Slaps, Punches, Pinches—But not Violence: Boundary-Work in Nursing Homes for the Elderly

Symbolic Interaction, Volume 25, Number 4, pages 515–536, ISSN 0195-6086; online ISSN 1533-8665.© 2002 by the Society for the Study of Symbolic Interaction. All rights reserved.Send requests for permission to reprint to: Rights and Permissions, University of California Press, Journals Division, 2000 Center St., Ste. 303, Berkeley, CA 94704-1223.

Direct all correspondence to Malin Åkerström, Department of Sociology, University of Lund, P.O. Box114, 221 00 Lund, SWEDEN; e-mail: [email protected].

Slaps, Punches, Pinches—But not Violence: Boundary-Work in Nursing Homes for the Elderly

Malin ÅkerströmUniversity of Lund

This article presents an analysis of boundary work in the context of care forthe elderly, where violence appears to be widespread but is still relativelyunacknowledged. Talk about aggressive patients was formulated in aparticular way among workers in a nursing home. Nursing home staffdescribed how the elderly residents sometimes slapped, pinched, or hitthem. Although staff members could describe these acts as intentional,although they could hold patients responsible, and although this violencecould end in injuries, demarcations were made such that aggressive actswere constructed as somehow not really “violence.” As “violent” is aninherently exclusionary label, this downplaying can be seen as an effort toavoid pushing persons outside the boundary of normalcy and of continuedacceptance. Placing the elderly’s violence outside the boundaries of vio-lence means that the elderly remain “care takers,” the staff “caregivers,”and the nursing home a “caring context.”

The ranked list of occupations that report violence at work is, according to StatisticsSweden, headed by psychiatric nursing staff and police of�cers: 35 percent and 34percent respectively of these workers state in interviews that they have been sub-jected to violence. Six percent of the auxiliary nurses (most of whom work at nurs-ing homes) stated that they had experienced violence directed against them (Statis-tiska Centralbyrån 1991, no. 66, p. 47). In my own quantitative interview study ofviolence and threats against health care personnel, the ranking differed markedly.Seventy-two percent of auxiliary nurses in nursing homes for the elderly and 71 per-cent of mental hospital staff responded af�rmatively to the question, “Have pa-tients ever tried or have they actually hit, pinched, or scratched you?” Moreover,one-third of those working in nursing homes reported that they were hit orscratched weekly, whereas the corresponding proportion in mental hospitals was 14percent (Åkerström 1993).

Why do staff report low rates of “violence” in one investigation but high rates of

516 Symbolic Interaction Volume 25, Number 4, 2002

“aggression” (as measured by the question above) in another? The solution to thispuzzle seems to lie in the ways the questions were formulated. Statistics Sweden’squestion was worded as follows: “During this last year, that is since . . . have youyourself experienced any of the following occurrences?” “Violence that led to visi-ble marks or injuries.” And the follow-up question was “Have you, in addition tothis, been subjected to violence that did not lead to visible marks or injuries?” Asonly 6 percent of the auxiliary nurses responded af�rmatively whereas 72 percentresponded that they had been hit, pinched, or scratched, they obviously did notrefer to such acts of aggression as “violence.”

In fact, nursing staff actively resisted this terminology. In one case when a nurseand an auxiliary nurse talked about the problem of new staff, the following ex-change occurred:

Nurse: We had this one trainee coming into one of our old-timers’ rooms and hetook his cane and [shows how he hit with it] she just ran out of the room. . . .And the next shift there were two of the new girls going in there. And it wasthe same commotion. [Laughter]

Auxiliary nurse: Yeah, some of them [patients] don’t give them [new staff] muchof a chance.

Interviewer: Is violence something you talk about, with the new ones?Nurse: They have to learn that this is not about me. It’s not really “violence,” I

wouldn’t call it that. It’s not directed to you personally.

This article discusses how nurses “normalize” elderly patients’ aggression bytalking about it so that it falls outside the boundaries of “violence.” They achievethis de�nition through different “defusing descriptions,” including, for instance, sym-pathetic accounts of the elderly’s aggression, as well as through claims that these ac-tions do not constitute “violence” even as they construct reasons to explain intent,attribute responsibility, and acknowledge injury. The ambiguity of the concept “vio-lence” is thus illustrated in the context of the nursing home. Staff have plenty ofroom for stretching meanings of violence and actions ordinarily associated with it.Because the occupational setting discussed here is contrasted with other settingswhere such acts are constructed as “violence,” this article demonstrates that “vio-lence” has perhaps less to do with acts of physical abuse than with cultural prac-tices, settings, and relationships. When we label some acts as “violence” and othersas “not violence,” our labels are actually products of de�nitional and interpretiveprocesses.

As Altheide (2000:7) points out, ethnographies belie claims that interactionismcannot deal with macro issues. This study is a case in point: the elderly’s “violence”is more or less invisible on a public societal level (as evidenced, for instance, in Sta-tistics Sweden), and the data from the nursing homes provide insight into the de�ni-tional process that generates this invisibility. In spite of this invisibility, ways of talk-ing about physical aggression—using the word violence or not—is not an issue ofsophistry. Wording holds important consequences. The very process of identifyingsomeone or some acts as “violent” is inherently exclusionary: the actor is cast as

Slaps, Punches, Pinches—But not Violence 517

deviant and made subject to more severe sanctions than if the acts were labeled“not violent.” We can hence assume that in some situations one wants to avoidpushing persons outside the boundary of normalcy and of continued acceptance.Placing elderly patients’ violence outside the boundaries of violence means thatthe elderly remain “care recipients,” the staff “caregivers,” and the nursing home a“caring context.”

THEORETICAL BACKGROUND

To analyze the distinctions and demarcations between that which staff consider vio-lence and in this case especially that which they do not, I will use the concept“boundary-work” (Gieryn 1983, 1995, 1999). Boundary-work is the efforts of de-marcation in which people become involved, to distinguish and separate activities,phenomena, objects, conditions, or people from each other. Through this workpeople bring some social objects inside a category or concept whereas they pushothers out of the de�nitional frame of this category or concept. Boundary-work is auseful concept because it allows us to see how the social construction of violenceproceeds by drawing lines or making distinctions between “violence” and “not vio-lence,” as, for instance, through the efforts and the work of distinguishing “a slap”from “violence.”1

In recent years, boundary-work in different areas has been analyzed from a con-structivist perspective. Researchers have studied how people construct boundariesaround various activities, conditions, or types of people. Gieryn’s studies centeredon the �rst: the boundaries constructed through “credibility contests” concerningsome intellectual activities as constituting science in contrast to nonscienti�c ortechnical activities. His approach is relevant to the issue of how a condition like vio-lence is constituted or contested, because it shows how the de�nitional process en-sues. However, other studies that do not use the concept of boundary-work demon-strate how distinctions are collectively constructed and maintained. For example,Loseke’s (1992) analysis of the construction of who is or is not a “battered woman”pointed to the importance of distinctions between types of people—an important as-pect because the term “violence” takes as given typi�ed actors: victims and villains.

Analyses that explicitly use the concept of boundary-work to make discursive ordescriptive distinctions among types of people include Berbrier’s (1998) study ofthe meaning of “Deaf” in the Deaf culture movement. This study illustrates howpeople may assign identities in ways other than the outside world may expect. Ad-vocates for Deaf culture argue that to be Deaf one must exhibit an “attitudinaldeafness,” acquired by embracing American Sign Language and by attending resi-dential schools for deaf children. Audiologically deaf people, on the other hand,who have no political commitment, personal identi�cation, or childhood socializa-tion into Deaf culture are simply “deaf.” They “might even be classi�ed as ‘hearing’in this scheme—that is as people who are audiologically deaf but culturally hear-ing” (Berbrier 1998:92).


Analyses of how people use boundary-work to make distinctions among them-selves also �t studies of interaction per se. In Gender Play, Thorne uses the concept“borderwork” to analyze how boys and girls, through contests, chasing, invasions,and tattling, activate gender boundaries and consolidate “the boys” and “the girls”into separate and rei�ed groups. Thorne is inspired by the anthropologist FredrikBarthe who studied boundaries between ethnic groups, and hence she argues that“one can gain insight into the maintenance of ethnic (and gender) groups by exam-ining the boundary that de�nes them rather than by looking at what Barthe calls‘the cultural stuff that it encloses’” (Thorne 1999:65).

Allen (2001) has argued for extending the concept of boundary-work to studyprocesses in the workplace, a relevant point for my study as this boundary-workwas also performed in an occupational setting. Her case involves the study of divi-son of labor among medical and care practitioners where “atrocity stories per-formed a dual boundary-work function—they constructed a boundary betweennursing and other occupations” (Allen 2001:98). Instead of the more diffuse con-cept “negotiation” that interactionists commonly use in their studies of demarca-tions between occupations, Allen argues that the analysis of more detailed mun-dane workplace talk can illustrate how boundary-work is accomplished.

My particular interest in boundaries in the workplace concerns the “talk work” toconstruct an act as an instance of “violence” or as “not violence.” Representatives forprofessions in the judicial system also may have an interest in de�ning acts so thatthey fall inside or outside the boundary of “violence.” Emerson (1994), for instance,has analyzed how staff at legal aid clinics tried to ensure that applicants for domesticviolence restraining orders �led the strongest possible petition. In interviews with cli-ents the staff searched for and identi�ed accounts of events involving physical vio-lence. They tried to establish whether these events were frequent and inquired abouttheir duration. Choices of words used in the application were important; descriptivephrases might thus be transformed by follow-up questions about “punching” whenthe clients had used the word “slapping.” Furthermore, in their written reports, thestaff would avoid words like “�ghts” or “quarrel” as this would convey mutuality; itshould appear that a victim is at the receiving end of unilateral aggressive acts. Staffadvocates would furthermore highlight consequences of violence (Emerson 1994:16–17). We �nd a similar process in Sarat’s (1993) study of a capital trial. Sarat states thatviolence is dif�cult to represent and pain is invisible, so asks: how was this done dur-ing the trial? He �nds that prosecutors emphasized detailed descriptions of woundsand weapons. In his effort to bring violence into discourse and to represent pain theprosecutor portrayed the sufferings of the victim: “[A]s wounds on the body, Janine’s[the murdered girl] body was partialized, objecti�ed, and marked in discrete ways[;]. . . violence is attached to particular parts of her body. It is inscribed as a hole in theneck, bite marks on the breast, blood from her private parts” (Sarat 1993:31–32). Thedefense attorney relied, in part, on the same tools—relating details about wounds andweapons—when trying to make an appeal for the accused through a detailed descrip-tion of an earlier story of violence that involved regular beatings by his stepfather.


Boundary-work on “violence” in occupations may also incorporate practices toexclude certain acts, as when workers are involved in potentially questionable ag-gressive actions. In such cases, workers evince interest in de�ning actions so thatthey fall outside “violence.” Hunt (1985:317), for instance, holds that what the pub-lic would regard as violence, the police often describe as “normal force, . . . depictedas a necessary and natural response of normal police work to particular situationalexigencies.” In my study, nursing staff also downplay aggression. This is the case,even though it is the clients and not the behavior of the workers themselves thatcome to the fore.

My analysis of boundary-work differs from other researchers who have explicitlyused the concept. This study does not concern an area of controversies as Gieryn’s,Berbrier’s, or Allen’s studies did, and the claims I study are not constructed in pub-lic or political settings as Gieryn’s or Berbrier’s were. Similar to Allen’s, my study il-lustrates the “micropolitical business that talk does” (2001:76) but with an implicitrather than explicit form of boundary-work.

METHODS

My initial interest in how nursing home staff does boundary-work on the topic of vi-olence by elderly patients began with previous qualitative research conducted in arange of health care settings. This research about violence and threats againsthealth care personnel in different contexts (nursing homes, emergency care, outpa-tient care, and mental hospitals) demonstrated that when asked which patientswere aggressive, health care employees mentioned the elderly (Åkerström 1991).When the Federation of Swedish County Councils asked me to do a quantitativestudy to measure how often, when, where, and so on, the staff reported such aggres-sion, I decided to ask about speci�c acts (acts that had been mentioned in earlier�eldwork) such as whether they had been hit, pinched, or scratched by patients,rather than about “violence.” Comparing my quantitative study with other investi-gations, as I indicated above, explicit use of the word violence in interview studies ofoccupational groups revealed different ways of making sense of incidents thanwords describing speci�c aggressive acts. When staff used the term “slap,” “pinch,”or “punch” instead of “violence,” they downplayed—as opposed to police or courtstatements in which concrete terms may be used to invest actions with a sense ofdrama (Emerson 1994; Sarat 1993). “Violence,” apparently, has connotations thatcannot simply be equated with those of punches, shoves, slaps,” and the like.

This study draws on this discovery and involves a reanalysis of the qualitativematerial, which consisted of group interviews, individual interviews, and �eld obser-vations, as well as some material collected later.2 The data used here are from inter-views with three groups of people. First, eighteen people were interviewed infor-mally (three nurses and �fteen auxiliary nurses) in two nursing homes during 1990.Second, interviews with twenty other employees who did not currently work innursing homes yet either worked in a context where they cared for elderly as well as


other patients (emergency and out patient clinics) or had previously worked innursing homes have also been included in this analysis. Although my analysis reliesprimarily on the �rst group, this second “sample” provided interesting data, espe-cially in terms of comparisons, analogies, and contrasts between different categoriesof patients.

These interviews, ranging from one to two hours, were tape-recorded and looselystructured and were carried out in available nursing home rooms suggested by thestaff. They contained open-ended questions starting with work in general and pro-ceeding to questions about “dif�cult patients,” and—if the subject had not yet comeup—the topic of aggressive patients. In early analysis of these interviews, it becameobvious that the topic of “dif�cult patients” and “violence” was not straightforward.Interviewees’ wordings and ways of describing aggression were interesting in theirown right. This led me to seek data that would be more sensitive to the complexitiesof boundary-work with regard to violence by elderly nursing home residents.

A third group of staff members was selected in January 2001. I interviewed anauxiliary nurse working at a nursing home who, in turn, was asked to conduct infor-mal interviews with four other colleagues who worked at nursing homes differentfrom those in which we had previously conducted interviews. She carried out the in-terviews in the homes of each respondent, and although the interviews focused onaggression only, they lasted from one to one and a half hours each.

In all interviews I used “active interviewing” (Holstein and Gubrium 1995) so asto elicit accounts that would highlight the border-work being done on physical ag-gression in this nursing home culture. This strategy permits what researchers tradi-tionally see as “leading” questions, because they form a natural part of any conver-sation; such questions evoke re�exive responses. In active interviewing “thestandard concern for contamination is replaced by the awareness of activeness. . . .The point is to capitalize on the dynamic interplay between the two to reveal boththe substance and process of meaning-making in relation to research objectives”(Holstein and Gubrium 1995:76). Interviewers, along with the interviewees, maycodify the meaning of various phenomena and pinpoint signi�cant features in alocal culture. In this case I, as well as the staff, became engaged in boundary-workwhen, in response to the telling of an incident, I would ask, “Is this violence?” Inone case a nurse at an outpatient clinic compared different “dif�cult” patients,some of them dif�cult because they were violent. I interrupted her and asked: “El-derly people who strike out at them [she had given one such example earlier], arethey violent?” She responded: “No, I don’t call that violence,” and explained why.Such accounts provide the basis of my analysis.

Because the nurses did not use the word violence, the only way to group howthey did their boundary-work was to ask outright. At times, as in any ordinaryeveryday conversation, this meant that we posed questions as spontaneous inter-ruptions but not, however, as drastic interruptions involving a change of topic.Rather, we asked probing questions suggesting that we would investigate the namingof the events and actors currently discussed.


Summing up, the physical acts encountered by auxiliary nurses working with theelderly might be similar to those encountered by occupational groups such as policeor mental nursing staff. However, they are not fed into public statistics because thenurses do not characterize these acts as “violence.” These acts are invisible in policeand court statistics, and barely discernible in statistics on “exposure to violence andthreats” published regularly by the National Board of Occupational Safety and Health(whose list is headed by mental nursing staff).

DEFUSING DESCRIPTIONS AS BOUNDARY-WORK

In this section I examine the various linguistic ways nursing staff talked about inci-dents and patients as boundary-work that involves a variety of downplaying strate-gies. This means that instead of highlighting the physical abuse, staff toned downthese actions in various ways. This boundary-work is not always manifested throughdistinct and clear-cut arguments. But the staff, through choice of wording and styles oftelling, still mark borders so that they keep “violence” apart from their situations. Si-multaneously, they construct physical aggression as slightly dramatic and as part ofthe routines of everyday occupational work that the nurses can handle professionally.

Categories of People Rather than Incidents

Through categorizations of other people and not their actions, individuals drawboundaries around phenomena. A conversation from staff members at an emer-gency room admission desk illustrates this tendency to assign offender status totypes of persons rather than according to their actions. I asked the staff about el-derly people who struck out at them, “Is that violence?”

Nurse 1: No, I don’t call that violence.Nurse 3: No, if an old man of eighty clocked me one I don’t suppose I’d have,

well, of course I’d have said to him, “What do you think you’re doing?” But Iwouldn’t have acted like I sometimes do when a forty-year-old drunk dots meone. That’s a different thing. Or a young guy or someone like that, or a girlwho’s had too many. But not an elderly person, no, I don’t call that violence.

Nurse 2: I’d agree with that.

At this point, I interrupted to ask the nurses about a situation they had referredto earlier involving an elderly patient who had given one of them “a box on theear.” Was that violence?

Nurse 2: No . . . but after all there’s a difference when an eighty-year-old whobarely has enough strength to lift his hand to slap your face, compared towhen a thirty-year-old does it.

Interviewer: Is it just a matter of the strength of the blow, or is it that they are . . . ?Nurse 2: Partly of course it’s that they’re old; they’re confused, and some don’t

have all that many brain cells left. Even when they’re not confused, they’renot quite normal. But if it’s a younger person, well, that’s different. That’s notthe same thing.


Interviewer: Then you may expect a different kind of behavior?Nurse 2: Yeah we must, regardless of whether they’re drunk or not, or high on

drugs or whatever they are when they get in here.Interviewer: So one can’t make excuses for drug addicts either?Nurse 1: No, that’s violence.

Even when the interviewer interrupted by asking “What about that box on theear?” the interviewees return to the issue of who delivered it—“the drunk,” “a youngguy,” “a girl who’s had too many”—and contrast the individual with “an elderlyman of eighty.” The nurses attend to a common cultural script: when members ofthe general public are asked to describe a person who uses violence, what immedi-ately comes to mind are youngsters and young men, drug abusers or alcoholics, andthose engaged in criminal activities. Criminologists likewise focus on these catego-ries of social actors (Barlow 1987; Fattah 1991; Wikström 1992). In addition, the“ideal victim” is an elderly woman, according to a classic criminology article(Christie 1986). This tendency to determine what a phenomenon is according towho is involved instead of what he or she does is not restricted to this study. Manyempirical studies dealing with very different situations and phenomena have shownthe same tendency, including, for example, Emerson’s (1966) study on court staff’sdecision making in work involving juvenile delinquency, Holstein’s (1993) investi-gation of determining “insanity” in involuntary commitment proceedings, andLoseke’s (1992) demonstration that “battered women” are not simply women whohave experienced violence.

By focusing on elderly people as violent offenders, this study reveals a situationin which types of people do not �t the image. Not only do nurses have dif�culty�tting the condition (violence) with the type of people that elderly are supposed tobe, but images of elderly people as victims in�uence researchers. For example, a re-searcher looking for studies on violence and the elderly will be directed to work onelderly people as victims. The searchword combination “violence/elderly” yieldsonly articles on elderly as victims in criminology, penology, and police science ab-stracts. As Holstein and Miller (1997:26) have argued, “Conventional victimology, itappears, presupposes that some persons or groups are objectively ‘victims’ withoutexplicitly considering the interpretive de�nitional processes implicated in assign-ment of victim status.” Researchers are thus no different from people in generalwho invariably use what Gar�nkel (1984:76–103) referred to as the “documentarymethod of interpretation,” a circular process whereby ideas about etiologyin�uence the way in which problems—that which is to be explained—are ap-proached. In other words, our initial assumptions form the framework within whichwe account for the world around us. We “see” violence more easily in certain placesand in some types of people.

The contrasting images of types of people that the nurses put forward constituteone kind of downplaying.3 Elderly people’s physical aggression does not belongwithin the framework of violence, not because their actions differ, but because ofhow elders are typi�ed. The elderly are associated with goodness, wisdom, and, in


the case of the very old, frailty (Baumann 1989). Moreover, while people often as-sume “addicts” and “drunks” somehow have chosen what they are, the elderly havenot and thus are more “sympathy-worthy” (Loseke 1999:115–16). Yet if one looksmore closely at the associations of the nurses’ typi�cations—“they are confused,”“even when they’re not confused, they’re not quite normal”—exactly the samecould be applied, in a different context, to categories of people expected to be proneto violence: drug addicts, inebriates.

Sympathetic Accounts

One way of dramatizing violence is to demonize the perpetrator (Sarat 1993). Inthe present context, descriptions are often of the opposite kind: nurses explain el-derly people’s aggression in mainly sympathetic terms. Staff members talk aboutthe motives underlying elderly people’s aggression in a way that normalizes it,makes it comprehensible, and morally justi�es it. Referring to two major thematicdimensions, this study shows how nurses become involved in character work inways that excuse the elderly of problematic behavior. In the �rst form of accounts,aggression is understood as a “natural response” in relation to the behavior of thestaff or the circumstances at the nursing home. In this case staff are appropriate ob-jects of aggression. In the second form of accounts, aggression toward the staff isnot “really” directed toward them; staff describe themselves as “the receiving sym-bols” of something else.

Aggression as Response to the Immediate Situation

The staff sometimes talked about aggression as a normal reaction to those in-fringements of integrity that form part of geriatric care: “But of course I can under-stand if you, say, have to wash and dress someone below. That they, men, that they. . . maybe they have tender spots, they’re sensitive and . . . That they don’t exactlyenjoy it.” Other situations in which aggression was said to occur was when puttingpatients to bed, putting diapers on them, or feeding them.

Auxiliary nurse: We have this lady who’s begun to hit us. When you’re puttingher to bed, change diaper, wash her or feed her or whatever she’ll punch you.So you’ll have to take off your glasses for example. Or you can sit and feedher and suddenly both the plate and everything come �ying.

Interviewer: How come?Auxiliary nurse: She’s deteriorating; she used to be able to take care of herself.

No one wants to have diapers, or be fed—but we have to do it, you know.

Another motive stated by the respondents was that aggression was a response topain that some patients experience as the staff perform various care work. For ex-ample, an auxiliary nurse said:

Rheumatoid arthritis is terribly painful, you know. Many of them are in tearsand feel bad about it hurting so much: “Ow, ow, ow.” But others become angry.


They don’t quite understand that we have to turn them and put on ointment. . . .But if they’re angry you can’t say it’s violence, because it’s responding to theirpain.

Staff also made the situation understandable when they describe the elderly asreacting to a new situation when foreign demands are forced on them. In these cir-cumstances, staff members argue, it is “wholly natural” for the elderly to react withanger when, for instance, they are forced to take a shower:4

[T]hey aren’t keen on water. Or not so used to it, as we are nowadays. Just thinkof all the showers we take. But elderly people aren’t that used to having bathsand showers, so it . . . naturally they give some sort of signal. (Auxiliary nurse)

Through such formulations, nurses place the elderly’s actions within a forgivingframework: aggressive actions are expressions of human nature, imperfect but un-derstandable. Again, their choice of words is signi�cant. The woman quoted aboverefers to how patients give “some sort of signal.” The euphemism is interesting in it-self. Violence in this kind of context is sometimes interpreted as a form of commu-nication. Another auxiliary nurse expressed it in the following manner: “They giveall the signs they can, you know. Both by saying things, but also by striking out andkicking. And biting and scratching and . . . Yes, well, they do make use of all their . . .”Sometimes anecdotes of patients start with descriptions of their aggression and endwith an appreciative comment, as in this nurse’s story:

We had an old senile lady who used to hurl insults at us in German for some rea-son. You had to look out for her teeth, hands and the pinches and so on, too, butat the same time I thought she was great. . . . She had some action in her. She camealive then, otherwise she sat mostly in her chair, she was ninety-six or something.But when she was on the go, she could start spinning and tell about the dinnersat the manor where she lived when she was young. In her case, the shouting andswearing showed there was life in her. So there are positive sides to it too.

The physical aggression of the elderly may thus be seen as life-af�rming. Nursescan talk about it as a sign of showing that they have not given up (Magnusson 1996).An auxiliary nurse explained, “Why, they’re ill. Of course it’s not nice that they �ght,but I mean, it’s great that they show their feelings, that we can see that they havethese aggressions. That’s a positive thing, after all.”

The Staff as Vicarious Punching Bags

In the other form of accounts, the staff describe themselves as “surrogates” or“symbols,” of convenient objects to whom patients may express aggression thatarises from anger or fear about something outside the care environment. For exam-ple, anger may be ascribed to a fear of death. As one auxiliary nurse states, “Andalso a lot of it’s ’cause they’re afraid of dying. So they take it out on the staff.”

Another theme consisted of constructing patients’ past lives. The nurses used home-sickness, for example, to account for aggression. Given the contention that nobody


wants to be in an institution—“we all want to go home, after all”—the nurses thennormalize their patients’ anger. In conjunction with the patients’ limited memoryand inability to grasp the impossibility of living at home, staff frame their reactionswhen hindered from leaving the nursing homes as expressions of natural frustra-tion. Furthermore, nurses describe these patients’ former relationships as crum-bling away; thus staff frame disappointments due to family members’ failure to tele-phone or visit as causing patients’ anger.

[I]t can be their own feelings about their families. That perhaps a family memberhasn’t been in touch and they’re lying there, waiting for a phone call. And thenwe turn up and say something like “Now it’s time to get up.” Then of course ithappens, we get all the aggression that wasn’t really meant for us. (Auxiliarynurse)

Nonetheless, staff also described the elderly as giving vent to bitterness and dis-appointment dating back to their earlier years. Staff members argue that elderly pa-tients permit themselves to do things that were previously impossible. In the follow-ing case, members of the nursing home staff once again depict themselves asrecipients of aggression that patients “really” meant for others. The staff seems toregard themselves as vicarious punching bags:

The act of vengeance, that’s several patients. It isn’t directed against those thatmaybe they’ve got something [against]. It’s directed against us. I mean there areother people who �ght back, so to speak. But those they’d like to pay back, theycan’t reach. And perhaps never could. And then we’re the ones who get theknocks. (Nurse)

Sociologists writing about “accounts” often focus on how individuals explaintheir own behavior, opinions, or relationships through talk or other behavior, orhow they make face-saving, remedial work in interaction (Buttny 1993; Scott andLyman 1968). Sometimes such accounts re�ect an interactive achievement when in-dividuals align their own explanations with that of a group, a local culture, or collec-tive (Rossol 2001); sometimes people construct accounts to defend opposing ver-sions of commonly adhered to understandings (Järvinen 2001). However, peoplealso get involved in accounting for others’ behavior, opinions, or relationships(Antaki 1987; Buttny 1993:15). The nurses’ explanations of the elderly’s aggressive-ness is an illustration of this practice.

Telling it with Laughter

Laughter and humor have been analyzed as drawing on a sense of incongruity(Katz 1999; Meyer 1997). What humor thus portrays in the accounts supplied by theauxiliary nurses is the clash between the interaction of “quiet, placid elderlypeople” and the competent nursing staff, on the one hand, and the picture of whatcan in fact happen, on the other. What humor achieves in our interviews is commu-nication of the contention—sometimes expressed among peers—that physical


aggression perpetrated by the elderly is nothing to get excited about; it is somethingthey can handle.

Different cultures develop particular conversational styles (Coates 1993; Tannen1981). A jocular tone, witty turns of phrase, and sarcastic challenges form part ofsuch styles in occupational cultures (Grif�ths 1998; Pogrebin and Poole 1988; Seck-man and Couch 1989). Members of a local culture may also deal with speci�c topicsin a jocular tone. In the present study, this was often the case when elderly people’saggression was discussed. Staff could construct accounts of particular patients, forexample, in a slightly facetious manner.

We had a man who was passed on to us from a geriatric long-term ward. Nowthat was a splendid specimen, I can tell you. He was mean, through and through.. . . And it didn’t matter what you had to do with him, there were always handsthere, and feet and a head and . . . goodness, yes. . . . If he sat in a chair and youwere putting his socks on, you could be sure of getting booted so you wentsmack into the wall [laughter].

This nurse used humor to underline the drastic dimensions of the story. But drama-tizing involves the situation as a whole, not the aggressive acts. If the latter hadbeen the primary object of the dramatizations, the nurse could have elaborated onthe concrete injuries and the problematic state of affairs, the social climate, theworking environment, and so on.

The laughter following the telling of an anecdote worked in many cases as arounding off, a conclusion that served as a comment on what the person had nar-rated (Gavioli 1995). In the talk that we overheard among nurses, this jocular styleoften evolved, indicating its support in local nursing cultures. In one group inter-view, for instance, the nurses told stories of patients who hit other patients. The su-pervisor �nished the exchange in this way: “Now that makes me think of that manwe got during my time. The one who was always sticking out a leg, tripping the pa-tients. We’ve never had so many fractures as we did during his time [laughter].”

Humor and laughter are, however, not always the same. Staff comment on somethings in a humorous way, but no laughter follows, as when nurses mention thatthey ought to demand a “scratch bonus” when working with especially aggressivepatients. Other times, laughter may accompany a story that is not told in a humor-ous way. For example, a kind of laugh signaled something else, something akin toembarrassment. The nurse below spoke about a man who kept running away fromthe nursing home. She had to go after him by car to try to pick him up:

Then you had to get him inside the car, either to drive him to Lummeboda [hishome] or back here. You really got it then. Kicks as well as punches. I got reallybeaten by that old man (farbrorn). Sometimes he managed to really deliver a, itwasn’t easy to hang in there and to receive (ward off). You could get realpunches [slightly smiling, small laugh] in the face.

This nurse spoke in a mild, kind tone; she used the respectful term farbror (whichliterally means “uncle” but is also used more generally as a polite and deferentialterm for old men). Her smile and laugh indicate another kind of de�nition than the


humorous laughs accompanying the jocular style that nurses use when relating an-ecdotes of aggressive patients. But her stance is still a form of downplaying. We caneasily imagine how other occupational groups dealing with other types of clientswould come across describing similar incidents (moral indignation, cynicism, or anattitude of masculine stoicism). The nurse’s response is an embarrassed laugh intelling a story about the public breakdown of control and competence. Right aftertelling the story above she continued to speak about other drivers going by, staringat her dressed in her white uniform trying to overcome an old man. No one helpsher but only glor, bara blänger (stares, glares a bit accusingly). This story thus con-veys feelings in a situation in which the outside world gets a peek at something thatis usually not talked about.

As Yoels and Clair (1995) have argued, laughter and humor may be seen as acentral dimension of social organization. Based on observations of interactionsamong physicians, nurses, and patients, they suggest that humor “often embodiesaspects of interactional con�ict in which ‘higher-ups’ af�rm what they see as theirrightful claims to control the work process when dealing with ‘lower-downs,’ whilecommiserating with one another about arduous, burdensome aspects of their workthat are little appreciated by outsiders” (1995:56). Whereas this may well be so,humor—as has been discussed here—may also communicate competence and con-trol; the patients are not “violent” and the nurses are still caregivers. The nursesshow humor in their stories rather than treat the matters as serious and demandingproblems of their work that outsiders do not appreciate and fully realize. On theother hand, laughter may accompany humorous stories and thus support the mes-sage of competence or in the case of embarrassed laughter, loss of control. In bothcases, though, nurses do not tell stories to communicate moral indignation, anger,or fear.

INTENT, RESPONSIBILITY, AND INJURY?

A common cultural construction of violence associates it with intention and respon-sibility on the perpetrator’s part and with injury on the victim’s (Arblaster 1994).These themes reoccurred in responses to the presentation of my �ndings. More-over, they appear in my data. When investigating the staff statements in light ofthese themes, a somewhat surprising picture emerges: the downplaying is no longerconsistent, as intent, responsibility, and injury are partly acknowledged. In spite oftelling about suspected intent and responsibility and actual more or less signi�cantinjury, the staff still drew the line so that the aggressive acts fell outside “violence.”

Intent?

Intent, that is, actions with design and purpose, concerned the staff. On the onehand, nurses framed elderly people’s aggressiveness as nonintentional expressionsof illness and confusion. In other words, nurses could frame incidents such that it


was the illness—not the person—that did the boxing, kicking, or biting. Staff work-ing in other care settings (outpatient and emergency clinics) commonly adhered tothis perspective when they compared their elderly patients’ aggression with that ofother patients. I also found such accounts in the general talk at the nursing homes.

But we could observe a degree of hesitancy as well. It seems as though the delib-erateness of patient actions is the object of continuous interpretive work. As Ding-wall and Murray (1983) argue in an investigation of emergency health care, intent—what they term “theoreticity”—is not easy to distinguish. Among nursing homestaff, the general view “they don’t know what they’re doing” is revised in concretecases. Are Astrid’s bites and pinches deliberate or not? It is not always completelyclear. One auxiliary nurse talks about a brain-injured patient who can “bite andpinch and pull your hair and that sort of thing.”

And it’s the same sometimes when she gets hold of your arms and then shepinches. I had large bruises here before, that I got after she’d pinched me. Sothen I was there the next day and said: “Do you know what that is?” Yes, sheknew all right. So I said: “Who did that?” “Me,” she said. So she knew she wasthe one who’d done it.

Another staff member told a story of a ninety-year-old man who had lashed outwildly when on an excursion. He had talked about seeing rats and said the staffwore green twigs in their hair. This, of course, could lead staff to argue, “He didn’tknow, has no functioning consciousness.” Even so, there was scope for gnawingdoubts: “Sometimes I think, well, it could have been a bit of sheer mischief. . . . [I]tcould have been on purpose. But that’s not really something you want to admit.”

Finally, nurses may interpret the fact that patients say they are “sorry” as indicat-ing that a violent action was deliberate. In other circumstances, an apology is likelyto be regarded as a sign that the guilty party has admitted a moral defect, where-upon he or she is excused (Goffman 1972; Hepworth and Turner 1982). In thepresent context, however, the act of apologizing is taken as evidence of guilt:

And this patient, why I think it may be a little bit deliberate: that’s because then,when you try to talk to her, when she’s punched you in the stomach and you’vedoubled up and say to her: “Now that hurt when you hit me.” “Yes, I’m sorry,”she’ll say. (Auxiliary nurse)

Responsibility?

The apportioning of responsibility appears to be a central theme in representa-tions of violence. To determine who may be held accountable, guilt and “victim-hood” must be clari�ed. Emerson (1994) has observed that people frequently stressthe element of one-sidedness in the words they choose. For example, in certain judi-cial accounts, actors may thus avoid a word such as �ght as it implies the interactionof two parties.

Was that the case here? Was any attempt made to apportion the blame to oneparty? Talking about nursing homes in general, responsibility could be placed on


the staff. Nurses believed that the climate on wards and a staff’s way of dealing withpatients would increase or mitigate patient aggression. Staff abuse of the elderly is atheme much discussed in the Swedish media, as well as in research from other coun-tries (Baumann 1989; Stannard 1973). The staff sometimes mentioned such prac-tices, blaming it on unsuitable staff: “Some nursing homes have staff that are notsuited to do this work, they don’t have the patience, they might overreact—then it’sonly natural that patients ‘answer back’” (Head nurse). Workers in the nursinghomes I studied claimed that well-educated staff and peace and quiet were essen-tial, as was keeping staff changes to a minimum. Moreover, helping hands should al-ways be the same hands, and staff should have time to sit and talk with their pa-tients. The staff members we talked to claimed that these characteristics typi�edtheir own institution when comparing it to others, where patients were said to bemore aggressive. Moreover, the interviewees claimed that the other institutionsadhered to a different philosophy of care, one that was placing greater stress oncleaning and purely physical nursing. “That’s not the way we do things here!” Thusthe responsibility may in theory rest with the institution, but we provide the bestpossible climate. In other words, the violence that exists here is not our fault.

If the staff describe themselves as not responsible, is the blame then laid on thepatients? Nurses mostly couched patients’ actions in exculpatory, tolerant rhetoricwith the major theme that their illness (“dementia”) caused their actions, that theydid not act as fully human people with intent to harm. In this sense the elderly couldnot be held responsible. Yet this did not hold for everybody. Some were declared“not confused”; they were just “mean to the bone,” “dif�cult,” or “provoking.”

You can’t help getting irritated when you deal with those you know understandwhat you’re saying, and you do something that don’t hurt them like helpingthem in the bathroom. You’re really careful when you do it and have told themin advance, they still pull your hair or twist your thumb.

The construction of responsibility is coupled with “accepting the consequences.” Inthis case the interviewee portrays a situation in which becoming irritated is morallyacceptable since she does not attribute the aggression to the patient’s illness (she“understands”), no pain is involved, and it is not a result of the actions of the staff .

Similarly, nurses may describe moralistic rebukes as legitimate, or at least under-standable, in relation to those considered responsible for their actions:

When I �rst got here and encountered it [aggression] for the �rst thousand times,I felt sorry for them and really tried to keep calm. You wanted to see it as an ill-ness. But we’re not more than humans: After a while you too begin to say thingslike “No that wasn’t nice, was it?” [in a stern voice] to those you think are notconfused.

The nurses thus construct situations in which blame on the part of the patient canbe morally justi�able. By doing this, they explain the limits of what they will excuseor allow. In this construction of at least partial responsibility nursing home staffseem to take an institutional perspective, involving a translation of clients’ behavior


into moralistic terms, similar to that of staff at the mental hospital studied by Goff-man (1961:86–87): “Although there is a psychiatric view of mental disorder. . . . free-ing the offender from moral responsibility for his offense, total institutions can littleafford this particular kind of determinism. . . . [B]oth desired and undesired conductmust be de�ned as springing from the personal will and character of the individualinmate himself, and de�ned as something he himself can do something about.”

Injuries?

The literature on physical aggression that has been subjected to boundary-workoften refers to injury as one criterion for the seriousness of violence. For instance,attempts have been made to distinguish among normal child rearing, corporal pun-ishment of children, and child abuse by looking at the severity of the injuries (Gil1970). When responding to presentations of my data, people have often accountedfor staff not using the label “violence” by claiming that the elderly are so frail thatthey can hardly cause any injury; in other words, the actions of the elderly cannot beconsidered violence. The downplaying observed in this study could thus be associ-ated with the actual nature of the violence: it was not very severe; that is, it had notangible consequences.

However, even if most injuries mentioned in the interviews are minor ones, thestaff does talk about injuries that generally are regarded as relatively serious, suchas femoral fractures, or consequences such as “going smack into the wall,” “head-butting,” “kicks,” and “punches.” Furthermore, nursing home staff make counter-claims concerning the elderly’s strength. Instead of descriptions of fragility andweakness, we repeatedly ran across statements that implied “the seniles” had spe-cial physical strengths. A few examples: “They are so strong in their hands,” “If youallow them to twist your thumb, you’re done,” and “You get a lot of bruises causethe seniles don’t let go. They pinch so hard, give it all they have.”

The informal interviews also showed that nursing home staffs often describetheir reactions as “emotional injuries.” One question in the quantitative interviewstudy was concerned with this dimension. Results revealed that nearly every fourthnursing home employee stated that they had been afraid of certain patients. In ad-dition, one-third reported that they had become angry with patients, and one-�fthreplied that they had become sad and upset when asked to describe the most recentsituation of aggression in which they were involved (Åkerström 1993). Often staffmix the two forms of injury—the physical and the emotional—in their descriptions:

Nurse: There was this old man who hit me with a cane, he fought with his cane,and hands and �sts, and everything. He just went crazy. . . .

Interviewer: Were you hit?Nurse: Yes, with the cane. If you want to keep someone away from you it’s real

effective to stick it into someone’s belly. . . . I kept my distance for a longwhile, didn’t dare to be close to him.

Frequently, as above, staff only imply their injuries (we have to imagine whatdamage the cane in the stomach in�icted) and seldom dwell on them. At times they


mention injuries in connection with other themes. At one point in an interview wetalked about whether staff members reported incidents as occupational injuries:“Once when I was hit in the eye, and it was sore for a while, I thought about it. Incase there are long term damages, but I never bothered, too many forms to �ll in[laughter].”

The most vivid descriptions of injuries concern patients or other occupationalgroups, not the staff. Two auxiliary nurses, for example, talked about the late after-noons that they consider a peak time for aggression:

Nurse 2: Then it doesn’t take much before they explode, to put it plain.Nurse 1: We have this one man who’s as clear as you and me, and then these two

very confused ladies, he’s banging them around, he gets irritated cause they, es-pecially one of them, gets a grip somewhere and don’t let go. They’ve beenblack and blue all over, one of them went smack into the wall the other day. . . .But the worst thing, you know who [turns to the other nurse, who nods] whenhe kicked the cleaner into the closet. He was lying in his bed and “tjiuff.” Sheruptured her spleen.

DISCUSSION

Underlying the term “violence” are a welter of images, feelings, and assumptions. Thesecan all be used, invoked, danced around with, emphasized, or downplayed so that ag-gressive actions are sorted inside or outside the cultural boundaries of “violence.”

Gieryn (1995) has suggested that studies on boundary-work would bene�t fromanalyses of exclusion, expansion, and monopolization, as boundaries are contextu-ally contingent and interest driven. Violence provides a rich and fertile ground forsuch studies. Monopolization is a process aimed at protecting the “ownership” of asocial problem (Gus�eld 1989; Loseke 1999:74–75). Leaving my own empirical casefor a moment, one can assume that some occupational groups have a vested interestin ensuring that “violence” is an issue for which they claim professional competenceand use to preserve the image of their jobs as dramatic and dangerous. Certainmental care nursing staff made this point evident in how they contrasted their workwith that of other health care personnel (Åkerström, 1993). Representatives of oc-cupations that may want to monopolize authority for handling violent encounterscan be expected to �nd the idea of staff in nursing homes experiencing violenceupsetting. At a conference where violence in working life was discussed, I presentedobservations from the nursing homes. After my presentation, a police of�cer got tohis feet and said, quite indignantly, “Surely that sort of thing doesn’t belong here,that’s not violence!” His comment not only represents a clear case of boundary-work, it also re�ects the powerful emotional charge tied to the concept ofviolence.

Expansion or, as Best (1990) calls it, “domain expansion” is a strategy wherebyactors expand a previously accepted social problem category so that the categorycan accommodate new constructions. We can observe this transformation in the suc-cessful struggle to rede�ne violence against women from “private family troubles” to


a social problem that engages public authorities, health services, and the judicialsystem (Loseke 1992). Aggression toward elderly family members has also been ex-panded from “private family troubles” to a public problem (Baumann 1989). Thisproblem is now constructed under the term “elder abuse” and has been given itsown journal, the Journal of Elder Abuse and Neglect.

The present study addresses a case of exclusion, that is, how a phenomenon thatmight be included in a category is instead kept out of it. The exclusion is, however,not the outcome or an illustration of a failed attempt to construct a social problem,since no effort has occurred in promoting a social problem of “elderly’s violence.”Nor can we de�ne the nurses’ rhetoric as counterrhetoric in the social problemsgame (Ibarra and Kitsuse 1993), since no claims have arisen against which to mounta counterattack. Exactly this lack of promotion efforts and counterrhetoric wassociologically interesting. Nurses unquestionably placed the elderly’s physical ag-gression outside the fence. Nurses had not deemed it necessary to formulate theirfrequent response “This is not violence” until I explicitly posed a question. Further-more, it was a statement nursing home staff held on to even though they claimedthat responsibility and intent were present at times and even though they describedinjuries—elements commonly associated with violence.

Gieryn’s approach leads us to examine what is at stake in this boundary-work.The bene�ts of constructing violence are obvious: workers can use exposure to vio-lence to argue for a higher salary by pointing to a heavy workload and dif�culty ofthe work, for example. The staff I interviewed at mental hospitals and in psychiatricwards made this case. They claimed they reported occupational injury in every in-stance entailing “the slightest bump or scratch” as work injuries due to violence(Åkerström 1997).

If constructing de�nitions of violence has bene�ts, what would be the drawbacksfor the nurses to describe their patients as “violent”? The meaning and identity oftheir occupation as well as their occupational skills are at stake. If the staff labeledthe elderly’s aggression as “violence” this de�nition would imply a description ofthemselves as “victims,” which would defy common cultural constructs, as woulddescribing the patients as “offenders.” Moreover, this de�nition would put theirskills in question. After all, the staff are professionals employed to take care of theelderly. To describe themselves as victims would suggest that they were not compe-tent. As Holstein and Miller (1997:43) noted: “[T]o ‘victimize’ someone instructsothers to understand the person as a rather passive, indeed helpless, recipient of in-jury or injustice. . . . In a sense, ‘victimizing’ a person ‘dis-ables’ that person to theextent that victim status appropriates one’s personal identity as a competent ef�ca-cious actor” (original emphasis).

Furthermore, to describe the elderly as “violent” implies that they are unman-ageable, and, hence, it follows that elderly patients should be excluded from nursinghomes. The very process of identifying someone or some acts as violent is inher-ently exclusionary, pushing that person outside the boundary of normalcy and con-tinued acceptance. Violent offenders should be locked up; violent spouses should


be abandoned or resocialized; violent schoolchildren should be removed from ordi-nary schools and placed in special education or correctional schools, and so on.

We can assume that nursing home staff avoid using “violence” because they wantto keep their elderly patients and continue to work with them. Therefore, they viewthem as manageable—in a way that protects the caring frame.5 Staff at nursinghomes work in a caring occupation in which manners and talk must re�ect their be-ing nonjudgmental, accommodating, and understanding. No one expects serviceproviders in prisons to like or respect prisoners, but one does expect staff in mentalhospitals to understand their patients, although control them. However, nursinghome staff must act and talk kindly about their charges. The former can thus calltheir clients “violent” and at times even legitimately “manage” them by respondingwith violence. Staff working with the aggressive elderly should, in contrast, managethem with no physical force, but if nurses resort to physical force, they must explaintheir actions in “care language.” A situation I observed may serve as an example. Twoauxiliary nurses were involved in changing diapers. One did the changing and theother held the patient’s hands in a �rm grip while the old lady kicked and shouted,“No, no, no. Stop it. Stop it.” Afterward one of the nurses told me: “Beata is a bit rest-less, she �ghts you a bit, so we have to be two when we help her.”

Gieryn (1999:34–35) noted that boundary-work is easiest to study in contexts ofcontroversies in which actors defend turf and de�ne explicit boundaries but “thattoo little is said about when and how some cartographies get stabilized as unques-tioned tacit assumptions or as uncontested old maps.” This article is an effort to con-duct such an analysis, with an implicit rather than explicit form of boundary-work.

Acknowledgments: I want to thank Bob Emerson for his extremely generous andextensive help. I also want to thank Johan Asplund, Kathy Charmaz, Jaber Gubrium,Jack Katz, Donileen Loseke, and Ann-Mari Sellerberg, and two anonymous reviewers.

NOTES

1. In the vast literature on family violence, one �nds many examples of conceptual distinctions be-tween “violence” and “force” (Gelles and Straus 1979) and between child abuse and punish-ment (Giovannoni and Becerra 1979). The victims themselves may get involved in such bound-ary-work, as for example women who are considering whether they were in fact “raped” in thecontext of a “date” (Wood and Rennie 1994). In other instances, the victim and the abuser mayengage in cooperative work on the cultural cartography of physical assault: Hyden (1995) hasshown how words such as “blows” (slag) and “battering” (misshandel) may be replaced by“row” (bråk) in Swedish marriage couples’ joint account of an incident.

2. The initial material consisted of sixty-� ve semistructured interviews that a research assistantand I conducted in various medical settings. Data from emergency wards are analyzed in Åker-ström 1996, and data comparing reporting of violence between mental health contexts andnursing homes are discussed in Åkerström 1997.

3. See Atkinson 1984 and Drew 1990 for illustrations of rhetorical contrastin g in othercontexts.

4. In Making Gray Gold, Diamond describes giving showers as one of the dif�cult and delicate


tasks in the nursing home he studied. He quotes one nursing assistant as explaining why she didnot like giving showers: “I can’t stand the screams” (1992:136).

5. Moreover, “excluding possibilities” are few, and some, like the legalistic sentencing alternative,would appear quite ludicrous. As one nurse put it: “We can’t very well call the police can we?”The resocialization alternative is hardly applicable since the elderly have very little future to besocialized into. The one “last resort” (Emerson 1981) is the psychiatric geriatric wards (referredto by my informants as “the scream wards”) where staff reluctantly send patients whom theycannot handle. (The reluctance seems due in part to such a decision being seen as an outcomeof the failure of staff at the current ward, in part to psychiatric geriatric wards having a reputa-tion for not being good care environments.)

REFERENCES

Åkerström, Malin. 1991. Vården möter våldet (Care-Givers’ Encounters with Violence). Stock-holm: Landstingsförbundet.

———. 1993. Våld och hot i sjukvårdsarbeten (Violence and Threats in Medical Settings). Lund:Network for Research in Criminology and Deviant Behaviour at Lund University.

———. 1996. “Waiting: A Source of Hostile Interaction in an Emergency Clinic.” QualitativeHealth Research 7(4):504–20.

———. 1997. “Att anmäla våld—en praktik inom lokala kulturer” (Reporting Violence: Practicesin Local Cultures). Nordisk Tidsskrift for kriminalvidenskab 84(2):124– 31.

———. 1998. “The Moral Crusade on Violence in Sweden: Moral Panic, or Material for Small-Talk Indignation.” Pp. 323–38 in European Criminology: A Handbook, edited by V. Rug-gerio, N. South, and I. Taylor. London: Routledge.

Allen, Davina. 2001. “Narrating Nursing Jurisdiction: Atrocity Stories and Boundary-Work. Sym-bolic Interaction 24(1):75–103.

Altheide, David L. 2000. “Identity and the De�nition of the Situation in a Mass-Mediated Con-text.” Symbolic Interaction 23(1):1–28.

Antaki, C. 1987. “Performed and Unperformable: A Guide to Accounts of Relationships.” Pp. 97–113 in Accounting for Relationships: Explanation, Representation and Knowledge, edited byR. Burnett, P. McGhee, and D. Clarke. New York: Methuen.

Arblaster, Anthony. 1994. “Violence.” Pp. 700–702 in Twentieth-Century Social Thought, edited byW. Outhwaite and T. Bottomore. Oxford: Blackwell.

Atkinson, Max. 1984. Our Masters’ Voices. London: Methuen.Barlow, Hugh. 1987. Introduction to Criminology. Boston: Little, Brown.Baumann, Eleen. 1989. “Research Rhetoric and the Social Construction of Elder Abuse.” Pp. 55–

74 in Images of Issues: Typifying Contemporary Social Problems, edited by J. Best. NewYork: Aldine de Gruyter.

Berbrier, Mitchell. 1998. “‘Being Deaf Has Very Little to Do with One’s Ears’: Boundary-Work inthe Deaf Culture Movement.” Pp. 79–100 in Perspectives on Social Problems, vol 10, editedby J. Holstein and G. Miller. London: JAI Press.

Best, Joel. 1990. Threatened Children: Rhetoric and Concern about Child-Victims. Chicago: Uni-versity of Chicago Press.

———. 1999. Random Violence. Berkeley: University of California Press.Best, Joel, ed. 1989. Images of Issues: Typifying Contemporary Social Problems. New York: Aldine

de Gruyter.Buttny, Richard. 1993. Social Accountability in Communication. London: Sage.Charmaz, Kathy. 2001. “Grounded Theory.” Pp. 335–52 in Contemporary Field Research . 2nd ed.

Prospect Heights, IL.: Waveland Press.Christie, Nils. 1986. “The Ideal Victim.” Pp. 17–30 in From Crime Policy to Victim Policy, edited by

E. Fattah. London: Macmillan.


Coates, Jennifer. 1993. Women, Men and Language: A Sociolinguistic Account of Gender Differ-ences in Language. London: Longman.

Coates, Linda, Janet B. Bavelas, and James Gibson. 1994. “Anomalous Language in Sexual AssaultTrial Judgments.” Discourse & Society 5:189–206.

Diamond, Timothy. 1992. Making Gray Gold: Narratives of Nursing Home Care. Chicago: Univer-sity of Chicago Press.

Dingwall, Robert and Topsy Murray. 1983. “Categorization in Accident Departments: ‘Good’Patients, ‘Bad’ Patients and ‘Children.’” Sociology of Health and Illness 5:127–148.

Drew, Paul. 1990. “Strategies in the Contest between Lawyer and Witness in Cross-Examination.”Pp. 39–64 in Language and the Judicial Process, edited by J. Levi and A. Walker. New York:Plenum.

Emerson, Robert M. 1966. Judging Delinquents: Context and Social Process in Juvenile Court. Chi-cago: Aldine.

———. 1981. “On Last Resorts.” American Journal of Sociology 87:1–22.———. 1994. “Constructing Serious Violence and Its Victims: Processing a Domestic Violence Re-

straining Order.” Pp. 3–28 in Perspectives on Social Problems, vol. 6, edited by J. A. Holsteinand G. Miller. Greenwich, CT: JAI Press.

Fattah, Ezzat. 1991. Understanding Criminal Victimization. Scarborough, NY: Prentice-Hall.Gar�nkel, Harold. 1984. Studies in Ethnomethodology. Cambridge: Polity Press.Gavioli, Laura. 1995. “Turn-Initial vs. Turn-Final Laughter.” Discourse Processes 19:369–84.Gelles, Richard J. 1977. “Demythologizing Child Abuse.” Pp. 385–94 in Family in Transition, 2d ed,

edited by A. Skolnick and J. Skolnick. Boston: Little, Brown.Gelles, Richard J. and Murray A. Straus. 1979. “Determinants of Violence in the Family: Toward a

Theoretical Integration.” Pp. 549–81 in Contemporary Theories about the Family, vol. 1, ed-ited by W. R. Burr, R. Hill, I. Nye, and I. L. Reiss. New York: Free Press.

Gieryn, Thomas F. 1983. “Boundary-Work and the Demarcation of Science from Non-Science:Strains and Interests in Professional Ideologies of Scientists.” American Sociological Review48:781–95.

———. 1995. “Boundaries of Science.” Pp. 393–443 in Handbook of Science and TechnologyStudies, edited by S. Jasanoff. London: Sage.

———. 1999. Cultural Boundaries of Science: Credibility on the Line. Chicago: University of Chi-cago Press.

Gil, David G. 1970. Violence against Children: Physical Child Abuse in the United States. Cam-bridge, MA: Harvard University Press.

Giovannoni, Jeanne M. and Rosina M. Becerra. 1979. De�ning Child Abuse. New York: FreePress.

Goffman, Erving. 1961. Asylums: Essays on the Social Situation of Mental Patients and Other In-mates. New York: Doubleday Anchor.

———. 1972. Relations in Public. New York: Harper Colophon.Goode, Erich and Nachmen Ben-Yehuda. 1994. Moral Panics: The Social Construction of Devi-

ance. Cambridge, MA: Blackwell.Grif�ths, Lesley. 1998. “Humour as Resistance to Professional Dominance in Community Mental

Health Teams.” Sociology of Health and Illness 20(6):874– 95.Gubrium, Jaber F. [1975] 1997. Living and Dying at Murray Manor. New York: St. Martin’s.———. 1986. Oldtimers and Alzheimer’s: The Descriptive Organization of Senility. London: JAI

Press.Gubrium, Jaber and James Holstein. 1990. What Is Family? Mountain View, CA: May�eld.Gus�eld, Joseph. 1989. “Constructing the Ownership of Social Problems: Fun and Pro�t in the

Welfare State.” Social Problems 36(5):431–41.Hepworth, Mike and Bryan Turner. 1982. Confessions: Studies in Deviance and Religion. London:

Routledge and Kegan.Holstein, James. 1993. Court-ordered Insanity. New York: Aldine de Gruyter.Holstein, James and Jaber Gubrium. 1995. The Active Interview. London: Sage.


Holstein, James and Gale Miller. 1997. “Rethinking Victimization: An Interactional Approach toVictimology.” Pp. 25–47 in Social Problems in Everyday Life edited by G. Miller and J. Hol-stein. Greenwich, CT: JAI Press.

Hunt, Jennifer. 1985. “Police Accounts of Normal Force.” Urban Life 13(4):315– 41.Hyden, Margareta. 1995. Kvinnomisshandel inom äktenskapet (Wife Beating in Married Life).

Stockholm: Liber.Ibarra, Peter and John Kitsuse. 1993. “Vernacular Constituents of Moral Discourse: A Interaction-

ist Proposal for the Study of Social Problems.” Pp. 21–54 in Constructionist Controversies,edited by G. Miller and J. Holstein. New York: Aldine de Gruyter.

Jarvinen, Margaretha. 2001. “Accounting for Trouble: Identity Negotiations in Qualitative Inter-views with Alcoholics.” Symbolic Interaction 24(3):263– 84.

Katz, Jack. 1999. How Emotions Work. Chicago: University of Chicago Press.Loseke, Donileen. 1992. The Battered Woman and Shelters. New York: State University of New

York Press.———. 1999. Thinking about Social Problems. New York: Aldine.Magnusson, Finnur. 1996. Janus-ansiktet (The Face of Janus). Stockholm: Carlssons.Mead, George Herbert. 1972. On Social Psychology: Selected Papers. Edited and with an Introduc-

tion by Anselm Strauss. Chicago: University of Chicago Press.Meyer, John C. 1997. “Humor in Member Narratives: Uniting and Dividing at Work.” Western

Journal of Communication 61:188–200.Miller, Gale and James Holstein, eds. 1993. Constructionist Controversies. New York: Aldine.Offer för vålds-och egendomsbrott 1978–1989 (Victims of Violence and Thefts). 1991. Levnads-

förhållanden. Rapport 66. Sveriges of�ciella statistik. Stockholm: Statistiska centralbyrån.Pogrebin, Mark and Eric Poole. 1988. “A Study of the Strategic Uses of Humor among Police.”

Journal of Contemporary Ethnography 17:183–210.Rossol, Josh. 2001. “The Medicalization of Deviance as Interactive Achievement: The Construc-

tion of Compulsive Gambling.” Symbolic Interaction 24(3):315– 41. Sarat, Austin. 1993. “Speaking of Death: Narratives of Violence in Capital Trials.” Law and Society

Review 27:19–58.Scott, Marvin B. and Stanford M. Lyman. 1968. “Accounts.” American Sociological Review 33:46–

52.Seckman, Mark and Carl Couch. 1989. “Jocularity, Sarcasm, and Relationships.” Journal of Con-

temporary Ethnography 18:327–344.Snow, David and Leon Anderson. 1993. Down on Their Luck: A Study of Homeless Street People.

Berkeley: University of California Press.Spector, Malcolm and John I. Kitsuse. 1987. Constructing Social Problems. New York: Aldine.Stannard, Charles. 1973. “Old Folks and Dirty Work: The Social Conditions for Patient Abuse in a

Nursing Home.” Social Problems 20:329–42.Statistiska Centralbyrån. 1991. Levnadschivåundersokningen (Standard of Living), no 66. Stockholm.Tannen, Deborah. 1981. “New York Jewish Conversational Style.” International Journal of the So-

ciology of Language 30:133–49.Thorne, Barrie. 1999. Gender Play: Girls and Boys in School. New Brunswick, NJ: Rutgers Uni-

versity Press.Wikström, Per-Olof. 1992. Våld—en kunskapsöversikt (Violence: A Scienti�c Overview). Stock-

holm: Brottsförebyggande rådet (Council of Crime Prevention).Wood, Linda and Heather Rennie. 1994. “Formulating Rape: The Discursive Construction of Vic-

tims and Villains.” Discourse & Society 5:125–48.Yoels, William and Jeffrey Michael Clair. 1995. “Laughter in the Clinic: Humor as Social Organi-

zation.” Symbolic Interaction 18:39–58.

Slaps, Punches, Pinches—But not Violence: Boundary-Work in Nursing Homes for the Elderly

Documents