The Future of Labeling Theory: Foundations and Promises
Charles F. Wellford Ruth A. Triplett
In the twentieth century tile history of criminological theory has been one of competing theoretical perspectives. Throughout this period numerous explanations for crime and deviance have emerged that have been offered as better explanations than other theories or perspectives.1 Research is done on one of the perspectives. The research demonstrates tile limited value of that theory or perspective and the community of criminological scholars moves on to develop a new theory, test a new theory, or lament tile stagnation in criminological theory.
In recent years, renewed and increased attention has been given to the need to organize a variety of theories into an interdisciplinary or integrated theory that captures tile contributions that call be made from the many explanatory approaches that have emerged over tile last one hundred years (Messner et al. 1989). This move towards integrated or interdisciplinary theory represents a new stage of development in the field and requires a careful reassessment of the perspectives that have formed the core of criminological thought.
Labeling theory represents for criminological thought a major school of social thought-symbolic interactionism. Labeling has as its foundation symbolic interactionist theory and, therefore, represents for criminologists one way in which a prominent school of social thought could be included into a comprehensive explanation of crime and deviance. In earlier analyses of labeling theory authors have, in tile tradition noted above, attempted to assess its value as an explanation of crime or deviance (e.g., see Wellford 1975, 1987; Mahoney 1974; Thomas and Bishop 1984). Our goal in this paper is not to establish that labeling theory on its own cannot explain ill any meaningful way significant amounts of criminal behavior (given Out- interest in integrated theory we take that to be axiomatic) but to assess tile essential elements of labeling theory, to review its likely contributions, and to describe the developments that we think are necessary to capture the advantages that labeling theory would bring to a comprehensive explanation of crime or deviance.
In this paper we will first briefly sketch the foundations of labeling theory. Then we will discuss the problems of labeling theory that have been faced in tile criminological literature. This will be followed by a review of empirical research indicating the elements of labeling theory that may be most critical to include in a comprehensive explanation of crime, and a discussion of the emerging trends in labeling theory research. Finally, we will conclude with an assessment of what we think should be done to advance labeling theory as part of a more comprehensive explanation of crime.
The Foundations of Labeling Theory
In one of the most thoughtful discussions of the origins of human behavior, Leslie White observed, "In the word was the beginning, the beginning of man and of culture" (White 1969, 22). White's analysis of the role of "the word" or as lie preferred to refer to it, "the symbol," represents a clear statement of an important tradition in social thought. Tills tradition emphasizes that the unique characteristics of humans and of their behavior is in tile use of symbols. Again to quote White,
it was the symbol which transformed our anthropoid ancestors into men and made them human. All civilizations have been generated and perpetuated only by the use of symbols. It is the symbol which transforms an infant of homo sapiens into a human being. All human behavior consists of or is dependent upon the use of symbols. Human behavior is symbolic behavior. Symbolic behavior is human behavior. The symbol is the universe of humanity. (White 1969, 22)
The symbol then represents that summation of tile characteristics of human beings that set them apart from their animal ancestors and contemporaries. The symbol represents an important element within social thought.
As defined by White the symbol is, "as a thing, the value or meaning of which is bestowed upon it by those who use it" (White 1969, 25). Thus, the symbol represents that which the individual attaches to something in tile environment or to oneself. Tile symbol can represent a physical thing, a social thing, a biological thing, or any element of the environment, internal or external, in which the individual operates. Tile critical element, then, of being human is the ability to interpret one's environment, to attach symbols to it.
This approach, made most explicit by Leslie White, was captured for American sociology much earlier by the concept of the looking-glass self as enunciated by Charles Horton Cooley. That concept emphasized for social theorists the role of an individual's perceptions (or in White's terms, symboling) of other people's behavior in the formation of that individual's behavior. To quote from Cooley (Cooley 1964, 189), "the looking-glass self consists of three parts: the imagination of our appearance to tile other person, the imagination of his judgment of that appearance, and sonic sort of self feeling, such as pride or mortification." For Cooley, tile symbolic process involves the manner in which the individual interprets tile social environment. The individual's characteristics and the social environment are neither in tension nor time ordered. They are related to each other in an interactive way, such that the self and its social environment emerge coterminously. For Cooley there is no distinction between the individual and the environment, they are inextricably related.
While Cooley's work is often cited in American sociology as a foundation for those theories that emphasize symboling, perhaps the more influential theorists in this tradition is George Herbert Mead. Mead's analysis was more complex because of his understanding of the relationship between the social environment and tile individual. Mead's concepts of I and me represent attempts to describe the way in which the individual both effects and is effected by the social environment through the process of interpretation. The concept of the generalized other, which represents an entity comparable to Freud's superego, provides another important theoretical component to Mead's work. For our purposes, there are two essential components of Mead's social thought. The first is tile recognition that roles and rules play a critical part in the formation of the self. The second is flow, after a self begins to interpret tile environment, react to and understand these rules, tile self then begins to act back on the environment. The potential of tension or conflict between these two aspects of self, present clearly in Mead and Freud but not in Cooley, represent an important advance in social thinking in this tradition-one of critical importance to labeling theory.2
For those who emphasize symboling, however, the critical element is the interpretation of others' activities. This subjective process of interpretation represents the critical element in this tradition in understanding the formation of an individual's understanding of oneself, tile determination of the link between self-assessments and behavior, and the role that the individual plays within a society. This tradition, while always having been recognized within American sociology, has not been a dominant one.
In American sociology, given tile emphasis during the last fifty years on tile scientific aspect of the discipline, those traditions that emphasized internal, unobservable, and individualized concepts tended to be given less status and less attention. As sociology began to emerge to develop its own unique identity, its emphasis on structural, cultural, external elements in the Durkheimian tradition placed symbolic interactionism at a much lower level of importance than other theoretical traditions. As noted earlier, our position is that such attempts to limit theoretical models is at best problematic. For a discipline such as criminology that is trying to explain a particular behavior rather than a discipline-created behavior, tile theoretical approach to explanation needs to be as comprehensive as possible.3 Therefore, even though it represents great difficulties for scientific analysis, Our contention is that symbolic interactionism and its representation within tile criminological field (labeling theory) should be given strong consideration as a component in a comprehensive understanding of crime and deviance.
It is not necessary to once again describe the essential elements of labeling theory. This has been done many times by many authors (e.g., Schrag 1971; Siegel 1986). Our interest is in understanding how labeling theory has emerged within criminology and why its contributions have been so minimal. Neither do we wish to review tile extensive research that has been done on labeling, most of which has noted its empirical shortcomings (for a recent summary see Adler et al. 1991). Rather we wish to discuss three reasons that we believe explain why labeling theory has not achieved the level of importance that it can have in a comprehensive theory of crime and deviance. These are first, the previously noted tendency in much of criminological theory to offer relatively narrow theoretical models to explain all of deviance; second, the tendency in the 1970s and 1980s for labeling to be connected with conflict theory; and third, the failure of labeling theorists and advocates to pay attention to the symbolic nature of labeling.
When Clarence Schrag (1971) Summarized tile propositions of labeling theory, tile first element he identified was that no act was intrinsically criminal. He further stated that a component of labeling theory was the proposition that laws arc made for and enforced in favor of more powerful interests. Nowhere in the symbolic interactionist literature or in the writings of those who have been most careful in delineating the labeling perspective, does one find these propositions. Still, Schrag's analysis was correct for that time. As labeling theory emerged, it emerged as a critic of government institutions established to control crime. Its central notion was that these institutions may, through their actions, in fact be perpetuating and exacerbating the problem they are established to control. This theme is consistent with tile writings of conflict theorists oil the formation of law and its operation. Tile similarity of these themes led many who were initially drawn to labeling theory to expand their thinking to include other aspects of critical criminology.
To the extent that labeling theory began to be a part of or merged with conflict theory (Davis 1972; Quinney 1973), it opened itself Lip to the obvious criticisms that have been made of that perspective. Both analytically and empirically these propositions turned out to be of questionable value in democratic societies and, therefore, labeling theory suffered from its affiliation with this approach. It is important, however, to emphasize that the labeling perspective is no more linked to critical or conflict theoretical models than it is to Structural, historical, or any other explanation of behavior.
In his analysis of the importance of symbolic interactionism, for understanding crime and deviance, Edward Lemert laid out clearly the ways in which labeling theory could be linked to other theoretical models (Lemert 1972). This linkage tends to be associated with the distinction between primary and secondary deviance, that is, primary deviance could be explained by other theoretical models, whereas secondary deviance would be explained by labeling. The need identified by Lemert to link labeling to other theoretical models was not developed by later theorists. On the contrary, the works of Schur (1971) and others (e.g., Scheff 1974) took a very narrow view of labeling, understanding it as the primary if not sole explanation of secondary deviation. The concept of primary deviation was generally left without explanation except by Lemert, who carefully considered its sources and relationship to secondary deviance (e.g., Lemert 1972, 48). Thus, while sonic labeling theorists recognized at least an understanding or the distinction between primary and secondary deviance and tile need to link labeling to other theoretical models, the primary thrust was to focus on labeling as (the) cause of crime and deviant behavior. As criminology has progressed we have, as noted above, come to understand the difficulties of such single-dimension theoretical models and have opted for interdisciplinary and integrated theories. The future of criminology clearly rests on this approach to theory building and the value of labeling will only increase as we begin to understand other theoretical perspectives and complete tile decoupling of labeling and conflict theories.
Another problem that has limited the contributions labeling has made to theories of crime and deviance is the failure of labeling theorists and researchers to understand or to remember the central elements of labeling theory. As a symbolic interactionist perspective, the primary concern in labeling should be on tile individual's interpretation of the response of others. Yet, labeling theorists focused primarily on the simple application of the label. There was no attention to the changes that were actually occurring in the self, how these might be effected by other characteristics of the individual, or how the strength of the label might also be a part of understanding its impact.
This tendency to deal only with the manifestation of the label and not its interpretation is similar to what has occurred in deterrence theory. As many have pointed out (e.g., Paternoster 1987), deterrence theory emphasizes primarily the role of informal and experiential deterrence rather than formal deterrence. And yet, much of the thinking and early work in criminology focused upon external, formal deterrents. Only recently have we moved to understanding the importance of informal deterrence and more importantly, the individual's perception of deterrence and its impact. Thus, in both deterrence and labeling theory we have seen a tendency to emphasize the external aspect without attention to tile internal. This is particularly troubling for labeling because tile central Value of labeling theory is its emphasis on tile interpretation that occurs when an individual responds to the social environment.
Therefore, labeling theory has suffered from the inattention to its central element, the symbolic process, by its tendency to drift into an association with critical and conflict criminology, and the tendency to treat labeling as a single explanation of secondary deviance. These are not necessary elements of labeling theory. In fact, in Lemert's analysis one finds none of these components and yet in tile interpretation and use of labeling theory by criminologists in the 1970s and 1980s these have been essential ingredients. As we move into the 1990s labeling theory has tile potential to become of much more value to theories of crime and deviance as it moves away from these misleading and misguiding approaches and we seek new ways to gain tile full value of a symbolic interactionist perspective for understanding crime and deviance.
The recent advances made in the development and testing of labeling or as it is increasingly referred to today the deviance amplification hypothesis may be divided into three different approaches. The first approach deals with specifying those factors on which the effects of formal labels are contingent. The second approach attempts an integration of aspects of labeling theory with other criminological theories. One such attempt integrates labeling with social learning theory. Another (Farrell 1989) focuses on a "psychological elaboration" of an integrated theory drawing on strain, interaction, and social cultural theories. Finally, the third development centers around changing the focus of tile type of label under consideration. As noted above, past research has focused almost exclusively on formal labels such as arrest, conviction, or sentencing (see Thomas and Bishop 1984 for an exception). Two new attempts at revitalizing labeling theory focus on tile effects of informal, subjective labels. This section of the paper discusses each of these developments.
In a recent review of labeling, Paternoster and Iovanni (1989) suggest that one reason empirical research is not supportive of labeling is that researchers have not specified tile conditions under which labeling is effective. This approach to labeling involves two areas of research and theoretical development. Tile first deals with specifying tile variables with which labels may interact such as group membership. Tile second area attempts to clearly define the process through which labeling works.
The group membership of an actor may interact with tile label to define the label's effect. Harris (1976) posits that labeling someone who by their "cast" does not already have full social membership in tile society is redundant (433). Thus labeling such an individual will have little effect on self-identity or behavior. Labeling someone who is accepted as a full member of society, however, may have dramatic effects. For this individual labeling devalues the individual's status and group membership.
Past research provides Some Support for these predictions, demonstrating that labeling effects are weakest among blacks and the lower class (Ageton and Elliott 1974; Harris 1975, 1976; Jensen1972; and Thornberry 1971). However, further development and testing of this hypothesis is needed. Besides tile need for verification of past findings it is important to continue theoretical development in this area. Future research might focus on further specifying the types of group membership which make labels redundant or ineffective. For example, Paternoster and Iovanni (1989) suggest that the group membership of both tile actor and the labeler is important in determining effects.
There is more to this approach, however, than simply specifying different groups for which labels' effects vary. The task also includes defining tile process through which labels enhance tile probability of delinquent behavior. One criticism of labeling is that official labels are neither necessary nor sufficient conditions for movement to a sustained pattern or criminal activity (Mankoff 1971). However, labeling theorists do not argue that deviance amplification is simply a matter of labels directly causing delinquent behavior. Ever since Lemert (1951, 1972), they hypothesize that the movement to secondary deviance is a process actors go through.
In describing the process, labeling theorists insist on a variety of conditions that must accompany or follow tile label for it to work as a deviance amplifier. For example, Becker (1963) emphasizes tile importance of membership in an organized group which promotes tile learning of deviant motives and interests. Lofland (1969) describes a variety of conditions which may "facilitate" the movement to secondary deviance. Sonic of tile conditions which lie specifics include self-orientation of tile actor, and degree of attachment of actor to normal or deviant others. A change in self-concept and role engulfment are other factors long associated with the movement to secondary deviance.
After an extensive review of the literature, Paternoster and Iovanni (1989) outline the most recent attempt in defining the process through which labels increase tile probability of delinquent behavior. The process begins with the labeling event that may or may not become publicly known. Tile label, if publicly known, call then lead to either exclusive or inclusive societal reactions. Exclusion from conventional activities and others can then cause an alteration in the identity of the actor to one or deviant or criminal.An alteration in identity leads tile actor to seek the support of deviant others. Being rewarded for deviant behavior leads to a greater probability of secondary deviance. Paternoster and Iovanni allow for tile possibility that in this process an actor may actively seek the label.
While there has been some recent theoretical development in this area, little attempt has been made to test these hypotheses. Thus, the greatest need for this approach is empirical testing. This presents a challenge to criminologists because of the difficulty in testing such a model as the one outlined by Paternoster and Iovanni (1989) such a test.
Related to specifying the process by which labeling works is tile attempt to integrate labeling with another criminological theory. Two such attempts at the integration of aspects of labeling with other theories are by Triplett (1990) and Farrell (1989).
Triplett (1990) argues that both Becker (1963) and Lemert (1967) saw labels as indirectly effecting deviant behavior through changes in the actor's reward and motivation systems. As previously discussed Becker (1963) saw labeling as only the first part in a process leading to secondary deviance. Membership in a group that supports deviant activity is another necessary step in the process. Lemert (1967) expands oil this idea when he proposes that the movement from primary to secondary deviance involves learning. Actors moving to secondary deviance learn how to exploit their new, degraded status by changing what they view as rewarding. Again, membership in a group that rewards deviant behavior promotes this learning.
Beginning then with the notion that labeling alone does not cause delinquent behavior, Triplett (1990) tests tile hypothesis that labels will only be effective if they cause a change in the actor's reward system, beliefs, motivations, and friends. She argues with Matsueda (1982) that variables such as attachments and involvement measure who we learn from and who rewards its. Triplett then tests For tile direct effects of labels on beliefs, attachment to family, proportion of delinquent friends, involvement with those friends and their effects on delinquent behavior. Using four waves of the Elliott National Youth Survey site finds that labels do have a significant impact oil tile proportion of delinquent friends one has. Tile proportion of delinquent friends then has a consistently, significant effect on delinquent behavior. The effects of labels on tile other variables was nonsignificant though in the expected direction. She concludes that while labels effect motivations to delinquent behavior by increasing social supports for delinquent behavior they do not effect beliefs or attitudes towards conventional behavior.
In 1989 Farrell presented what lie termed a "psychological elaboration of an integrated systems model." In this work Farrell centers on integrated theory which draws upon interaction, strain and social and Cultural Support theories around tile personality characteristic of ambiguity tolerance. He predicts that individuals come under a certain amount of conflict in terms of the way they arc defined by others. This creates a tension or strain for individuals that can be reduced by redefining and reorganizing their self image. The less individuals can tolerate ambiguity the more susceptible they arc to the strain and also to the stereotypical imputations of others. Their reaction may be to accept this stereotypical identity, associate with those similarly labeled and thus to increase their deviant activities.
The question surrounding these two approaches is tinder what conditions does formal action by the society (a label) deter delinquent behavior, increase delinquent behavior or have no effect. This question was raised by Thorsell and Kicnike in 1972, repeated by Tittle in 1975 and yet has received little attention. Past research either ignores the question or suggests we make a theoretical choice between labeling and deterrence. The answer, however, does not lie ill such a choice but in determining tile conditions tinder which the effects of formal criminal justice agency contact vary.
While tile first two approaches deal with specifying the process in which labeling causes deviant behavior, the third approach takes a different turn. In response to several damaging criticisms of labeling (see Tittle 1980 and Wellford 1987) and the writings of labeling theorists, researchers have begun to focus on the effects that informal, subjective labels have on self-identity and behavior. Such efforts have led us to one of the more promising developments in labeling-the rediscovery (see discussion above) of the importance of informal labels.
Informal labels are defined as
an attempt to characterized a person as a given "type" (such as "good student"or "troublemaker") by persons who are not acting as official agents of formal social control agencies, and in social situations that are not formal social control "ceremonies." (Paternoster and Triplett 1989, 6)
Informal labels then arc, for example, those which parents give when they perceive their children as "dumb" or "bad" and transmit these feelings to their children.
The emphasis on informal labels answers several damaging criticisms of labeling. Informal labels may be applied long before actors have contact with formal agencies of control and before self-concept has completely developed. It is important to note that labeling theorists themselves did not ignore the possible effects or informal labels (see for example Lemert 1951; Schur 1971; and Tannenbaum 1938). However, their effect may have been downplayed because of the historical context in which labeling theory developed (see Paternoster and Triplett 1989 for a more detailed discussion).
Another long ignored aspect of labeling is the subjective interpretation of that label (see Mead 1974 for an early statement on tile importance of subjective interpretations). Tile importance of focusing on subjective labels, the actor's perception of how others perceive them, derives from the fact that the actor's perception may differ from the objective label, which is the perception of the one assigning tile label. Thus, while it is important to know that parents label their child "criminal" it is equally important to know how the child interprets this label. If a child is labeled "good" but perceives a different label, "criminal," which one will have the effect? Testing for the effects of subjective labels takes into consideration the possibility of misinterpretation between actors.
Another development in this area is tile distinction between types of labels. Orcutt (1973) notes that informal social reactions to deviant behavior may be exclusive or inclusive in form. Inclusive reactions arc attempts by the social group to bring tile deviant "into conformity with the rules of the group without excluding him from it." Exclusive reactions arc those attempts at social control which operate to reject the rule-breaker from the group and revoke his privileges and status as an ordinary member" (260). While Orcutt was concerned mostly with the determinants of inclusive versus exclusive reactions this distinction also has importance for deviance amplification. It might be predicted that exclusive reactions and inclusive reactions will have opposite effects on the probability of future delinquent behavior.
One early test of the effects of informal labels is that of Thomas and Bishop (1984). They test the effects of self-reported Suspension or being thrown out of a classroom on delinquent involvement, self-identity, and perceptions of risk. Thomas and Bishop find no significant effect of informal labels and conclude that there is little support for labeling's predictions. However, there is a difference between informal sanctions and informal labels. Sanctions given by an informal authority, like a parent or teacher, (to not necessarily imply a label. A teacher, punishing a child for inappropriate classroom behavior, may not generalize that punishment to describe tile child's general behavior or the child.
Two recent attempts at testing labeling have tested the impact of informal, subjective labels. One is Triplett's (1990) attempt at integration discussed previously. Another attempt is Paternoster and Triplett (1989).
Paternoster and Triplett (1989) develop a theory of informal labeling that focuses on the subjective interpretation of informal labels, both inclusive and exclusive. They propose that informal labels that actors interpret as signifying that they arc delinquent result in a weakening of conventional ties while simultaneously strengthening unconventional ties. Informal labels which are interpreted as inclusive, signifying the actor is "sick" instead of "bad," however, work to reestablish the offender's ties with the conventional community.
They use the child's perception of tile parental label "delinquent- as a measure of exclusive labels and "sick" as inclusive. They then hypothesize that exclusive labels will decrease family and school involvement, school grades, conventional aspirations and beliefs while increasing the proportion of delinquent friends and tile commitment to those friends thus increasing delinquent behavior. Inclusive reactions are predicted to have the opposite effect. Paternoster and Triplett (1989) find that indeed exclusive reactions work in the predicted manner. However, their measure of inclusive reactions did not consistently work as predicted. While they increased school involvement and conventional beliefs and decreased commitment to delinquent friends as predicted, they also negatively effected conventional aspirations and school performance (grades): "Perceptions that others think of one's self as I messed up' and 'having personal problems' may constitute a more temperate form of a negative label than a perception that others view one as a 'trouble maker,' but it appears to be a derisive label nonetheless" (27).
The focus on informal, Subjective labels is perhaps the most promising area of development in labeling research. However, this preliminary work requires further careful testing and development. One particular difficulty of this approach is the measurement of informal labels. As mentioned in connection with Thomas and Bishop's (1984) study it is difficult to specify when an informal response is a label. The possible distinction between label and sanction that does not pose a problem for formal labels poses considerable difficulty when discussing informal labels.
Another difficulty that needs to be addressed is tile meaning that these labels have for both the giver and the one being labeled. As measured by Paternoster and Triplett (1989) and Triplett (1990) we know when a child perceives a negative label by their parents but not what this label means to the child. The work of Link and his associates (Link 1982, 1987; Link and Cullen 1983; Link et al. 1987, 1989) addresses the area of the expectations of the labeled person and the effects these expectations have on behavior. Their work, though focusing oil mental illness, has implications for criminology. Any discussion of advances in labeling would not be complete without a discussion of their work.
Link and his associates developed a modified labeling perspective towards mental illness based on the work of Scheff (1966). They propose that our society has formed conceptions of what it means to be a mental patient that involve tile ideas of devaluation and discrimination. These conceptions arc known by all tile members of the society but become personally relevant when they arc applied to oneself. Those labeled respond depending on their expectations of societal reactions. Their responses include secrecy, withdrawal, or education. The labeled person's response may have negative consequences for self-esteem, earning power, or ties with society. Lack of self-esteem, lower earning power and a change in one's social network can then lead to increased vulnerability to a new disorder (Link et al. 1989).
In testing this model Link et al. (1989) examine tile expectations of patients and untreated community residents, their strategies to cope with labeling or the possibility of labeling, and the effects of these strategies. They find that both samples believe mental patients face devaluation and discrimination. In terms of coping strategies, Link et al. discovered that tile people in their sample endorse secrecy, withdrawal and education. Testing the effects of these responses on social support networks they find they create a greater reliance on household members for support and lessening of reliance on nonhousehold members. This effect remains even when controlling for behavior. They conclude, "Our study strongly challenges the notion that labeling and stigma are inconsequential in tile lives of psychiatric patients" (1989, 421).
One point that Link (1982) makes is that labels have debilitating effects that go beyond the creation of deviant behavior. Awareness and exploration of these effects as well as a movement away from exploring only those factors considered to be the primary cause of criminal or delinquent behavior may be one of the many contributions of labeling to criminological theory.
The Future of Labeling Research and Theory
One of the most thoughtful contributors to and analysts of labeling theory, Edwin Lemert (especially 1972), described tile emergence and future of labeling theory in ways that if they had been followed would have prevented many or the problems that we have noted in this paper. Lemert observed that there was no necessary link between labeling and theories that saw social control as arbitrary, discriminatory, and as the sole source of deviance. He also recognized tile importance of other theoretical perspectives in the development of comprehensive theories of crime and deviance. He understood how these could complement and be integrated with labeling. Most importantly, from our perspective, Lemert recognized the significance of symbolic interactionism and the work of George Herbert Mead for labeling theory. Unfortunately many labeling theorists, researchers, and critics have not paid careful enough attention to the richness of Lcmert's analysis. The future of labeling theory as we have outlined it and as we see it developing depends very Much on its ability to in a sense "get back on track." The period of tile 1970s and early 1980s has seen little development in labeling theory as critics noted its obvious shortcomings and other perspectives became more prominent.
The future of research on labeling theory seems quite clear. Such research must be longitudinal; it should focus on the labeling process in tile family, school and community as well as by more formal labelers; it should emphasize early labeling; it must necessarily have a developmental component to provide for the ways in which a developing self may in fact react to labels and interpret them; and it must avoid the problem of treating labels as objects rather than symbols of objects. In recent years criminologists have come to understand the importance of longitudinal research (Tonry et al. 1991) and comprehensive longitudinal studies are underway in tile field of delinquency (see the 1991 Symposium issue of tile Journal of Criminal Law, and Criminology) and, more generally, for delinquency and crime (Earls and Reiss 1990). These studies demonstrate the Importance of longitudinal designs in sorting out complex casual models. We believe such efforts should be incorporated into the models organizing tile research measures of the components of labeling theory.
We have argued throughout this paper that official labels are less important than informal labels in part because formal labels occur late in the period when one's self-conception is forming. Therefore, It Is critical that future research oil labeling better understand the process of labeling as it occurs in preschool and elementary school years as the child's conception and understanding of self begins to take form. This requires a truly developmental perspective to allow for an understanding of the interaction between emerging self and the social environment and requires that tile researcher understand that the important measure of a label is the individual's perception of labels, whether formal or informal, and not the actual behavioral manifestation of tile label by an external source. Finally, this research, as it develops, will begin to allow us to focus oil the interaction between informal and formal labels as these change through the development of an individual. Clearly, this type of research has begun. The work noted earlier by Triplett (1990) represents an attempt to address the longitudinal component and to focus on informal labels. Our hope is that as we begin to understand the contribution labeling theory can make to the understanding of crime that other longitudinal studies focusing oil early development and its relationship to crime and deviance will pay greater attention to tile labeling perspective.
While the agenda for research development is clear, the future of labeling theory and its development raises a much more complex set or concerns. First, as we noted at the very beginning of this paper, criminological theories have often been presented as comprehensive explanations of crime and deviance with little recognition of the limits the theories carry with them. This certainly applies to labeling theory. Self-perceptions, self-interpretations of self and environment, the actions of those who provide reactions and labels are all in some sense limited by the characteristics of individuals and their environment. As a simple example, no matter how intense one is, no matter how one perceives oneself, no matter how tile environment reacts, it is generally impossible for someone under four feet to dunk a basketball. This simple example is offered to note that even biological characteristics may have an impact on the influence that labels and labeling can have. This is certainly the case also with other characteristics of individuals and environments. This, of course, is not to say labeling has no effect; simply, that it will have, as any single theory or theoretical perspective must have, limits on its ability to explain behavior. This brings, us to our earlier point that for labeling theory to fully develop, it must be seen as part of a larger theoretical model.
It has recently been argued that a distinctive characteristic of criminology is its commitment to an interdisciplinary of integrated theory of criminal behavior (Wellford 1989). While many have debated the value or possiblity of such theories, there is no debate about their potential value. The approach that seems to offer the best foundation for a model for interdisciplinary theory is one that provides for sources of explanation interacting together from cultural, social, psychological and biological spheres of knowledge. Labeling theory clearly represents an important element in the interaction between social and psychological levels of explanation. Still, it needs to be integrated into not only other social and psychological theories (as has been Suggested by, for example, Farrell, 1989), but also with theories that include biophysical characteristics and components representing cultural and subcultural aspects.
This approach should not lead us into a distinction that marrs the contributions of Lemert-that is that labeling theory explains secondary deviance and all other theories explain primary deviance. Such an approach has proven to be inadequate. Certainly the impact on secondary deviance can come from theories that also explain primary deviance and the basic factors of symbolic interactionism and labeling may well be important in understanding sonic forms of primary crime or deviance. It is the integrated model that would explain through the operation of the same conceptual elements, perhaps differently weighted, perhaps operating at different times, the sources of primary and secondary deviance.
Finally, advances in labeling theory must provide more information on the process by which informal and/or formal labels actually effect delinquent or criminal behavior. Most of the research that we have reviewed and most of the theoretical development has focused on whether a label, either real or perceived, has sonic relationship to subsequent behavior. The theoretical question that needs further development is how a label has an effect. Is its effect primarily on the individual or the environment; is the effect primarily in terms of its subsequent change in the behavior of the individual; is its effect primarily on the way in which the individual is ' responded to by certain segments of the environment; and so on. The models that have been developed so far leave the connection between the label and the behavior relatively undeveloped.
We consider the future of labeling theory to be quite promising. We believe that as a representation within the criminological field of a major source of social thought (symbolic interactionism) it will be an important part of our understanding of crime and delinquency. Its value is significantly enhanced by the degree to which we are able to remove it from any direct association with critical or conflict criminology. Its value is becoming more apparent as research addresses its essential symbolic elements. Its ability to provide explanations will be enhanced as it is integrated with other theories in a comprehensive model of crime and delinquency. While it is easy to criticize much of the past theory and research on labeling, it is also easy to understand its importance. It is the importance that we wish to encourage other criminologists to consider and, in so doing, to begin to develop a better understanding of crime, delinquency and their prevention and control.
1. Particularly in discussions of libeling, but also for other approaches, there is the discussion of whether it is a theory (Schur 1971) or a perspective (e.g., Gibbs 1966; Tittle 1975). We do not find this discussion very illuminating or important. Certainly almost any current explanation of crime could be more fully and formally developed. Labeling and other approaches are offered as explanations of crime and deviance. Whether they meet some definition of a theory or not is separate from the potential value of the explanation and the steps necessary to further develop and/or test the explanation.(back)
2. For those who want to more fully understand Cooley and Mead we suggest the following original works-for Cooley, Human Nature and Social Order (19(A)-, Social Organization (1962) and for Mead, Mind, Self and Society (1934). Excellent summaries and analyses of their contributions can be found in Rieff's (1962, 1964) introductions to the above noted books by Cooley and Coser's (1971) analyses of Cooley and Mead.(back)
3. Many social science disciplines have been able to define a fundamental dependent variable that is appropriate to a discipline specific theory. For example, the concept of rational economic behavior created by economists has allowed the development of simple economic theories that utilize few concepts, models, etc. from other disciplines. Criminology, however, attempts to explain criminal behavior- a set of behaviors whose definition is not controlled by the discipline (see Gottfredson and Hirschi, 1990 for a discussion of this point).(back)
Adler, Freda, G. Mueller and W. Laufer
1991 Criminology. New York: McGraw-Hill.
Ageton, Suzanne and Delbert Elliott
1974 The effect of legal processing on delinquent orientations. Social Problems 22: 87-100.
1963 Outsiders. New York: Free Press.
Cooley, Charles Horton
1962 Social Organization. New York: Schocken.
1964 Human Nature and Social Order. New York: Schocken.
Coser, Lewis A.
1971 Masters of Sociological Thought. New York: Harcourt Brace Jovanovich.
1972 Labeling theory in deviance research. Sociological Quarterly 13): 447-474.
1989 Cognitive consistency in deviance causation: A psychological elaboration of an integrated systems model. Pp. 77-92 in Steven Messner, Marvin Krohn and Allen Liska (eds.) Theoretical Integration in the Study of Deviance and Crime: Problems and Prospects. Albany: State University of New York Press.
Felton, Earls and A. Reiss
1990 Program on Human Development and Criminal Behavior. Cambridge, Mass: Harvard School of Public Health.
Gibbs, Jack P.
1966 Conceptions of deviant behavior: The old and the new. Pacific Sociological Review, 9: 9-23.
Gottfredson, Michael R. and T. Hirschi
1990 A General Theory, of Crime. Stanford, Cal.: Stanford University Press.
1975 Imprisonment and the expected value of a criminal choice: A specification and test of aspects of the labeling perspective. American Sociological Review, 40: 71-87.
1976 Race, commitment to deviance and spoiled identity. American Sociological Review 41: 432-42,
1972 Delinquency and adolescent self-conceptions: A study of the personal relevance of infraction. Social Problems 20: 84-103.
1951 Social Pathology. New York: McGraw Hill.
1967 Human Deviance, Social Problems and Social Control. Englewood Cliffs, N.J.: Prentice Hall, Inc.
1982 Mental patient Status, work and income: An examination of the effects of a psychiatric label. American Sociological Review 47: 202-15.
1987 Understanding labeling effects in the area of mental disorders: An assessment of the effects of expectations of rejection. American Sociological Review 52: 96-112.
Link, Bruce and Francis Cullen
1983 Reconsidering the social rejection of ex-mental patience: Levels of attitudinal response. American Journal of Sociology 11: 26173.
Link, Bruce, Francis Cullen, James Frank and John Wozniak
1987 File social rejection of former mental patients: Understanding why labels matter. American Sociological Review, 92: 1461-1500.
Link, Bruce, Francis Cullen, Elmer Struelning, Patrick Shrout, and Bruce Donrenwend
1989 A modified labeling theory approach to mental disorders: An empirical assessment. American Sociological Review 54: 40023.
1969 Deviance and Identify. Englewood Cliffs, N.J.: Prentice-Hall, Inc.
Mahoney, Ann R.
1974 The effect of labeling upon youths in the juvenile justice system: A review of the evidence. Law and Society Review, 8: 583-614.
1971 Societal reaction and career deviance: A critical analysis. The Sociological Quarterly 12: 204-18.
1982 Testing control theory and differential association: A causal modeling approach. American Sociological Review 47: 489-504.
Mead, George H.
1934 Mind, Self and Society. C. Morris (ed). Chicago: University of Chicago Press.
Messner, Steven F., M. D. Krohn, and A. E. Liska
1989 Theoretical Integration in the Study of Crime: Problems and Prospects. New York: SUNY Albany Press.
1973 "Societal Reaction and the Response to Deviance in Small Groups." Social Forces 52: 259-67.
1987 The deterrent effect (if the perceived certainty and severity of punishment: A review of the evidence and issues," Justice Quarterly 4: 173-218.
Paternoster, Raymond and LeeAnn Iovanni
1989 The labeling perspective and delinquency: An elaboration of the theory and assessment of the evidence. Justice Quarterly 6: 359394.
Paternoster, Raymond and Ruth Triplett
1989 The Effect of Subjective Labels on Delinquency. Paper presented at American Society of Criminology.
1962 Introduction. C. 14. Cooley. Social Organization. New York: Schocken.
1964 Introduction. C. Ff. Cooley. Human Nature and Social Order. New York: Schocken.
1966 Being Mentally Ill. Chicago: Aldine.
1974 Labeling theory of mental illness. American Sociological Review, 39:444-52.
1971 Crime and Justice: American Style. Washington, D.C.: Government Printing Office.
1971 Labeling Deviant Behavior. Englewood-Cliffs, N.J.: Prentice Hall.
Siegel, Larry J.
1986 Criminology. St. Paul: West Publishing
1938 Crime and the Community. New York: McGraw Hill.
Thomas, Charles and Donna Bishop
1984 The effect of formal and informal sanctions on delinquency: A longitudinal comparison of labelling and deterrence theories. Journal of Criminal Law and Criminology 75: 1222-46.
1971 Punishment and crime: The effect of legal dispositions on subsequent criminal behavior. Unpublished Ph.D. Dissertation, University of Pennsylvania.
Thorsell Bernard and Lloyd Klensky
"The labeling process: Reinforcement vs. deterrence." Law and Society Review 6: 393-403.
1975 Deferrents or labeling. Social Problems 22 :332-45.
1980 Labeling and crime: An empirical evaluation, Pp. 241-68 in Walter Gove (ed.) The Labeling of Deviance: Exploring a Perspective, Second Edition. Beverly Hills, Cal.: Sage Publications.
Tonry, Michael, L. E. Ohlin and D. F Farrington
1991 Human Development and Criminal Behavior. New York: Springer-Verlag.
1990 Labeling and Differential Association: The Effects on Delinquent Behavior. Unpublished Ph. D. Dissertation, University of Maryland. Wellford, Charles F.
1975 Labeling theory and criminology: An assessment. Social Problems 22: 226-41.
1987 Delinquency prevention and labeling. Pp. 257-267 Prom Children to Citizens, Volume 3 (J. Q. Wilson and G. C. Loury, eds.). New York: Springer-Verlag.
1989 Toward an integrated theory of criminal behavior in S. F. Messner et al. Theoretical Integration in the Study of Crime New York: SUNY Albany Press.
White, Leslie A.
1969 The symbol. Science and Culture (L. White) New York: Farrar, Strauss, Giroux.