Abercrombie & Longhurst (1998), this oscillation constitutes a continuum between “dominant text and dominant audience” approaches. More recently, as media have become ubiquitous in North America, with society increasingly described as mediatized (e.g. Couldry, 2005; Dahlgren, 2005; Livingstone, 2005), conceptions of the audience have become more complex. As Peterson (2005) says, “people are never only audiences” (p. 130), and the relationship of individuals to media can no longer be thought of only in terms of consumption and interpretation of specific media texts. Ross and Nightingale (2003), for example, describe media as a “means by which people keep abreast of current affairs and contemporary trends, entertain themselves, relax, take time out, become involved in the cultural life of their community, and make themselves into interesting people” (p. 5). Abercrombie and Longhurst (1998) offer the concept of the diffused audience, in which being an audience is not connected to specific media texts or events. Instead, the media are inextricably interwoven into everyday life and everyone is an audience member all the time. As Livingstone (2005) says, there are “few if any aspects of social or personal life unaccompanied by, or independent of, the media” (p. 19). The audience is, as Bird (2003) asserts, “everywhere and nowhere” (p. 3).

Various scholars have responded to this entanglement of media in everyday life by proposing approaches that go beyond the dominant text/dominant audience binary. Couldry (2004) suggests that analysis should focus on “the study of the open-ended range of practices focused directly or indirectly on media” (p. 117). He sees this approach as a way of avoiding the intractable problem of having to demonstrate conclusively that media do in fact have “effects.” Instead, he suggests that we should study “how media are embedded in the interlocking fabric of social and cultural life” (p. 129). Peterson (2005) suggests attending to intertextuality, which he defines not as a characteristic of texts, as it is usually understood, but as “an active social process involving the extracting of a discourse … from one setting and … inserting it into another” (p. 130). It is a public and social process of which its performers may be quite unaware. He suggests that discourses appropriated from mass media enter into social interaction even when speakers have not seen the original media texts. Dahlgren (2005) also regards the media as providing resources for individuals. He sees media messages as “but one step in a larger communication chain that includes how the media output is received, made sense of, and used by citizens in their interaction with each other” (p. 321). The focus in this study is therefore not on how audiences receive and interpret media output, but on how they use it in social interaction with each other.

Emotion discourse

The focus groups in this study offered us an ideal opportunity to study social interactions in which people talk about homelessness and to examine how people use their knowledge about homelessness in interaction with each other. A particularly striking aspect of the focus group discussions (as reflected in the transcripts) was the frequent use of emotion words, such as “shocked” or “frustrated,” and the frequent descriptions of emotional states, both those of the speaker and those of the homeless people the speakers were describing. This drew my attention to the role of what Lutz and Abu-Lughod (1990) call emotion discourses in social interaction. There has been a flowering of interest in emotion across the social sciences and humanities since the early 1980s (e.g., Stets & Turner, 2006; Lewis, Haviland-Jones, & Feldman Barrett, 2008), including in the field of media studies (e.g., Döveling et al., 2011). Scholars have attended to the evolutionary and neurological aspects of emotion (e.g., Turner, 2007), the identification and definition of emotions (e.g., Thamm, 2006), the ontological status of emotion (e.g., Frijta, 1994), the expression of emotion (e.g., Matsumoto, Shoita, O’Sullivan, & Frank, 2008), how emotions are engendered in engagements with media (e.g,. Grossberg, 1992; Zillman 1988), and how emotions are presented and managed in media products (e.g., Pantti, 2011). In this study, I align myself with those who see emotions not as internal, irrational, natural biological experiences, but as sociocultural constructs, embedded in cultural values and constituting an essential aspect of social activity (e.g., Abu-Lughod & Lutz, 1990; Edwards, 1999). Individuals do, of course, experience emotions as subjective and embodied, but emotions can be more than simply personal experiences. They provide a set of flexible resources (Edwards, 1999) that people use in social interaction to construct identity and establish social accountability. As Hepburn and Jackson (2009) say, emotion is not separate from nor does it underlie discourse; rather, it is something that is “managed and made accountable in discourse” (p. 182). It is “invoked, described, and displayed for the purposes of social action” (p. 183). Emotion discourse, in this view, is implicated in the production of social relations: “Emotion talk must be interpreted as in and about social life” (Abu-Lughod & Lutz, 1990, p. 11).

Various scholars have illustrated the usefulness of this approach to emotions. For example Hepburn’s (2004) examination of crying as a social practice focuses on the interactional role of crying, rather than on crying as a result of a personal emotional state. Locke and Edwards (2003) analyze Bill Clinton’s grand jury testimony to show how Clinton uses emotion discourse to manage blame and responsibility. These studies have demonstrated the value of analyzing emotional expressions as social practices with a role both in “getting things done” (Hepburn & Jackson, 2009, p. 189) in social interaction and in re-circulating larger cultural understandings of the moral and social order. Studies that take this approach to emotion have, however, not explicitly discussed how emotion discourse might be related to media discourse. Media scholars have typically examined the emotional reactions of audiences to media (e.g., Pantti, 2011), but not the use of emotion discourse in social interaction among audience members. I therefore bring these two areas together to analyze how expressions of emotion are managed in the social interactions of the focus groups, with a view to understanding what people are accomplishing when they use emotion discourse in their talk about homelessness and how this relates to the representations of homelessness that circulate in the media and in society more generally. 


Focus groups, in which researchers interact with audience members in small group settings, have been used by many researchers to study media audiences, starting with Morley’s (1980) well known study of Nationwide audiences. They have since been used to talk to audiences about a range of topics, including news representations of social science (Fenton et al., 1998), AIDS (Kitzinger, 1993), and child sexual abuse (Kitzinger, 2000). As Gauntlett (2007) points out, focus groups are not without their problems, particularly if researchers want to make claims about what people “really” think. Focus groups are social settings in which people interact with others, constructing moral identities through, in the case of our groups, performing knowledge and caring about homelessness. I therefore follow Alasuutari’s (1999) and Hartley’s (1988) advice to take a discourse analytic approach to the study of audience and to focus on a specific aspect of participants’ talk—namely, emotion discourse. I regard the focus group interaction not as an indicator of individuals’ beliefs about homelessness, but rather as an opportunity to analyze the participants’ use of emotion discourse and media-derived knowledge to construct identity and to manage accountability. This focus offers a way to understand how knowledge about homelessness circulates in society.

We conducted seven focus groups (comprising a total of 43 people) in the summer of 2009. Participants attended because they were concerned about homelessness and saw this research project as an opportunity to contribute in some small way to doing something about homelessness. They were recruited through posters and advertisements at our university and in various shopping areas in our city and were in no way a random or representative sample. The groups ranged in size from two to eight participants. Participants ranged in age from 18 to 59, with most in their twenties and thirties. Of the 43 participants, 23 were women and 20 were men. A variety of ethnic backgrounds were represented, although none of the participants identified themselves as Aboriginal. Two identified themselves as having a high school diploma, and all the rest either were currently attending or had completed postsecondary education. Thirty-three (77%) identified themselves as readers of Calgary newspapers. Two said they worked or volunteered in the homeless sheltering industry. One focus group took place with staff at a café that welcomes homeless people. The sessions lasted between 60 and 90 minutes and were relatively unstructured in order to encourage a wide range of discussion. The moderator used a script to ask similar questions in each focus group, but as participants took the discussions in very different directions, it was not possible to ask exactly the same questions in each group.

Before the sessions, focus group participants were emailed copies of recent articles on homelessness from a local newspaper, and during the sessions, they were shown photos of homeless people from the same newspaper. One of these articles was a report about the inappropriate use of the indoor public walkway system in our city by homeless people who were apparently drug users. Another reported comments made by Canada’s governor general about Canada’s poor record in solving the problem of homelessness. Another described five individuals at various stages of “recovery” from homelessness, from someone still homeless to someone working and living in his own apartment. The last reported homeless people’s reactions to fines being issued under a new public behaviour bylaw. Participants were also emailed a YouTube clip about a “bumbot” created by an Atlanta bar owner, a mobile device that shines lights, makes noise, and sprays water to deter loitering outside his bar by apparently homeless people who were said to be dealing drugs.

Not all participants read the materials or viewed the clip before the sessions, so printed copies of the articles were available at the sessions. Some participants commented directly on the media materials, but for the most part, these materials simply served as prompts for a general discussion of homelessness. Participants were also asked to share their personal stories about homelessness and to respond to each other’s statements. All group members received a $25 gift card for their participation. The discussions were recorded, transcribed by a professional transcriptionist, and then coded using HyperRESEARCH, a computer program for qualitative data analysis. All names were changed in the transcripts to ensure anonymity.


In this analysis, I examine the data for how the emotions discussed or expressed reflect particular conceptions of homelessness; I then compare these to the findings from previous studies that analyzed journalists’ perspectives (Schneider, 2013) and news items on homelessness (Schneider, et al., 2010). I examine the discursive work that descriptions and expressions of emotion perform in the focus group context. I argue that the use of emotion categories allows people to construct moral identities and to manage interactional dilemmas. In doing so, people reproduce various discourses about homelessness that circulate widely in the media and in society and contribute to the reproduction of existing social inequalities.

Sympathy and social inequality

Participants constructed a moral identity in the focus group context by expressing their sympathy for homeless people. As Clark (1997) points out, “we have made sympathy giving a key ingredient in ‘niceness’ and institutionalized it in the social role of the ‘good person’” (p. 11). Sympathy for those who are somehow less fortunate is part of our moral code. Speakers in the focus groups demonstrated that they care about homelessness, using emotion discourse in several different ways. These expressions of sympathy closely parallel expressions of sympathy available in newspaper representations of homelessness. Paradoxically, they function to reproduce social inequalities rather than interrupt them.

A number of participants described themselves as having been surprised or shocked when they learned something about homelessness from the media items we provided for discussion. For example, one said he was shocked after reading the article about the bylaw that allowed police to give tickets for the use of public space in ways deemed inappropriate by civic authorities. This seemed to him to target only homeless people. After reading another of the newspaper items, several others described themselves as shocked that the number of homeless people in our city is as large as it is. Another described herself as shocked that it is so difficult to find a bathroom in the downtown area that she can use without having to buy something first and compared that to the even greater difficulty homeless people would surely have. Another described the appearance of a homeless man he was talking to as shocking. Another described himself as furious after he viewed the “bumbot” video. These speakers conveyed sympathy for homeless people through their use of emotion words.

Another way in which people expressed sympathy for homeless people was by characterizing homeless people as having emotions. One woman described homeless people as proud, not necessarily looking for handouts. Several said that homeless people feel shame at the situation they find themselves in and so will not make eye contact. They also were described as feeling sadness and anxiety and feeling worthless or bad about themselves. Others were described as being grateful for the help they receive, or as having moments of joy, such as when finding great stuff in the garbage. In describing homeless people as experiencing commonly felt emotions, speakers aligned themselves with homeless people and asserted them to be worthy members of the human community.

Participants also demonstrated their sympathy by distancing themselves from the emotions of people who apparently do not feel such sympathy. One described the unsympathetic opinions of people who do not consider homeless people worthy of help, asking of them, “Are you a human being?” Another described the rudeness of a non-homeless man at the recycling depot toward a homeless man who was also recycling bottles, with the words that only someone who is “an uncaring piece of whatever” would not feel sympathy for homeless people. Another described a person who thought of homeless people as “pests” as “unhuman.” Another recounted seeing housed working people angrily confronting homeless people on the street and telling them to get a job.

In these descriptions of others’ emotions, the speakers used a discursive strategy identified by Snow & Anderson (1987) as distancing. Distancing is a way of asserting a favourable personal identity and making moral judgments about those from whom one is distancing oneself. In the statements identified above, speakers distanced themselves from those who are apparently unsympathetic and, by implication, asserted their own sympathy for homeless people. Distancing not only constructs a particular identity for the speaker, but also allows speakers to introduce other (i.e., negative) points of view about homelessness without taking responsibility for those points of view; in this sense, it reinforces negative perceptions of homeless people. Expressions of sympathy are thus a double-edged sword, simultaneously promoting both positive and negative understandings of homelessness.

This “double” function of sympathy is also evident in newspaper reports of homelessness. Journalists who were interviewed for this project (Schneider, 2013) indicated that the production of sympathy is one of their goals in reporting on homelessness. They care about homelessness and believe that they contribute to solving this social problem by generating sympathy for homeless people through “telling the stories” of individuals in desperate circumstances. As one said, “I actually believe the most important thing we do is we tell stories, and the reason the people will read a story is because there’s a face and a name, and it’s the story that makes people care about homelessness.” Such personal stories may indeed generate some sympathy, as they generally describe the desperate situations homeless people face on the streets or in shelters, often estranged from family and without secure employment. However, they typically highlight aspects of homeless people’s lives that housed people fear or, at least, do not condone—for example, fights in shelters, drug and alcohol addictions, and interactions with police—and so simultaneously undercut that sympathy.

News accounts that focus on the desperate situations of particular individuals also promote a view that individual factors, such as mental illness, addictions, or personal incompetence, push people into homelessness, rather than structural factors, such as unfair wages and over-priced housing. This tendency has been noted repeatedly in the literature on media reporting of homelessness (e.g., Buck, Toro, & Ramos, 2004; Hodgetts, Cullen, & Radley, 2005). A framework that casts individuals as personally responsible for their homelessness releases society from making any structural changes that might address problems of poverty and homelessness.

Clark (1997) describes another one of sympathy’s double functions. On one hand, it connects the fortunate to the unfortunate, creating social bonds and decreasing social distance. Sympathy, she says, is an aspect of what binds people together, playing a role in the creation of a larger social order (Clark, 1997). However, this larger social order consists of multiple divisions and hierarchies, and sympathy also plays a role in increasing social distance and sustaining relations of inequality. Sympathy is provided only by “us” to “them,” placing recipients in a one-down position and saddling them with a debt of sympathy they may not be able to repay. Likewise, recognition of the essential humanity of marginalized Others is often cited as a positive thing, something that will help to reduce stigma. If we think someone is just like us in some way we are more inclined to be sympathetic and less likely to marginalize them. But, as Schneider and Remillard (2013) have shown, positive attitudes and statements about marginalized groups and individuals have the same stigmatizing effects as negative ones. Sympathy reinforces negative perceptions of homeless people and therefore also recreates social structures of inequality.

Expressing sympathy in appropriate ways—that is, following what Hochschild (1983) has called feeling rules—is one aspect of constructing an identity as a person who cares about homelessness. Expressions of sympathy allowed participants to manage the interactional situation of the focus group and to construct a moral identity.  Simultaneously, their expressions of sympathy reproduced the media’s focus on individual rather than structural causes of homelessness, and thus paradoxically reproduced the social inequalities they claimed to want to change.

The deserving and undeserving poor

In this and the next section, I discuss the discursive use of emotion categories to mitigate interactional dilemmas and manage issues of accountability that arise in speaking about homelessness. In addition to sympathy, focus group participants expressed a number of negative emotions toward homeless people when describing what they saw as problematic behaviours. Expression of these negative emotions placed the speakers in an interactional dilemma in the context of a focus group made up of people who “care about” homelessness. As one participant said, “There is this huge gap there. … We all want to do something about this problem, have good intentions, but the real physical person in front of us is a problem.”

Several speakers described having felt bullied by aggressive, apparently homeless people who asked them for money. One said he felt “pissed off” by a very persistent panhandler who was “bothering” his sister. One speaker described a situation in which he was approached at a bus station in another city by a man claiming to need $40 to get his wife and child a bus ticket. He gave the man the money and then watched as the man approached another person with exactly the same story. He said he had been “had” and felt betrayed. Another described himself as “a socialist and sort of left-leaning liberal,” but nevertheless said that he gets upset when he sees an apparently able-bodied young person panhandling, since he assumes that they could live with their own family or at least make their own living. Two speakers explicitly acknowledged the interactional dilemma of expressing negative feelings about homeless people. One said he felt ashamed to admit that some homeless people are aggressive and annoying. Another said that he “felt awful” that he had reservations about the possibility of a formerly homeless person living next door to him. These speakers acknowledged the interactional dilemma and managed it by using the emotion categories of shame and guilt to describe their own reactions to their negative feelings about homeless people.

As Edwards (1999) points out, emotion discourse “works back upon the nature of prior events” (p. 279). That is, the use of emotion words is a way of specifying the character of the event being described. The bus station incident could, for example, be described with admiration at a poor person’s ingenuity in finding ways to survive. The persistent panhandler could be described as having personal characteristics that might be useful in a business setting. Instead, the description of these events as upsetting or annoying characterizes the events as self-evidently objectionable and the perpetrators as problematic. This suggests that, on the one hand, people feel sympathy for homeless people and describe themselves as shocked if homeless people are treated in ways that seem to marginalize them. On the other hand, they feel upset and annoyed when homeless people behave in ways that they disapprove of. These two positions mirror the long-standing, widely circulating division between the deserving and the undeserving poor. As scholars such as Hopper (1988) and Feldman (2004) have noted, one version represents the poor as victims of circumstances beyond their own control and who therefore need and deserve help. The second version represents them as having freely made bad choices that led to their present unfortunate circumstances and as being therefore undeserving of help. They are “matter out of place” (Douglas, 1966).

These competing conceptions of homeless people as deserving and undeserving overlie each other in complex ways in discourse about homelessness, including in newspaper representations. As we showed in Schneider et al. (2010), journalists’ desire to help homeless people manifests itself in the generally neutral or positive representations of homeless people in news items (only one in five were negative). At the same time, nearly four out of five news items mentioned negative associations with homelessness (for example, addictions, panhandling, or squatting), with an average of four associations in each item. Homeless people were presented, on the one hand, as needing and deserving help, and simultaneously as having chosen activities that housed people do not want to see in public. Threading through the coverage is an overarching narrative of homeless people as requiring regulation and control, with implications for their access to citizenship and social inclusion.

Taking action (or not) on homelessness

Participants also used emotion categories to deal with the interactional dilemma presented by their own role, or lack of role, in doing something about homelessness. Participants voiced their frustrations with governmental or other institutional policies on homelessness. One was frustrated that it seemed that a homeless person could only get help if they had a personal advocate (in this case, the speaker). Another was frustrated that so many people were on the street because of an undetected or untreated mental illness. Another was frustrated that there seemed to be money available to pave roads and put up flower boxes, but not to address homelessness.

The speakers, however, did not see themselves as a possible source of solutions to the problem of homelessness. On the one hand, in their words, “You can’t help but feel sympathetic and that you should do something.” On the other hand, homelessness felt “like way too big of an issue to deal with on your own.” The problem, said one, is so huge it is overwhelming. Another said she had been to protests and carried signs, but said “the feeling of hopelessness is almost too much” for her, because activist groups have been working on the problem for a long time with no resolution. Several said they find fundraising campaigns annoying and so do not often give money. Another said the suggestion that she should donate money, when she is a student and is barely getting by herself, “left me feeling bitter. … I don’t know what I can do to help them.”

The use of emotion categories again works back to legitimize both the position that government policy on homelessness is inadequate and the idea that individuals themselves are not really in a position to do something about it. The notion that homelessness is a problem to be solved by someone else is one that is widely, although not explicitly, present in media discourse about homelessness. Schneider et al. (2010) note that more than 70% of quotations in newspaper items on homelessness are attributed to expert sources, including government officials, politicians, homeless agency representatives, and academics. They argue that the predominance of quotations from experts produces a sense that homelessness is a problem to be solved not by ordinary newspaper readers but by professionals and experts (Schneider et al., 2010). Solutions were mentioned in over half of the items studied; but these were not solutions that ordinary housed people can implement—establishing temporary shelters or constructing affordable permanent housing. These are solutions that service providers, politicians, and governments must implement. Ordinary domiciled people, while they may be upset by the existence of homelessness, are effectively off the hook.

Nevertheless, people sympathetic toward homeless people tend to think that they themselves should be doing something. While emotions are often contrasted with rationality, as Edwards (1999) points out, emotions are in fact an aspect of rational accountability. They provide a warrant for the reasonableness of particular actions or thoughts. The use of emotion categories allows speakers to manage the interactional dilemma of the focus group situation by shifting responsibility for addressing homelessness away from individuals, thus making their own inaction seem rationally accountable.


In this article, I have linked how people talk about homelessness to journalists’ reporting practices and to the content of newspaper items on homelessness. I have shown how people use their knowledge about homelessness in interaction with each other in ways that reproduce the various and competing cultural discourses about homelessness and homeless people.

Participants used emotion discourses in the social interaction of the focus group setting to construct moral identities and to manage interactional dilemmas and social accountability. Although the people in our study came to the focus group because they “care” about homelessness, their emotion discourses nevertheless reproduce existing social relations of social inequality and exclusion that are also evident in media representations of homelessness. It is difficult to say whether media discourses produce public understandings, or whether journalists simply take up and recirculate existing public discourses. The media are certainly not all-powerful in the sense understood by media effects theorists, but individuals are also not entirely independent of mediated discourses. As Couldry (2001) articulates, media power is a social process that requires constant reproduction through the practices of social actors at every level of social life, including members of dispersed audiences. Studying social interaction allows scholars to avoid being drawn into the irresolvable debate between media effects and audience reception and to focus instead on interactional practices and how those relate to media representations and larger social formations. Despite the sympathetic intentions of both journalists (Schneider, 2013) and the people in our focus groups, both media and audience discourses about homelessness reproduce existing social relations and inequalities and ensure the persistence of homelessness as a social problem.

In this study, I was able to draw conclusions only from the focus group data I had available. It is possible that if the composition of the focus groups had been different—for example, in gender balance or in ethnicity, particularly if there had been Aboriginal participants—other ways of talking and expressing emotion about homelessness may have emerged. Both gender and ethnicity are not just personal attributes but are aspects of identity that are performed in interaction and may interact with emotion discourse in specific ways. In this study, I did not observe differences in how gender interacts with emotion discourse and was not able to draw conclusions about the interaction of ethnicity and emotion discourse. I suggest both of these as areas for further investigation.

Though media are ubiquitous in the public sphere, the functioning of the public sphere is predicated on interaction among people (Dahlgren, 2005). Studying social interaction, particularly in relation to media representations, is thus essential to understanding how social issues come to be understood in one way rather than another. It allows us to see that while people experience their thoughts, attitudes, and emotions as deeply personal and display them as such in social interaction, this display also reproduces larger mediated discourses of social inequality, even when participants may think and hope that they are not doing so. It also calls into question the distinction between formal political talk and informal talk (Schudson, 1997), demonstrating that even in informal, relatively unstructured talk, such as that among people who care about homelessness as a social issue, larger political discourses are produced and reproduced. As Dahlgren (2003) says, “the political and politics are not simply given, but are constructed via word and deed” (p. 155).


The author would like to thank the Social Sciences and Humanities Research Council (SSHRC) for its support in this research.


Abercrombie, Nicholas, & Longhurst, Brian. (1998). Audiences: A sociological theory of performance and imagination. London: Sage.

Abu-Lughod, Lila, & Lutz, Catherine. (1990). Introduction: Emotion, discourse, and the politics of everyday life. In C. Lutz & L. Abu-Lughod (Eds.), Language and the politics of emotion (pp. 1–23). Cambridge: Cambridge University Press.

Alasuutari, Pertti. (1999). Rethinking the media audience. London: Sage.

Bird, Elizabeth. (2003). The audience in everyday life: Living in a media world. New York, NY: Routledge.

Buck, Philip O., Toro, Paul A., & Ramos, Melanie A. (2004). Media and professional interest in homelessness over 30 years (1974–2003). Analyses of Social Issues and Public Policy, 4(1), 151–171.

Clark, Candace. (1997). Misery and company: Sympathy in everyday life. Chicago: University of Chicago Press.

Couldry, Nick. (2001). The place of media power: Pilgrims and witnesses of the electronic media age. London: Routledge.

Couldry, Nick. (2004). Theorizing media as practice. Social Semiotics, 14(1), 115–132.

Couldry, Nick. (2005). Media rituals: Beyond functionalism. In E. W. Rothenbuhler & M. Coman (Eds.), Media Anthropology (pp. 59–69). Thousand Oaks, CA: Sage.

Dahlgren, Peter. (1995). Television and the public sphere: Citizenship, democracy and the public sphere. Thousand Oaks: Sage.

Dahlgren, Peter. (2005). The public sphere: Linking the media and civic cultures. In E.W. Rothenbuhler & M. Coman (Eds.), Media Anthropology (pp. 318–327). Thousand Oaks, CA: Sage.

Douglas, Mary. (1966). Purity and danger: An analysis of the concept of pollution and taboo. Abingdon: Routledge.

Döveling, Katrin, von Scheve, Christian, & Konijn, Elly A. (2011). The Routledge handbook of emotions and mass media. London: Routledge.

Edwards, Derek. (1999). Emotion discourse. Culture & Psychology, 5(3), 271–291.

Feldman, Louis. (2004). Citizens without shelter: Homelessness, democracy, and political exclusion. Ithaca, NY: Cornell University Press.

Fenton, Natalie. (2007). Bridging the mythical divide: Political economy and cultural studies approaches to the analysis of media. In E. Devereux (Ed.), Media studies: Key issues and debates (pp. 7–31). London: Sage.

Fenton, Natalie, Bryman, Alan, & Deacon, David. (1998). Mediating social science. London: Sage.

Frijta, Nico H. (1994). Emotions require cognitions, even if simple ones. In P. Eckman & R. J. Davidson (Eds.), The nature of emotions (pp. 197–202). New York: Oxford University Press.

Gauntlett, David. (2007). Creative explorations: New approaches to identities and audiences. London and New York, NY: Routledge.

Grossberg, Lawrence. (1992). Is there a fan in the house? The affective sensibility of fandom. In L. Lewis (Ed.), The adoring audience: Fan culture and popular media (pp. 50–68).  London: Routledge.

Hartley, John. (1988). Critical response: The real world of audiences. Critical Studies in Mass Communication, 5(3), 234–238.

Hepburn, Alexa. (2004). Crying: Notes on description, transcription, and interaction. Research on Language and Social Interaction, 37, 251–290.

Hepburn, Alexa, & Jackson, C. (2009). Rethinking subjectivity: A discursive psychological approach to cognition and emotion. In D. Fox, I. Prilleltensky, & S. Austin (Eds). Critical psychology: An introduction, 2nd ed. (pp. 188–219 ). London: Sage.

Hochschild, Arlie. (1983). The managed heart: Commercialization of human feeling. Berkeley: University of California Press.

Hodgetts, Darren, Cullen, Andrea, & Radley, Alan. (2005). Television characterizations of homeless people in the United Kingdom. Analyses of Social Issues and Public Policy, 5(1), 29–48.

Hopper, Kim. (1988). More than passing strange: Homelessness and mental illness in New York City. American Ethnologist, 15, 155–167.

Kitzinger, Jenny. (1993). Understanding AIDS: Researching audience perceptions of acquired immune deficiency syndrome. In J. Eldridge (Ed.), Getting the message: News, truth and power (pp. 271–305). London: Routledge.

Kitzinger, Jenny. (2000). Media templates: Patterns of association and the (re)construction of meaning over time. Media, Culture, and Society, 22(1), 64–84.

Kitzinger, Jenny. (2007). Framing and frame analysis. In E. Devereux (Ed.), Media studies: Key issues and debates (pp. 134–161). London: Sage.

Lewis, Michael, Haviland-Jones, Jeannette M., & Feldman Barrett, Lisa. (Eds.). (2008). Handbook of emotions, third edition. New York, NY: Guillford Press.

Livingstone, Sonia. (2005). Introduction. In S. Livingstone (Ed.), Audiences and publics: When cultural engagement matters for the public sphere (pp. 9–16). Bristol, UK: Intellect Books.

Locke, A., & Edwards, D. (2003). Bill and Monica: Memory, emotion, and normativity in Clinton’s grand jury testimony. British Journal of Social Psychology, 42, 339–256.

Lutz, Catherine, & Abu-Lughod, Lila. (Eds.). (1990). Language and the politics of emotion. Cambridge: Cambridge University Press.

Matsumoto, D., Keltner, D., Shoita, M. N., O’Sullivan, M., & Frank, M. (2008). Facial expressions of emotion. In M. Lewis, J. M. Haviland-Jones & L. Feldman Barrett (Eds.), Handbook of emotions, third edition (pp. 211–234). New York: Guilford Press.

Morley, David. (1980). The Nationwide audience: Structure and decoding. London: British Film Institute.

Pantii, Mervi. (2011). Disaster news and public emotions. In K. Döveling, C. von Scheve, & E.A. Konijn. (2011). The Routledge handbook of emotions and mass media (pp. 221–236). London: Routledge.

Peterson, Mark A. (2005). Performing media: Towards and ethnography of intertextuality. In E.W. Rothenbuhler & M. Coman (Eds.), Media Anthropology (pp. 129–138). Thousand Oaks, CA: Sage.

Press, Andrea, & Livingstone, Sonia. (2006). Taking audience research into the age of new media: Old problems and new challenges. In M. White & J. Schwoch (Eds.), Questions of method in cultural studies (pp. 175–200). Oxford: Blackwell.

Remillard, Chaseten, & Schneider, Barbara. (2010). Picturing homelessness: Visual representations of homelessness in a Canadian newspaper. Social Development Issues, 32(3), 77–91.

Ross, Karen, & Nightingale, Virginia. (2003). Media and audiences: New perspectives. Maidenhead, England: Open University Press.

Schneider, Barbara. (2012). Sourcing homelessness: How journalists use quotations to frame homelessness. Journalism: Theory, Practice, Criticism, 13(1), 71–86.

Schneider, Barbara. (2013). Reporting homelessness: Practice, product, profession. Journalism Practice, 7(1), 47–61.

Schneider, Barbara, Chamberlain, Kerry, & Hodgetts, Darren. (2010). Representations of homeless in four Canadian newspapers: Regulation, control and social order. Journal of Sociology and Social Welfare, 37(4), 147–171.

Schneider, Barbara, & Remillard, Chaseten. (2013). Caring about homelessness: How identity work maintains the stigma of homelessness. Text & Talk, 33(1), 95–112.

Schudson, Michael. (1997). Why conversation is not the soul of democracy. Critical Studies in Mass Communication, 14(4), 297–309.

Snow, David, & Anderson, Leon. (1987). Identity work among the homeless: Verbal construction and the avowal of personal identities. American Journal of Sociology, 92, 1336–1371.

Stets, Jan, & Turner, Jonathan. (Eds.). (2006). Handbook of the sociology of emotion. New York: Springer.

Thamm, Robert. (2006). The classification of emotion. In J. Stets & J. Turner (Eds.), Handbook of the sociology of emotion (pp. 11–37). New York, NY: Springer.

Turner, Jonathan. (2007). Human emotions: A sociological theory. London: Routledge.

van Dijk, Teun. (2000). New(s) Racism: A discourse analytical approach. In S. Cottle (Ed.), Ethnic Minorities and the Media (pp. 33–49). Milton Keynes, UK: Open University Press.

Zillman, Dolf. (1988). Mood management: Using entertainment to full advantage. In L. Donohew, H. E. Sypher, & E. Tory-Higgins (Eds.), Communication, social cognition, and affect (pp. 173–194). Hillsdale, NJ: Lawrence Erlbaum.

  •  Announcements
    Atom logo
    RSS2 logo
    RSS1 logo
  •  Current Issue
    Atom logo
    RSS2 logo
    RSS1 logo
  •  Thesis Abstracts
    Atom logo
    RSS2 logo
    RSS1 logo

We wish to acknowledge the financial support of the Social Sciences and Humanities Research Council for their financial support through theAid to Scholarly Journals Program.