LABELING THEORY CAPSTONE PAPER

victims of crime and labeling theory: a parallel process?

J. Scott Kenney Department of Sociology and Social Anthropology, Dalhousie University, Halifax, Nova Scotia, Canada

Labeling theory tends to focus largely on the offender. Yet, implicit in interactionist theories of deviance is a concern for the social situation as a whole. This logically includes the victim of crime. This article explores the potential of extending the interactionist perspective on deviance to the experiences of victims of crime. Speci®cally, I outline a parallel labeling process for victims in which differential social reactions to this status, ¯owing from varying attributions of sympathy worthiness, have an impact on the behaviors, adjustment, and identities of the individuals concerned. This process is further distinguished from the related labeling of emotional deviance. I then present the results of a qualitative study of individuals who have suffered the murder of a loved one. Through an empirical examination of the varying social reactions to these individuals by extended family, friends, acquaintances, and the community, as well as victims’ varying responses thereto, I indicate how familiar terms such as accommodation, labeling, primary, secondary, and tertiary deviance each have their conceptual counterpart in the experiences of victims.

Address correspondence to J. Scott Kenney, Department of Sociology and Social Anthropology, Dalhousie University, Halifax, Nova Scotia, Canada B3H 3J5. E-mail: [email protected]

Deviant Behavior: An Interdisciplinary Journal, 23:235±265, 2002 Copyright # 2002 Taylor & Francis 0163-9625/02 $12.00 + .00

235

Crime creates problems for people. In the past much research in this vein, particularly from the labeling perspective, has largely focused on offenders. Yet, victims of crime often experience the labeling process as well. This paper delineates the processes and consequences underlying such labeling by empirically examining the experiences of individuals who have suffered the murder of a loved one. In this way, the labeling perspective, as a theory rooted in the interactionist tradition, is broadened in a direction implicit in its philoso- phical antecedents.

PRIOR LITERATURE

The labeling perspective is grounded in the conception that deviance is de®ned by social reactions, and that the frequency and character of the deviation, as well as the role of the deviant, are largely shaped by interactional response (Lemert 1951; Becker 1963). As such, it draws attention away from the idea that there is anything objective in the actions of individuals that is deviant, focusing instead on changes in social de®nitional processes and their consequences. Yet this perspective need not be restricted to the study of deviance.1

Taylor, Wood, and Lichtman (1983) note the victims often experience negative social consequences, asserting that ``interpersonal reactions to victims are at best ambivalent, and at worst hostile and rejecting’’ (1983:23). Indeed, they make a distinction, parallelling Lemert (1951), between primary victimization, that is, the initial victimizing circumstances, and secondary victimization, the negative social reactions of hostility, derogation, and rejection that can follow Taylor et al. (1983:23). This often results in a ``master status’’ (Hughes, 1945) where attributes related to victimization are highlighted and others downplayed. Ultimately:

Even the best social responses to victimization may be aver- sive to the victim. The need to accept aid from others and the

1 It has already been extended in the study of illness by Conrad and Schneider (1980:17± 20 et. seq). As such, the parallels outlined by examining victims merely refer to one possible analogy among others. Indeed, while this paper’s specific focus is on extending labeling theory, one can easily look to other theoretical traditions, such as Parsons’ (1951) work on the sick role, for additional insights.

236 J. S. Kenney

accompanying emotional reactions such as pity may indicate the condescension of the other and underscore the loss of power or status on the part of the victim. Help-seekers lose face and self-esteem, and they risk evaluations of incompe- tence by the helper. Secondary victimization can occur, then, even when the non-victim has the best intentions (Taylor et al. 1983:24±5).

Many who are treated in such ways ®nd their victim status almost impossible to reverse. Since others respond to victims on the basis of the label, victims may come to internalize these responses and perceptions, and begin to think of themselves in the same way (Taylor et al. 1983:24). Indeed, such individuals may feel social pressure to withdraw further into the world of similar victims, making a change in status even more dif®cult. ``Whether or not the victim has lost self-esteem due to the primary victimizing circumstances, then, the secondary victimization of social labeling, rejec- tion, and isolation can itself lower self-esteem’’ (Taylor et al. 1983:24±5).2

Taylor, Wood, and Lichtman’s (1983) work is exciting and suggests more detailed theoretical and empirical examination of a parallel labeling process for victims. However, while theoretically pointing to such an interesting extension of labeling theory, these authors fail to present any data of their own, merely reviewing existing theoretical and empirical work suggestive of their model.

A related literature exists on social responses to victims of undesirable life events (Lepore et al. 1996; Holman and Silver 1996; Wortman and Lehman 1983; Silver and Wortman 1980; Wortman and Dunkel-Schetter 1979). This examines situations as diverse as accidents, incest, cancer, and non- homicide bereavement and reviews the impact of social responses on affected individuals. Theoretically, for example, Silver and Wortman (1988) argue that an examination of (1) perceived social support, and (2) the relative opportunity for free expression of feelings may further understanding of

2 Such inhibiting responses run counter to the psychoanalysis literature that bases treatment on support for the ventilation of feelings (Freud 1957:146±58). Indeed, recent psychiatric literature discusses homicide bereavement with reference to the possibilities for sharing offered by peer support groups (Klass 1988:125±39; 151±77).

Victims of Crime and Labeling Theory 237

people’s responses to stressful outcomes and enhance our ability to predict the nature, sequence, duration, and intensity of their reactions. They point to studies demonstrating a correlational relationship between perceived social support and effective adjustment, but suggest that more emphasis should be placed on the deleterious consequences of beha- viors that are intended to be supportive (1980:310, 313). They also note that despite a desire to talk, ``people who are in the throes of serious life crises often report that they are encouraged to be `strong’, are prevented from free expression of their feelings, and are thus denied the ventilation they frequently need and want’’ (1980:314). This is despite evi- dence that many ®nd the opportunity to express their feelings helpful, and that ``lack of communication may intensify the strain of victimization’’ (1980:315).

Particularly interesting is the work of Wortman and Lehman (1983), who assert that other’s well-intentioned efforts to provide support may be regarded as unhelpful by the reci- pient, result in negative consequences, or both. They focus on three basic reasons why people may respond to victims of life crises in such ways: (1) for many reasons, people appear to hold negative feelings about others who are suffering or distressed (particularly because such encounters make potential supporters feel threatened and vulnerable); (2) Most people have little experience dealing with others in the throes of a life crisis, and may experience a great deal of uncertainty about what to say or do; and (3) Others hold a number of misconceptions about how people should react. Beliefs about how much distress should be experienced and displayed, and how long the effects of the crisis should last, will likely in¯uence the kinds of support attempts made. To the extent that these beliefs are in error, others’ support attempts may be inappropriate or counter-productive (Wort- man and Lehman 1983:464). Signi®cantly, the authors assert that all of these factors ``are much more likely to predominate when the consequences of the victimization are serious’’ (1983:467).

Wortman and Lehman (1983) note that because of these feelings and beliefs, individuals engage in three types of behaviors intended to be supportive, but that are often det- rimental: (1) they discourage open expression or discussion of feelings about victimization; (2) they encourage recovery or

238 J. S. Kenney

movement to the next life stage; and (3) they fall back on automatic or scripted support attempts which may seem to dismiss or trivialize the victim’s problem, such as giving advice, providing a reason for what has happened, or telling the victim that they know how he or she feels (Wortman and Lehman 1983:468).

Considering the impact of such behaviors on the victim, Wortman and Lehman (1983) state that there are three rea- sons why these particular support tactics may be ineffective and harmful: (1) they may make victims feel isolated and alone because they curtail opportunities to share concerns with others; (2) in many cases, they may seem to dismiss the victims’ feelings as insigni®cant or unimportant; and (3) they imply that the victim should not feel as bad as he or she does. Indeed, there is evidence that victims worry about whether their feelings of distress are normal (pp. 473±4). The authors deduce that those who are closest to the victimÐand therefore who have the greatest stake in the victim’s recov- eryÐmay have the least tolerance for the victim’s distress. Indeed, they may be especially likely to engage in the inap- propriate support behaviors noted above.

A number of empirical studies have supported these arguments (Lepore et al. 1996; Holman and Silver 1996; Wortman and Lehman 1983; Wortman and Dunkel- Schetter 1979). All of this is consistent with the labeling of victims.

Finally, a more speci®c literature also exists on social reaction to homicide bereavement in which further evidence of labeling is strongly suggested. Following the violent death of their loved one, individuals often turn to family and friends for help and support. Yet, in many cases they do not receive what they are looking for, and these interactions may do more harm than good. Published accounts of homicide survivors suggest that while family and friends are usually supportive at ®rst, this often gives way to the unspoken imperative that ``they should get on with their lives’’ (Young 1991:37). In some cases, grieving nuclear families have been avoided by family and friends who do not how to deal with their con- tinuing grief (Ontario Government 1984). Indeed ``families of homicide victims speak of losing up to 90 percent of their friends because no one wants to talk about the crime and no one wants to hear about the victim’’ (Young 1991:37).

Victims of Crime and Labeling Theory 239

Therapists corroborate such accounts. Klass (1988:151±3) has noted that while problems with the social support system form the basis for one element of psychotherapy with all bereaved parents, he has observed in his experiences with homicide support groups that the loss of support systems surrounding parental grief ``seems to be exaggerated in the case of murder’’ (1988:130).

Both the discussion of social reactions to victims of life crises and the literature on those bereaved by homicide have implications for labeling theory. First, following Taylor, Wood, and Lichtman (1983), there may be many cases where individuals are not reacted to as deviant, but as victims. In such cases, this literature suggests that the quality of indivi- duals’ social relationships following traumatic events has important consequences for their adjustment to traumatic experiences. In some cases, social constraints ¯owing from the label of victim, such as stigmatization, uncertainty, and misconceptions about appropriate response (Wortman and Lehman 1983), can inhibit people from discussing their traumatic experiences and increase the positive association between intrusive thoughts and depressive symptoms (Lepore et al. 1996). This potentiality intensi®es characteristics typical of victimization. In other cases, the victim label is apparently associated with unconstrained, accommodative social rela- tionships that facilitate the processing of traumatic events and emotional recovery. It is necessary to clarify the circum- stances under which each response occurs.

However, it is important to further consider the issue of stigmatization, as this constitutes the porous boundary between the label of victim and that of deviant. Wortman and Lehman (1983:464±5), on the one hand, review the literature suggesting that many victims are stigmatized as such due not only to the unsettling feelings of vulnerability and help- lessness they evoke in others, but the common belief that we live in a world where people get what they deserve and deserve what they get. In particular, the work of Lerner (1980) suggests that if an individual can believe that others do not suffer unless something is wrong with them, or there are weaknesses in their behavior, he or she will feel protected from undeserved suffering in the future. On the other hand, Thoits (1990) has introduced the term ``emotional deviance’’ to cover the labeling of behaviors that violate some standard

240 J. S. Kenney

of appropriate emotional response in a given situation.3

Intermediate between these positions are authors such as Coates and Winston (1983) who consider the long term depression associated with victimization to itself be deviant in cultural terms (1983:171±77). It will thus be important to see if stigmatization as victims, as deviants, and as emotional deviants may be separated empirically, and, if so, to examine the circumstances in which each occurs.

The above issues are helpfully illuminated by the work of Clark (1987). In her work on sympathy, Clark found evidence that crime victims may be either considered ``sympathy worthy’’ or ``blamed’’ for their plight (1987:298). She pro- poses a theoretical model for distinguishing individuals who receive sympathy from those who are blamed, which revolves around the concept of ``sympathy margin’’ (i.e., the amount of leeway a given individual has for which he or she can be granted sympathy and not blamed). As such it is similar to the concept of accommodation to deviance (Lynch 1983; Rubington and Weinberg 1987; Yarrow et al. 1955; Jackson 1954). Sympathy may either be ascribed to strangers or those with whom one is acquainted. In the ®rst instance, status factors such as age, social class, sex, and type of problem are signi®cant (Rubington and Weinberg 1987:291, 298, 300±1).4 As for those in one’s social network, those in close relationships have an obligation to create wider sympathy margins for each other than do mere acquaintances. Regardless of the context, however, Clark (1987) notes that anyone who has been ascribed margin has the right to sen- timent, empathy, and display of sympathy. Throughout, sympathy accounts or margins do not remain constant, but are continually negotiated. In this process, these may be ``increased, decreased, replenished, or used up entirely’’ (Clark 1987:302). One may cash in the credits built up throughout his or her ``sympathy biography’’ in a dif®cult

3 This relates to Hochschild’s (1983) discussion of ``feeling rules.’’ Indeed, Thoits (1990) draws heavily on Hochschild’s work in elucidating this concept. There is also a link with the ``doubly deviant’’ character of HIV=AIDS victims, whose illnesses place them in an ideologically different space from the socially sanctioned deviance of the Parsonian model (Llamas 1994; Kowalewski 1988).

4 This may relate to studies of charitable giving, where presumed culpability is identified as the major factor affecting sympathetic response to disasters (Russell and Mentzel 1990).

Victims of Crime and Labeling Theory 241

situation, but should not drain the account completely. In such cases, the potential sympathizee may not only lose sympathy, but have to look elsewhere.

Indeed, Clark (1987) outlines four rules of ``sympathy eti- quette,’’ which include: (1) not making false claims to sym- pathy; (2) not claiming too much sympathy; (3) claiming some sympathy in appropriate circumstances; and (4) reci- procating to others for gifts of sympathy (pp. 303±13). She also identi®es what she calls ``deviant sympathizers,’’ who either ``underinvest’’ by not recognizing others’ rights to sympathy, or ``overinvest’’ by giving sympathy to others who are not worthy, whose plights are not worthy, or who do not adhere to the rules of sympathy etiquette (pp. 313±6).

In sum, the label of victim has three possible trajectories:

1. Victims may be reacted to as such, ascribed sympathy, and be offered unconstrained, accommodative support.

2. Victims may be reacted to as such, ascribed sympathy, but others respond in ways that indicate uncertainty or mis- conceptions about how to interact.

3. Victims may be reacted to as such, initially ascribed sympathy, but others may eventually stigmatize them as ``helpless victims’’ who are unable to cope.

As for labeling as deviants, victims experience this in two ways:

1. Victims may be reacted to as emotional deviants as the result of their ``inappropriate’’ emotions in a particular social context. This is related to stigmatization as a help- less victim, and may follow from it, but contains the ad- ditional element of inappropriate behavior in public.

2. Victims may be stigmatized squarely as deviants who are blamed for their plight in the ®rst place.

For all of these, Clark’s work may be hypothetically incorporated into the labeling perspective as a set of sensi- tizing concepts to help distinguish between different labeling responses to victims, as well as between circumstances where individuals are labeled as victims or as deviant.5

5 There may be a parallel with M. Douglas’ (1970, 1978) Grid=Group theory, in which a two-dimensional schema for classifying social relations is related to varying beliefs and accounts. While not employing a formal grid=group analysis in this paper, the data herein clearly illustrate that variations in respondents’ social relations impacted their reputed labels.

242 J. S. Kenney

Finally, in an attempt to further elaborate Taylor, Wood, and Lichtman’s (1983) theoretical ideas about victimization and labeling, the differential labeling inherent in social reaction to victims necessitates consideration of their ``careers’’ that follow (Becker 1963). Of particular interest is how ``primary victimization’’ and ``primary deviance’’ are accommodated by others (Lynch 1983; Rubington and Weinberg 1987; Yarrow et al. 1955; Jackson 1954), and, if these break down, whether individuals begin to employ their behaviors, or a role based upon it, as a means of defense or adjustment to the overt or covert problems created by this social reaction (Lemert 1951). Moreover, such ``secondary’’ reactions (e.g., adopting the ``helpless victim’’ role) must be contrasted with the potential for survivors to engage in ``tertiary’’ responses where they may collectively ``confront, assess, and reject the negative identity imbedded in victimi- zation, and transform that identity into a positive and viable self-image’’ (Kitsuse 1980:9).

Thus, this article will focus on the idea that the type of support offered (or not offered) to victims may sometimes contribute to grief and powerlessness rather than alleviate it. Following Lo¯and (1985:173), it is argued that grief is ``pro- foundly socially shaped.’’ Indeed, there is reason to believe that, in the interactional construction of survivors’ experi- ences, there is a dynamic relationship between others’ con- ceptions of appropriate behavior, on the one hand, and the type of support (or lack of it) given to survivors on the other (Emerson and Messinger 1977). By a careful examination of the differential labeling that occurs in informal settings,6 and the ways survivors individually and collectively respond, labeling theory can be enriched and extended along a whole new dimension.

METHODOLOGY

The methods underlying this study were qualitative in nature. Recognizing the dif®culty of accessing highly personal

6 This article specifically focuses on the labeling dynamics faced by victims in informal settings, such as with friends, family, acquaintances and the community. Their encounters with formal agents (i.e., medical=psychiatric professionals, the criminal justice system) are exceedingly complex, and each warrant separate papers in their own right.

Victims of Crime and Labeling Theory 243

information from a traumatized population, I used a wide variety of direct and indirect approaches, including extensive volunteer involvement with a prominent victims’ organiza- tion, advertising for interview and survey respondents with a variety of others, reviewing published accounts, and seeking information through the legal system.7 Taken together, while some of these strategies were more successful than others, they ultimately resulted in a large volume of rich, qualitative data. The overwhelming majority of data fell into three gen- eral categories: (1) intensive interviews with 32 individuals; (2) mail-back surveys from 22 respondents; and (3) 108 homicide ®les obtained under an agreement with a Provincial Criminal Injuries Compensation Board (C.I.C.B.), including information on 145 individuals. Aside from the C.I.C.B. data, in which there was a con®dentiality agreement negotiated for transcription and content analysis of case ®les, interview and survey respondents were informed in advance about the subject matter of this study, con®dentiality, and given opportunities to ask questions before proceeding.

Possible threats to validity were dealt with by having both academic colleagues and key members of the victims’ orga- nization in which I was workingÐincluding survivorsÐread and comment on the research instruments before they were used. The valuable input received, and changes made on the basis of these suggestions, buttressed my con®dence that the instruments did indeed get at the issues they were intended to elicit. In addition to face validity, the fact that the sought after issues emerged through the use of these instruments was suggestive of construct validity as well.

7These strategies included: (i) volunteering, conducting fieldwork, and doing interviews for a year with a prominent Canadian victims’ organization; (ii) networking=advertizing with six other Canadian victims’ rights organizations for interview=survey respondents; (iii) advertizing with a national U.S. support=advocacy group for the homicide bereaved for survey respondents; (iv) attending conferences, rallies, and protests on victims’ issues to observe=seek respondents; (v) contacting a prominent bereavement support organization for potential subjects; (vi) contacting the provincial chapter of an organization devoted to combatting impaired driving for comparative data; (vii) subscribing to the newsletters of the above organizations; (viii) reviewing biographical and journalisitic accounts of survivors’ experiences; (ix) unsuccessfully attempting, through formal access to information requests, to gain access to ``victim impact statements’’ submitted by survivors at the sentencing of offenders; and (x) writing all Criminal Injuries Compensation Tribunals in Canada for information on applications by survivors.

244 J. S. Kenney

With regard to reliability, rather than rely on only one question per issue, research instruments were constructed such that they generally utilized several different questions to measure or get at the same thing. As such, problems of consistency in responses were reduced or negated.8

While the sample was neither random nor clearly repre- sentative of any speci®ed population,9 all efforts were made to keep it as diverse as possible within the researcher’s ability to gain access to data.10 Hence, the ultimate conclusions are less likely to carry an obvious bias toward any speci®c demographic group.11 Still, broader generalizations must be quali®ed with this in mind.12

8 More detailed information on research instruments may be obtained from the author. 9 Facing the problem of drawing random sample from an unknown population (Berg

1995:178±9), theoretical sampling and Glaser and Strauss’ (1967) ``constant comparison method’’ were utilized. A purposive sample was developed (Berg 1995:179), and, subject to my ability to gain access to willing respondents, efforts were made to cover an ever wider variety of survivors as evidenced by both the literature and completed fieldwork. This involved such demographic variables as race, class, gender, age, marital status, and relationship to the deceased. Other dimensions included circumstances surrounding the murder, plus variations in survivors’ interactions with family, friends, the community, self- help and ``victims’’ organizations, mental health professionals, and legal institutions. This strategy was further buttressed once the C.I.C.B. data became available. I was then able to compare the purposive sample emerging through thoretical sampling with the entire population of homicide applications decided by a provincial C.I.C.B. between 1988 and 1993.

10 Attempts were made to clarify and validate observations, to resolve anomalies and contradictions, and to fill in parts of the social process that had not yet been observed (Schwartz and Jacobs 1979:29). Sampling was thus ``an ongoing procedure’’ (Williamson et al. 1977), and utilizing more than one source enabled the critical evaluation of information provided by ``checking out’’ each form of data gainst the other (Douglas 1976). Indeed, at times, the data from several sources even provided useful contrasting information on the same individuals. By triangulating data from various sources, a much more well- rounded interpretive understanding of survivors’ experiences was obtained than would otherwise be the case. Sampling continued in this fashion until reaching the point of ``theoretical saturation’’ (i.e., the point where incoming data revealed nothing new).

11 As the C.I.C.B. data helped round out the largely white, middle-class composition of the interview and survey data, the latter helped balance the potential biases inherent in the demand characteristics of the C.I.C.B. process.

12 Response rates to the directly solicited interviews and surveys were low. Through contacts with victims’ organizations it was clear that there were many more survivors than were willing to be interviewed. Of the 60 mail-back surveys distributed in Canada, and 35 to the U.S., only 8 and 14 were received back, respectivelyÐresponse rates of 13.3% and 40%. This reflects the extremely sensitive nature of the subject, and subjects’ understandable distrust. Yet, comparing the unsolicited C.I.C.B. and field data helped round out these sources.

Victims of Crime and Labeling Theory 245

Once collected, data were transcribed verbatim and entered into the qualitative analysis software Q.S.R. NUD IST. Using a procedure of open coding, a systematic ®ling system was developed. Coding and analysis initially pro- ceeded according to various common and special classes (e.g., survivor’s relation to the deceased for the former; involvement with the C.I.C.B. for the latter). As the analysis progressed, these materials were coded and recoded into theoretical categories that emerged from the data as well (Berg 1995:185). Throughout, emerging theoretical cate- gories and the various sources of data were continually ``cross-checked’’ and subjected to ``negative case testing’’ (Berg 1995). If inconsistencies were located, emergent ideas were either discarded or reformulated until practical certainty was achieved. Having outlined methodology and the sample, it is time to discuss the results of this study.

PRESENTATION OF THE DATA

The varying social responses to survivors following a homi- cide interactionally shaped the course of their experiences and strongly suggested differential labeling dynamics at work. I will brie¯y discuss these results in three parts: (1) extended family and friends; (2) acquaintances, strangers, and the community; and (3) subjects’ responses.

First, a minority of respondents experienced widespread, ongoing support from the majority of their extended family and friends.

I’ve always had people really look after me. I’ve never been left to do this aloneÐand that made a huge difference. My family and friends were there morning, noon, and night for at least two years, which made me feel goodÐto know that they were there for me (Survey #19: Female, age 45).

These individuals reported a number of sympathetic, and reportedly helpful responses such as others visiting and staying, providing ongoing emotional support, handling responsibilities for them, and a wide variety of helpful com- munication involving the ability of others to pick up subtle cues regarding when, and how, to offer support.

246 J. S. Kenney

Running through such responses is the implicit theme that these individuals were considered ``sympathy worthy’’ by others (Clark 1987). On the one hand, as extended family and friends are relatively close to survivors, they may have built up, or feel it necessary to extend a wider ``sympathy margin’’ than others, at least initially. However, as these responses were only mentioned by a minority in the sample, it may be that this also has something to do with survivors adhering to the rules of ``sympathy etiquette’’ such as making ``legitimate’’ claims to sympathy in ``appropriate’’ circum- stances.’’ These matters are closely interrelated with survivors ``victim’’ status, and these sympathetic behaviors cannot be understood outside of this context. Indeed, in Clark’s (1987) terms, such helpful responses, often extending for consider- able periods of time, suggest that suffering the murder of a loved one prima facie constitute, for these sympathizers, a legitimate claim for sympathy, and that the victim label, in some instances, even broadens their de®nitions of what is too long for victims to accept sympathy and their obligation to reciprocate.

It also may be argued that much of the above ®ts clearly into the idea of accommodation to deviance (Lynch 1983; Rubington and Weinberg 1987; Yarrow et al. 1955; Jackson 1954). Grief, in such unusual and violent circum- stances, is frequently extreme and persistent (Rynearson and McCreery 1993; Knapp 1986; Klass 1988), and the data in this study certainly bore this out. Family and friends who respond to such behavior as above are attempting to make adjustments to accomodate survivors’ powerful emotions without making an explicit labeling of deviance. If anything, it may be the situation that is seen as unusual, not survivors’ reactions to it, which are considered normal for victims. As such, the label of victim may not only play a role in accommodation to deviance, it may indicate the entry point of a parallel labeling process revolving around the term vic- tim. Accommodation in one sense is thus interchanged by labeling in the other.

On the other hand, the stigmatization of victims and the labeling of emotional deviance generally proceed when accomodation quickly breaks down, or doesn’t occur in the ®rst place. Together, these were the most common patterns, and had a far different impact on survivors’ experiences.

Victims of Crime and Labeling Theory 247

Generally, survivors who reported a lack of support from the majority of their extended family and friends over time felt that this made their experience worse. One survivor noted:

The only thing I remember is that I was aloneÐall the time. All the time. There was nobody came around. My family hasn’t been supportive. I haven’t seen any of my family. None of them have come, and that hurt . . . It’s been two years, and I think you are the third person that’s been in this house (Interview #10: Female, age 60) (Emphasis added).

Similar comments were made by survivors about their friends:

We were ostracized in a sense. In our own minds we felt ostracized anyway by everybody we knew (Interview #14: Male, age 54) (Emphasis added).

This was a time for me to ®nd out who my real friends were, and it sometimes hurt a lot to see that you friends are not always the one that you had considered. It is a shock (Survey #3: Female, age 58) (Emphasis added).

These individuals reported a number of unhelpful respon- ses including initial lack of support, rapidly disappearing support (e.g., after the funeral), inappropriate attention and harassment, widespread avoidance by others, numerous problems with communication such as others using ``stock’’ responses and attempting to identify with their feelings, and, in some cases, overt con¯ict with family and friends. Taken together, these corroborate and expand, in a new empirical context, many of the problems noted by Wortman and Leh- man (1983).

Most interstingly, some considered that these were the result of their being labeled as victims, others as deviants. Survivors’ rationales for these responses were instructive in separating these:

Some labeled as victims asserted how many of their extended family and friends were afraid to do or say anything that might upset them further and avoided contact as a result. This involved ascriptions of sympathy worthiness, labeling as victims, but an inability to provide overt expressions of sympathy out of uncertainty as to appropriate response.

248 J. S. Kenney

I know that my family just damn well didn’t know how to cope and how to deal with me [sounds angry]. They were scared. In fact we talked about it. We went home and it was discussedÐand they said that they had no idea. They really didn’t know how to deal with me, and were afraid to even phone me [sounds tearful] (Interview #15: Female, age 49).

Like my friends are afraid to keep bringing it up, and they shy away because they’re afraid you’re going to bring it up, `cause they’re afraid they’re going to say something that’s going to hurt you. `So they shy away. Maybe they don’t know that you want it. That it’s all right to talk about it. So they’re afraid to. I think that’s what happens with a lot of them, they’re afraid to start . . . Cause, I had a few people that I’ve bumped into, and I’ve said: `Oh, you haven’t been around.’ `Well [respondent’s name], it’s not that we didn’t want to. It’s just that we just don’t know what to say, and we didn’t want to say the wrong thing and hurt your feelings’ (Interview 21: Female, age 45).

This was further exacerbated in some cases where family and friends were considered to be so upset about the murder themselves that seeing the survivors evoked unpleasant sympathetic outbursts among all concerned. This response exhibited an excess of sympathy for survivors labeled as ``victims’’ whereby others became so upset that they simply couldn’t handle seeing the survivors:

My family, for the most part, kept their distance. And, occa- sionally when we’d come by, they would just sob and cry, and they’d be so upset. The few that did come by, they’d just break down. Horrible. I think they felt they were just upsetting us more and would be all apologetic. Then they wouldn’t come by (Interview #31: Female, age 46) (Emphasis added).

Our friends? It varied. I think that the reason why some of them kind of were standof®sh, was because they couldn’t handle it themselves. I ®gure that because they knew [the deceased] personally, they just couldn’t handle it. It was too close to reality for them, so they kind of backed off. I was a little bit taken aback by that, and I was very hurt. It’s been dif®cult because you really do need your friends at a time like that (Interview #1: Female, age 47).

Victims of Crime and Labeling Theory 249

These responses were particularly notable among individuals previously close to respondents.

There also were survivors, initially labeled as victims, who noted how the initially sympathetic responses of others eventually gave way to privately urging them to ``get on with your lives.’’ This signi®es a breakdown in others’ ascriptions of sympathy worthiness and the beginning of stigmatization as ``helpless victims.’’ These are individuals who have broken one of the cardinal rules of sympathy etiquette: not to claim too much sympathy for too long.

We had some very close friends who came out to visit us, but after a while their recommendation to us was, `You’ve got to put this in the past, like, bury this right now’ [sarcastically]. The closest friends we had! He’d say to me `[survivor’s name], you’ve got to put this behind, get back to work, and you’ve got to get on with your life.’ You know, `You’ve got to maintain your business, you’ve got to . . .’ and I’m not thinking any of this! (Interview #14: Male, age 54).

Some individuals came to feel that they were stigmatized as emotional deviants by others due to inappropriate behavior in public settings. For example, some talked about how their inability to talk about subjects other than the murder, or to remain calm, drove others away and resulted in uncomfor- table encounters at social events. Others talked of their own upset driving family and friends away. This is more suggestive of labeling as emotional deviants:

They started changing the subject a lot, so we would just pick up on it that `OK, we won’t talk about this any more.’ We caught the vibes, right? Bad vibes. Some pulled back from us, and this continues right up to the present. Nobody really wants to talk (Interviews #8: Female, age 45).

I think if I’m showing signs of emotion, I mean showing tears or emotional stress to others, they don’t feel very comfor- table, because they don’t know how to handle it. And so, if you are at a stage where you can’t speak about your son or daughter clearly, openly, and without showing an awful lot of undue stress, it drives people away (Interview #5: Male, age 50).

250 J. S. Kenney

Finally, some individuals were simply stigmatized as deviants. In these cases, survivors attributed others’ reactions to the perceived stigma or discrediting that went along with having a murder in the family, a stigma that suggested shared deviance:

The rest of my relatives were worthless and acted like a stigma was attached to us, like something about us caused us to have a murder victim (Survey #7: Female, age 38).

Some asked me `How could you let him kill the kids?’ A few other people I counted as close friends suddenly were distant. The press had picked up that when [the offender] was arrested he shouted that I had done it. Some of these `friends’ seem to have doubts about my involvement (Survey #6: Female, age 37).

These were individuals who felt blamed or thought they were somehow viewed as contributing to their plight. In some cases this stigmatization referred to ostensibly questionable circumstances surrounding the murder (e.g., the murder of drug dealers; families of women who remained with abusive partners). However, this was not necessarily the case, and many may have simply been the result of individuals’ attempting to apportion blame in line with a cultural belief that ``people get what they deserve’’ (Lerner 1980). Essen- tially, these were individuals who broke (or were perceived to break) another cardinal rule of sympathy etiquette: not mak- ing ``false claims’’ to sympathy.

The ultimate result of such interactions, whether motivated by stigma, uncertainty regarding how to behave, survivors’ or others’ own upset, or generally not wanting to further upset each other, was that survivors tended to become socially isolated when they needed support. Moreover, while this often upset them further, many chose to withdraw themselves as ``people in general don’t understand . . . it’s too diff®cult’’ (Interview #29: Female, age 37).

Summing up this topic, survivors indicating a wide variety of perceived support from their extended families and friends reportedly fared better. These survivors reported few insensitive comments and obvious avoidance behaviors from those previously close to them, along with support and encouragement. This clearly relates to the literature on

Victims of Crime and Labeling Theory 251

accommodation (Lynch 1983; Rubington and Weinberg 1987; Yarrow et al. 1955; Jackson 1954), but further suggests that sensitive, overt expression of sympathy worthiness associated with the label of victim played a signi®cant role. Ultimately, such survivors tended to report milder grief experiences overall, supporting the literature suggesting that ``social support is the most important factor in helping parents ®nd new social and psychic equilibria’’ (Klass 1988:179).

However, the majority of survivors who experienced many of the unhelpful, unsupportive interactions with friends generally assessed themselves as faring much worse than those with perceived long-term support. Indeed, they may experience additional losses. These interactions not only gave survivors more reasons to be emotionally upset, such accu- mulated indignities were often interrelated with the various labels associated with survivors (Wortman and Lehman 1983). In some cases unhelpful behaviors were associated with uncertainty, unexpressed or inappropriately manifested sympathy, and the label of victim; in other cases these behaviors were associated with stigmatization as victims or deviants, or stigmatization as ``emotional deviants’’ (Thoits 1990). Regardless of their label, such survivors were avoided, stigmatized, encountered dif®culties in communication, or had increased con¯icts. Moreover, survivors’ resulting social and emotional isolation, which was commonly employed as a defense or reaction to this labeling, could be cited as evi- dence of secondary victimization (Taylor, Wood, and Licht- man 1983), secondary deviance (Lemert 1951), or secondary emotional deviance, (Thoits 1990; Lemert 1951).

Next, and in contrast with their extended family and friends, respondents often noted receiving remarkable sup- port from mere acquaintances. This included many of the matters discussed earlier, but especially emotional support.

It was really strange because you ®nd out who you could lean on. You know, and sometimes it’s the people you least expect (Interview #16: Female, age 56).

We found some friends that we considered to be close friends never showed up again. Yet, there were other people that came out that we had sort of considered acquaintances, and we became very close (Interview #24: Male, age 47).

252 J. S. Kenney

It is hypothesized that acquaintances who labeled respon- dents as victims considered them as legitimately sympathy worthy as did sympathetic friends and family. However, their expression of sympathy was not as readily blocked by their own upset, familiarity, and personal grief.

Interestingly, respondents sometimes noted a groundswell of support from strangers in the community as well. Strangers sometimes volunteered to search for the deceased, sent cards, ¯owers, food, raised money, erected memorials, organized petitions, and urged respondents to take action. Two factors were associated with the wide sympathy margins ascribed to such survivors: First, either the deceased, or their survivors, were well known, had much prior community involvement, or both:

[the deceased] had a lot of friends. She was a very popular girl. She had a lot of friends from school. And I was very active in the community. I belonged to the Optimist Club, I coached baseball, I coached hockey, and so onÐso I was very active, as far as that goes, with kids and all that all the time. I worked in other various organizations, working bingos and fundrais- ing. So, I mean, there were a lot of people that it stunned, too. Then, as the trial went on, they were reading the newspaper and were getting upset with what was being said, and most of the people got together and wanted to do something. So, basically that’s how it started (Interview #23: Male, age 49).

The second factor was widespread media coverage sympa- thetic to the survivors, illustrating the impact of the mass media on sympathy margins:

It was not until the Saturday that it made the papers. The local paper did a good story, and people started coming to the door then, and we had literally hundreds of total strangers just arriving at out door in tears, very upset. It was obviously that their prayers were with us. Food started arriving in trays, ¯owers, baskets of fruit, it was just incredible. From 10 a.m. to 10 p.m. there was just a steady stream of traf®c, and probably thrity to forty people at all times (Interview #31: Female, age 46).

Indeed, it appeared that several respondents were cast into the role of victim advocates in such a context. One woman,

Victims of Crime and Labeling Theory 253

who has made a career as a moral crusader for victims’ rights, recounts:

I went on the radio [to thank the community search teams]. A policeman phoned in and said `I searched for [the deceased], now what are you going to do?’ And that’s how I started. I wrote a petition, and this whole thing kept going and kept going. So that is how [this survivor’s organization] happened. People just kept coming to us and saying `what are you going to do?’ There was never a day when the phone didn’t ring thirty times (Interview #17: Female, age 50).

Such widespread community support, in several cases, was the genesis of more enduring victims’ rights organizations. Moreover, in such altercasting (Weinstein and Deutschberger 1963) we may see the potential social basis for such survivors’ tertiary victimization (Taylor, Wood, and Lichtman 1983; Kitsuse 1980) or tertiary emotional deviance (Thoits 1990; Kitsuse 1980), where the victim role is converted into something positive, meaningful, and socially ef®cacious.13

However, in cases where the deceased and=or survivors were not well known, sympathy margins were lessened. Where interactions with the media also did not go well, there was little interest in the story, or where the press’ investigation of the crime resulted in an unfavorable portrayal, the miti- gating effect on sympathy margins, community support, and survivors was magni®ed. One woman asked: Why isn’t our daughter being talked about? (Field notes: Female, age 50). Another added:

The publicity made it worse. The fact that the murder had to deal with drug dealing made it much worse, as there was no sympathetic community support. It was almost ``like AIDS’’ (C.I.C.B. #91: Testimony of Female, age 46).

Aside from generalized lack of sympathy, such circum- stances and related statuses were frequently characterized by an increased potential for stigmatization, either as a victim,

13 While the issue of applying the term ``deviant’’ to those organizing lobby self-help groups may be questioned by some, this is firmly in line with how these individualsÐand othersÐperceive them (i.e., as somehow different). However, this does not mean that they, or their emotional self-presentation, must be seen as immoral. In the words of one such man ``our normals are different now.’’

254 J. S. Kenney

deviant, or emotional deviant. Three types of negative inter- action followed. First, survivors experienced harassment:

Like, we got obscene phone calls to the house. The minute that we became public, I got phone calls that ``If you and your family had belonged to the right religion, these horrible things wouldn’t happen to you. God is punishing you.’’ We’d get phone calls like that. One day this guy phoned and asked if I was the mother of one of these children that had been mur- dered. I said I was, and he says ``Well, would you stay on the line while I masturbate?’’ And like this is for real! This is going on like every hour around here. LikeÐnot just sporadically days apartÐthis is for real! And it’s going on constantly (Interview #15: Female, age 49).

Secondly, survivors experienced a problem with blaming, particularly in a context of ongoing speculation, innuendo, and gossip. For example, one mother comments:

One young girl came to the funeral home. I think when you have that type of thing, some come just to be nosy, like, to see what they can see. She had just come with these other girls, and said ``Well, the little slut got what she deserved.’’ People had to hold us back! (Interview #21: Female, age 45).

Finally, notoriety resulted in uncomfortable interactions when going about one’s business in public. This became a problem for some survivors:

You know, you’re a bit of a freak for a while because people don’t really understand: ``Oh, that’s the woman that lost her daughter.’’ People point you out. When they hear the name, they say, ``Oh, are you the [respondent’s ®rst name]?’’ Like, that kind of thing. And you go ``No, I’m just me.’’ I kind of lose it there (Interview #16: Female, age 56) (Emphasis in original).

A male respondent also said:

We went on a cruise, a company cruise. When we’d come down to dinner, we’d be the last, so everyone’s sitting around their tables laughing and having a good time. Whatever table we’d pick, Whew! Dead silence. In the end we sat by our- selves in the corner. Cause when you sit at a table with people who are feeling that way, I had to start the conversation, and you know what it’s like to try and talk . . . I mean it’s bullshit

Victims of Crime and Labeling Theory 255

that you have to come up with these little weak things to keep the conversation going (Interview #26: Male, age 61).

Ultimately, survivors who reported encountering such negative responses from acquaintances, strangers, and the community were far more likely to feel that they were faring worse emotionally. Indeed, it could be argued that these negative interactions revictimized those who reported ex- periencing them, leading again to secondary victimization (Taylor, Wood, and Lichtman 1983).

Summing up, these ®ndings regarding differential support from acquaintances, strangers, and the community are intri- guing, particularly insofar that such individuals appeared in many cases to respond more sympathetically than extended family and friends. It is hypothesized that, despite agreement on the sympathy-worthiness of survivors among these groups, supportive acquaintances and strangers were not as nervous about upsetting survivors, or becoming upset themselves. Of further interest was the potential for popular protest, fed by the media, to not only provide meaningful support to survivors, but also to inculcate positive new role identities (e.g., victims’ advocate) indicative of tertiary victi- mization (Taylor, Wood, and Lichtman 1983; Kitsuse 1980) or tertiary emotional deviance (Thoits 1990; Kitsuse 1980). Of course, the more negative interactions simply revictimized survivors, adding to the secondary labeling process noted earlier.

Turning, ®nally, to subjects’ responses, it is important to note that survivors in the wake of a murder essentially faced one major decision: either to attempt to deal with the murder primarily on their own, or to seek out help from other indi- viduals, groups, and institutions. Their responses generally involved an interaction between their initial, gendered orientation toward seeking help and a variety of incentives and disincentives found in social interaction. The ®rst strategy was more frequently utilized by men, who often avoided seeking help from medical professionals, self-help, and=or victims groups. Women appeared to be far more inclined to seek out help from others, and this was re¯ected in the data in their reports of far greater levels of seeking out medical and psychiatric care, involvement in victims’ and self-help groups, and economic support from social service agencies

256 J. S. Kenney

when unable to maintain employment. To give just one example, a survivor, who also happens to be a practicing therapist who treats victims, stated that her clientele was approximately ``one quarter men, the rest women’’ (Interview #6: Female, age 46).

In addition to survivors’ initial orientation to seeking help, they faced a variety of incentives and disincentives to do so related to the labeling process. Frequently, there was an obvious relation between perceived or actual level of support and survivors seeking outside help. For example, there were survivors who eventually sought outside help because of increasing social isolation:

I did go to a psychiatrist on a regular basis, because [sighs] after a while your friends don’t want to hear about it any more. They just don’t want to talk about it, or they’re uncomfortable talking to you about it (Interview #16: Female, age 56).

If it wasn’t for the group, I would be so isolated, and I wouldn’t have a network, and I wouldn’t have . . . soul companions . . . Because our families won’t let us talk about it, cause they’re dealing with guilt or something of their own, and they won’t allow us to really express ourselves and hear this stuff, you know? So we do it amongst ourselves (Interview #19: Female, age 53).

Conversely, others felt they did not need to seek help as they had suf®cient support already:

We didn’t require any counselling. We’ve got good family around us, and that’s how we’ve dealt with it (Interview #24: Male, age 47).

A few times I thought I should’ve gone to the hospital, but I didn’t want to. I’m lucky I have friends. It’s nice to have a support system behind you. You know, a support system is very important to mental health (Interview #1: Female, age 47).

These examples illustrate how sympathetic labeling as victims, as in the latter case, can accommodate survivors in such a way that they are not isolated in subcultures or rele- gated to formal control agents. On the other hand, social stigmatization as either victims, deviants, or emotional

Victims of Crime and Labeling Theory 257

deviants can increase survivors’ chances of contact with subcultures (such as support groups), or more formal agents of social control (e.g., those in a position to apply medical labels). Yet, level of support did not tell the whole story. Going hand in hand with this, survivors also spoke of encouragement to seek help. In some cases, survivors noted that othersÐparticularly those with whom they could iden- tifyÐtactfully and altruistically encouraged them to do so:

I have a friend who lost his sonÐI guess it was some sort of cancer. Well, because of his loss, he had sought help with [a self-help group]. So, knowing what had happened to us, he says ``I’d like to talk to you for a minute.’’ We talked a bit, and he went on to tell me about the group. He told me about himself, and how his grief had been helped. He highly recommended it, saying that he wouldn’t have survived without them. It was interesting, so I approached my wife (Interview #4: Male, age 56).

Yet other survivors, most often men, appeared to resent such ``interference’’ and avoided seeking help, which sug- gests that gender plays a role as well. Finally, as this suggests, there was the element of choice. Consider the words of the following woman who, after unsupportive interactions with her family during the funeral, and avoidance thereafter, chose to become involved in a newly formed survivors’ support group:

I had people there who I could talk to, and that was good for me. In fact, right or wrongly, I made the decision that would be the only place I’d ever talk about this. I would not talk to my friends or family unless I had to. But fortunately, you know, a survivor will ®nd a wayÐand that’s exactly what I did. I felt ``Hey, who needs them?’’ (Interview #18: Female, age 55) (Emphasis added).

Theorists have argued that labeling can have both adaptive and maladaptive consequences (Plummer 1979:118). Here it appears that informal labeling was important in either directing survivors to therapy and self-help or ``forcing’’ them into further social withdrawal. It should be noted that a breakdown in perceptions of sympathy-worthiness, and the corresponding aspect of the victim label, undermines accomodation. This results in either a shift in emphasis to the

258 J. S. Kenney

stigmatized, helpless side of the victim label, or the labeling of individuals as deviant, in this case emotionally deviant. Indeed, in some cases where they are not perceived as sympathy worthy at the outset, victims can be blamed and seen as deviants deserving of their fate. All of these outcomes can have a signi®cant impact on their self-identity (Rubington and Weinberg 1987:289±384). In such cases, the concepts of secondary victimization (Taylor, Wood, and Lichtman 1983) and secondary emotional deviance (Thoits 1990; Lemert 1951), in which survivors begin to employ their victimiza- tion, deviant behavior, or a role based upon it as a means of defense or adjustment to the problems created by this social reaction, are strongly suggested.

Finally, it is important to consider that, both within and outside their respective subcultures (Rubington 1982), these individuals sometimes embraced, at other times distanced themselves from these various roles, or aspects thereof (Goffman 1961). Indeed, it appeared that a form of self-pre- sentation (Goffman 1959) was sometimes utilized, which I term ``volitional gerrymandering.’’

Summing up, survivors’ initial orientation to seeking help were partially rooted in traditional gender roles. However, social incentives and disincentives appeared to play a sig- ni®cant part in survivors’ coping choices. Overall perceived level of support from family, friends, and the community, coupled with gender, the relative maintenance or breakdown of accommodation by others to the emotional upset of sur- vivors, and individual coping choices all played a role in the evolution of survivors’ interactions beyond the ``unof®cial’’ sphere, to those with more ``formal’’ helping agents. Often, the labeling process appeared to be at work here (Plummer 1979; Lemert 1951). This was re¯ected in both secondary and tertiary victimization (Taylor, Wood, and Lichtman 1983; Kitsuse 1980) and corresponding aspects for deviance (Lemert 1951; Thoits 1990; Kitsuse, 1980). In either case survivors employed the victim role as a shield or as a sword in inter- actions, as well as in their involvement with subcultures (Rubington 1982). While all of this can have a signi®cant impact on an individual’s self-concept, particularly when a ``deviant identity’’ is being inculcated (Rubington and Weinberg 1987), one must also bear in mind the potential for

Victims of Crime and Labeling Theory 259

self-presentation. evinced by some subjects’ use of ``volitional gerrymandering.’’ 14

DISCUSSION AND CONCLUSION

The results of this research con®rm and elaborate many of the themes in the labeling literature, but extend them theoreti- cally by identifying a parallel labeling process for victims. Moreover, the intricate interplay between this and the deviance labeling process were empirically illustrated through a study of homicide survivors.

First, the varieties of helpful interactions may be con- ceptualized as re¯ecting the sympathy margins ascribed to survivors by others (Clark 1987). In those cases where wide sympathy margins are ascribed and helpfully expressed to survivors, survivors are labelled as victims but accom- modated as deviants. In these cases, it is not so much indi- viduals’ grief that is de®ned as deviant, but the circumstances that initiated it. Survivors are recognized as victims, not blamed as such.

Secondly, it is important to note that such labeling as vic- tims and recognition of sympathy worthiness is often also the case when survivors do not ®nd others’ behavior helpful. In these circumstances sympathetic others either do not know what to say, avoid survivors, or express their sorrow and care for survivors in well-meaning but hurtful ways in victims’ eyes. The difference between the responses of extended family and friends, on the one hand, and acquaintances, strangers, and the community, on the other, was particularly interesting. This corroborates and extends the literature on social response to victims of life crises to a new empirical context (Lepore et al. 1996; Holman and Silver 1996; Wortman and Lehman 1983; Silver and Wortman 1980; Wortman and Dunkel-Schetter 1979).

14 The issue of whether labeling amplifies or creates victimization must be answered on two levels. On one hand, the labeling described above has been shown to have a significant effect on both subjects’ self-identity and their subsequent behaviors (e.g., social withdrawal, participation in counseling, self-help, and lobbying activities, along with differential self- presentation). On the other hand, the question of whether they are more likely to be victims of future violent crimes is unclear, as it is beyond the scope of the data utilized in this study.

260 J. S. Kenney

Third, it is clear that survivors are not always recognized as victims deserving of sympathy, or as such inde®nitely. Sym- pathy margins vary in width, and are governed by the rules of sympathy etiquette. In those cases where survivors break the rules of sympathy etiquette, either by claiming too much sympathy for too long, not making legitimate claims to sym- pathy, and the like, they may risk stigmatization. This may take several forms as illustrated above.

The data illustrating sympathetic social support, on the one hand, and active avoidance of survivors by their extended family, friends, and members of the community, on the other exempli®ed these labeling variations. This was supplemented by survivors’ own rationales of why this occurred, including varying perceptions of stigmatization, and others’ fear of upsetting them. Indeed, survivors’ additional reports of poor communication and inappropriate attention such as harass- ment and unwelcome comments further supported these variations in labeling. Signi®cantly, when accommodation of survivors’ upset broke down in these ways, many survivors became socially isolated and utilized others’ responses as a rationale and defense. This is clearly evidence of secondary victimization (Taylor, Wood, and Lichtman 1983) or sec- ondary emotional deviance (Thoits 1990; Lemert 1951), depending on the predominant focus of the labeling.

In other cases, some survivors, such as those who became caught up with others in community protests and ®ghting for change, showed evidence of engaging in tertiary victimiza- tion (Taylor, Wood, and Lichtman 1983; Kitsuse 1980) or tertiary emotional deviance (Thoits 1990; Kitsuse 1980). In such cases, a new, positive self-image emerges in interaction such that victims advocates are seen as working toward positive ends for all, using their role as a sword in the process.

Ultimately, however, once accommodation breaks down and labeling begins, regardless of whether survivors use the victim role as a shield or as a sword, this becomes relevant to the literatures on subcultures (Rubington 1982) and deviant identity (Rubington and Weinberg 1987). Yet, the fact that some embraced, while others distanced themselves from these identities, and employed them selectively, suggests that volitional gerrymandering also has signi®cance for self-pre- sentation and micropolitics (Goffman 1959; Emerson and Messinger 1977; Clark 1990).

Victims of Crime and Labeling Theory 261

Now that the literature on accommodation and labeling has been found relevant to the experience of victims, it is necessary for future researchers to follow up these insights. For example, a study of the subculture of victims’ lobby and support groups is suggested. In particular, it would be inter- esting to examine in more detail the degree to which the victim role is used as a sword or a shield, by whom, and under what circumstances. It may be that some individuals have largely internalized one or the other of these as a ``master status’’ (Hughes 1945); others may exhibit role dis- tance and rhetorically present the most useful role in each interaction (Goffman 1959, 1961). A focused, observational study of such volitional gerrymandering would be most illuminating.

In addition, a study of the medicalization of victims’ behavior, keeping in mind Conrad and Schneider’s (1980) discussion of the implications of the medical model, seems to be warranted for interactions between victims, support groups, and mental health professionals. Finally, a study of ``victims’’ encounters with these, and other formal control agents such as police and the justice system, may shed further light on the consequences of the breakdown of accom- modation, the formal labeling of victims versus deviants, and the incidence of crime generally. Only by including the vic- tim, in such a truly interactional manner, can we begin to live up to the potential, as well as the philosophical antecedents, of the labeling perspective.

REFERENCES

Becker, Howard. 1963. Outsiders: Studies in the Sociology of Deviance. Glencoe: Illinois The Free Press.

Berg, Bruce L. 1995. Qualitative Research Methods for the Social Sci- ences, (2nd ed.) Needham Heights, MA: Allyn & Bacon.

Clark, Candace. 1987. ``Sympathy Biography and Sympathy Margin.’’ American Journal of Sociology 93(2):290±321.

Clark, Candace. 1990. ``Emotions and Micropolitics in Everyday Life: Some Patterns and Paradozes of `Place’.’’ In Research Agendas in the Sociology of Emotions, edited by Theodore D. Kemper. New York: SUNY Press.

Coates, Dan and Tina Winston. 1983. ``Counteracting the Deviance of Depression: Peer Support Groups for Victims.’’ Journal of Social Issues 39(2):169±94.

262 J. S. Kenney

Conrad, Peter and Joseph W. Schneider. 1980. Deviance and Medica- lization: From Badness to Sickness. Columbus, OH: Merrill.

Douglas, Jack D. 1976. Investigativ e Social Research. Beverly Hills, CA: Sage.

Douglas, Mary. 1970. Natural Symbols: Explorations in Cosmology. London: Barrie & Rockliff.

Douglas, Mary. 1978. Cultural Bias. London: Royal Anthropological Institute.

Emerson, Robert M., and Sheldon L. Messinger. 1977. ``The Micro-Pol- itics of Trouble.’’ Social Problems 25(2):121±34.

Freud, Sigmund. 1957. ``Repression’’ Pp. 146±58 in The Standard Edition and Complete Psychologica l Works of Sigmund Freud. vol. 14, edited by J. Strachney. London: Hogarth.

Glaser, Barney and Anselm L. Strauss. 1967. The Discovery of Grounded Theory: Strategies for Qualitative Research. Chicago: Aldine.

Goffman, Erving. 1959. The Presentation of Self in Everyday Life. New York: Doubleday.

Goffman, Erving. 1961. Encounters : Two Studies in the Sociology of Interaction. Indianapolis, IN: Bobbs-Merrill.

Hochschild, Arlie. 1983. The Managed Heart: Berkeley, CA: University of California Press.

Holman, E. Alison and Roxanne C. Silver. 1996. ``Is it the Abuse or the Aftermath? A Stress and Coping Approach to Understanding Responses to Incest’’ Journal of Social and Clinical Psychology 15(3):318±39.

Hughes, Everett C. 1945. ``Dilemmas and Contradictions of Status.’’ American Journal of Sociology L (March):353±59.

Jackson, Joan K. 1954. ``The Adjustment of the Family to the Crisis of Alcoholism.’’ Quarterly Journal of Studies on Alcohol 15 (Decem- ber):564±86.

Kitsuse, John I. 1980. ``Coming Out All Over: Deviants and the Politics of Social Problems. ’’ Social Problems 28(1):1±12.

Klass, Dennis. 1988. Parental Grief: Solace and Resolution. New York: Springer.

Knapp, Ronald J. 1986. Beyond Endurance: When a Child Dies. New York: Schocken.

Kowalewski, Mark R. 1988. ``Double Stigma and Boundary Maintenance: How Gay Men Deal with AIDS.’’ Journal of Contemporary Ethno- graphy 17(2):211±28.

Lemert, Edwin M. 1951. Social Pathology. New York: McGraw-Hill. Lepore, Stephen, J., Camille B. Wortman, Roxanne C. Silver, and Heidi

A. Wayment. 1996. ``Social Constraints, Intrusive Thoughts, and Depressive Symptoms Among Bereaved Mothers.’’ Journal of Person- ality and Social Psychology 70(2):271±82.

Lerner, Melvin J. 1980. The Belief in a Just World: A Fundamental Delusion. New York: Plenum.

Victims of Crime and Labeling Theory 263

Llamas, Ricardo. 1994. ``The Reconstruction of the Homosexual Body in the Time of AIDS.’’ Revista Espanola de Investigacione s Sociologica s 68 (October±December): 141±71.

Lo¯and, Lyn H. 1985. ``The Social Shaping of Emotion: The Case of Grief.’’ Symbolic Interaction 8(2):171±90.

Lynch, Michael. 1983. ``Accommodation Practices: Vernacular Treat- ments of Madness.’’ Social Problems 31(2):152±64.

Ontario Government Consultation on Victims of Violent Crime Report. 1984. Toronto: Provincial Secretariat For Justice.

Parsons, Talcott. 1951. The Social System. New York. The Free Press. Plummer, Kenneth. 1979. ``Misunderstanding Labelling Perspectives.’’

Pp. 85±121 in Deviant Interpretation s, edited by David Downes and Paul Rock. Oxford: Martin Robertson.

Rubington, Earl. 1982. ``Deviant Subcultures.’’ Pp. 57±60 in The Sociology of Deviance, edited by Michael Rosenberg, Robert A. Stebbins, and Allan Turowetz. New York: St. Martin’s Press.

Rubington, Earl and Martin S. Weinberg. 1987. Deviance: The Interac- tionist Perspective. 5th ed. New York: MacMillan.

Russell, Gordon W. and Robert K. Mentzel. 1990. ``Sympathy and Altruism in Response to Disasters.’’ Journal of Social Psychology 130(3):309±17.

Rynearson, E. K. and Joseph M. McCreery. 1993. ``Bereavement After Homicide: A Synergism of Trauma and Loss.’’ American Journal of Psychiatry 150:258±61.

Schwartz, Howard and Jerry Jacobs. 1979. Qualitative Sociology: A Method to the Madness. New York: The Free Press.

Silver, Roxanne L. and Camille B. Wortman. 1980. ``Coping With Undesirable Life Events.’’ Pp. 279±340 in Human Helplessness: Theory and Applications, edited by Judy Garber and Martin E. P. Seligman. New York: Academic.

Taylor, Shelley E., Joanne V. Wood, and Rosemary R. Lichtman. 1983. ``It Could Be Worse: Selective Evaluation as a Response to Victimiza- tion’’ Journal of Social Issues 39(2):19±40.

Thoits, Peggy A. 1990. ``Emotional Deviance: Research Agendas.’’ Pp.180±203 in Research Agendas in the Sociology of Emotions, edited by Theodore D. Kemper. New York: State University of New York.

Weinstein, Eugene and Paul Deutschberger. 1963. ``Some Dimensions of Altercasting.’’ Sociometry 26:454±66.

Williamson, John B., David A Karp, and John R. Dalphin. 1977. The Research Craft: An Introduction to Social Science Methods. Boston: Little, Brown.

Wortman, Camille B. and Christine Dunkel-Schetter. 1979. ``Inter- personal Relationships and Cancer: A Theoretical Analysis.’’ Journal of Social Issues 35(1):120±55.

264 J. S. Kenney

Wortman, Camille B. and Darrin R. Lehman. 1983. ``Reactions to Victims of Life Crises: Support Attempts that Fail.’’ Pp. 463±89 in Social Support: Theory, Research and Applications, edited by Irwin G. Sar- ason and Barbara R. Sarason. Boston: Martinus Nijhoff.

Yarrow, Marian R., Charlotte G. Schwartz, Harriet S. Murphy, and Leila Calhoun Deasy. 1955. ``The Psychological Meaning of Mental Illness in the Family.’’ Journal of Social Issues 11(4):12±24.

Young, Marlene A. 1991. ``Survivors of Crime.’’ Pp. 27±42 in To be a Victim: Encounters with Crime and Injustice, edited by Diane Sank and David I. Caplan. New York: Plenum.

Victims of Crime and Labeling Theory 265