THE INFORMAL SOCIAL CONTROL
The relationship between state and nonstate forms of social control has long been one of the central issues of legal sociology. Whether discussed in terms of "legal pluralism" (e.g., Griffiths 1986), "custom" (e.g., Diamond 1971), "indigenous law" (Galanter 1981), "community justice" (Henry 1985), or - the designation used here - "informal social control" (see Schwartz 1954), the similarities and differences between various types of normative regimes remains one of the enduring questions of the discipline. In recent years, much of the literature has centered around the problem of influence: how legal rules and institutions both affect and are affected by the informal order (Merry 1988: 879-886). The present article, by contrast, compares the two systems as types of social control, focusing on how different groups and individuals respond to a particular type of human behavior (see especially, Black 1976; 1989). The behavior in question is homicide, the empirical base being provided by interviews with men and women incarcerated for murder or manslaughter.
Little is presently known about the informal social control of criminal defendants. Though the interview data presented in these pages only illuminate a single category of deviance, this has the advantage of allowing law and informal social control to be compared more exactly. Concentrating on homicide cases [page 32] alone reduces the amount of variation in the defendant's behavior and thereby isolates some of the purely sociological forces at work.
Viewed in this light, the two systems of social control appear remarkably similar. Despite clear procedural differences, a common social morality appears to underlie both law and at least one type of informal social control. Though people who kill and their intimates are sometimes subject to self-help in the form of threatened or actual violence, the most frequent informal sanction for homicide is avoidance, the reduction or elimination of contact with the killer or the killer's intimates. In the cases sampled, those homicides which attract the most severe sentences also attract the most avoidance. Killings between blacks, for instance, generate less severe criminal sentences and less informal avoidance than those between whites. Moreover, cross-cultural data disclose that social control systems in other societies discriminate on similar grounds. There seems, therefore, to be a core set of variables which operates in both systems and which distributes different sanctions similarly.
The relationship between law and avoidance in these cases is one of similarity not identity, however. One significant contrast is that while killings between strangers generate more severe legal sentences they do not attract more avoidance than those between intimates. The reason for this divergence seems to be that informal social control is confined to smaller segments of social space than law. The more extensive jurisdiction of law has itself a general consequence: law subjects those too distant to be touched by the informal order to a common system of control, a fact of considerable significance in a world of intermingling strangers.
The overall picture, then, is one of law as informal social control writ large. Law operates on a wider stage, bringing larger numbers of socially distant people within the same system of authority. But in doing so, law selects out the same kinds of cases for leniency and for severity. In this way, law reproduces rather than replaces the discriminations inherent in non-legal forms of social control.
Two trends in social control
The case-level study of social control - the response to deviant behavior - has undergone two major evolutions in recent years. The first is an expansion beyond law, or state-centered social control (see Black 1984). Law continues to attract a lot of scholarly attention, but it is no longer the exclusive focus of theoretical or empirical attention. One reason for this shift is the realization that, despite the popular and scholarly attention law receives, people invoke the legal system relatively rarely. Victimization surveys, for instance, routinely reveal that less than half of all crimes are reported to the police (e.g., Shim and DeBerry 1988). A study of civil dispute processing among contemporary Americans discloses that out of every 20 cases in which $1000 or more is in issue the aggrieved party complains to the other side in only 14, contacts a lawyer in two, and files suit in one (Trubek et al. 1983: s18-20). Moreover, the ethnographic literature provides additional reasons for not over-estimating the importance of the legal system. Not only has law has been absent from the egalitarian hunter-gatherer societies in which most of human history has been played out, but even in those societies that do have law, people often prefer to handle their disputes by other means anyway (e.g., Gibbs 1963).
Building on the rich cross-cultural literature on feuding and warfare (e.g., Hasluck 1954; Koch 1974) negotiation (e.g., Gulliver 1979) mediation (e.g. Ekvall 1964), gossip (e.g., Merry 1984), witchcraft (e.g., Winans and Edgerton 1964) avoidance (e.g., Woodburn 1979) and other forms of conflict management, scholars have begun to investigate non-legal ways of handling disputes in modern society. Among the settings analyzed have been: a wealthy suburb (Baumgartner 1988), illegal markets (Reuter 1984), the higher echelons of corporate management (Morrill 1994), a city neighborhood (Thomas-Buckle and Buckle 1982), a rural ranching community (Ellickson 1991), a day-care center (Baumgartner 1992a), an urban beach (Edgerton 1979), and an egalitarian manufacturing corporation (Tucker 1992). As a result of this work, theoreticians have been able to specify the central conditions associated with a range of forms of social control (Black 1990; see also Horwitz 1990).
As interest has moved beyond legal cases, a second evolution has occurred. Initially, theoretical work tended to emphasize the social characteristics of the principal parties to disputes in explaining the outcome of cases (see especially, Black 1976). More recently, though, the role of third parties has come to the fore. Third parties may relate to cases in one of two broad ways: as partisan supporters or as nonpartisan settlement agents (Black and Baumgartner 1983). Theory now exists to predict the amount of partisan support and the authoritativeness of the settlement behavior that third parties engage in on the basis of their social characteristics and relationships to the principals (Black and [page 34] Baumgartner 1983; Black 1993: Chapters 7 and 8; see also Baumgartner 1988: 66-71; Cooney 1988: 53-60, 1992; Horwitz 1990: Chapters 8 and 9; Baumgartner 1993).
The present article lies at the intersection of these two developments. Exploring both the legal and informal social control of homicide, it draws upon the social characteristics of the principals and third parties to make two points. First, law and avoidance respond to at least some of the same social characteristics of cases. In practical terms this means that defendants who get severe sentences also tend to experience the most avoidance. Second, the overlap is not precise. Strangers receive the heaviest legal sentences but not the greatest amount of avoidance. The reason for this is that different kind of people bring legal and informal sanctions to bear against killers. Legal officials are more organized and more distant from defendants than the family members, friends, and acquaintances who avoid them. Law, consequently, has a wider social jurisdiction; it encompasses larger networks of people.
Data for the present paper come from an interview study conducted by the author with 75 people convicted of homicide in Virginia, U.S.A. and a further 42 interviews with a member of some of their families. The homicide offenders were interviewed in prison; the family members by telephone.
The defendants interviewed were randomly selected from all those admitted to the Virginia correctional system during 1988 following a conviction for murder of manslaughter (voluntary or involuntary). Because of the time lag between offense and sentence most of the homicides occurred in 1987. Thirty nine (52 [page 35] percent) of the defendants were black, and sixty three (84 percent) were male, figures almost identical to those reported by the Uniform Crime Reports for homicide arrestees nationally in 1987 (52% black; 85% male: Federal Bureau of Investigation 1988: 9).
The interviews, which typically lasted about two hours, focused on three points: (1) the circumstances and context of the homicide itself, (2) the behavior of others toward the defendant following the homicide, and (3) the legal processing of the case. Interviews were semi-structured in the sense that although the defendants were all asked the same minimum set of questions, they were also encouraged to expand and add more detail about their own individual cases. Written notes were taken but a tape-recorder was not used. The interviews were conducted in as informal a manner as the setting and occasion permitted.
Since the present paper focuses on the legal and informal response to the homicide, rather than the more controversial issue of the reasons for it, the veracity of the information provided by the defendant is not a major concern. This is, however, generally supported by information obtained from two other sources: the 50 defendant Pre-Sentence Investigation Reports that were available, and the telephone interview with a member of defendant's family.
At the conclusion of the prison interview, the defendants were asked if they would agree to having the interviewer contact a member of their family by telephone in order to learn about the impact of the case on those close to the defendant. Fifty of the 75 defendants agreed. Of the 50 people named by the defendants, it was possible to contact 40. In two instances, a second family member was interviewed, bring the total number to 42. These interviews, also semi-structured, were conducted by telephone. The person interviewed was typically a female relative, of whom the most common was the defendant's mother. Given the greater degree of social distance involved, it is not surprising that these interviews exhibited greater variation in quality than the prison interviews. Although only one person refused to be interviewed, a number were guarded in their responses. On the other hand, some of those contacted were extremely informative, discussing the case and its impact on their lives and those close to them in considerable detail. The variable quality of these interviews is reflected in their variable length: the shortest was 15 minutes and the longest 2.5 hours.
The telephone interview tended to confirm and, sometimes, to expand upon the information about the informal social control experienced by the defendant. Even so, it is important to note that this study does not capture the full range of social control generated by homicide. First, the study is confined to cases in which somebody is convicted for a killing. Those cases in which an offender escapes [page 36] conviction ‑ probably about 50 percent of homicides known to the police ‑ are not included in the study. Second, the study is confined to the three-year period following the killing when the person convicted is still in prison. It does not, therefore, cover the post-release phase. Third, no data were obtained from the victim's relatives who may have engaged in social control of which the defendant and his or her relatives were unaware (e.g., the dissemination of gossip).
The informal order
The defendants in the sample were serving sentences ranging from five years imprisonment to the death penalty. The modal sentence was 20 years with life imprisonment being the second most frequent sentence. In Virginia, as in most other states, prisoners do not serve the full amount of their prison terms. The precise length a person actually spends behind bars is determined by a number of factors, such as the number of previous prison terms served, the nature of the offense, and the prisoner's conduct. The ultimate decision to release is usually made by a parole board whose decisions have not been scientifically analyzed. But prison officials say that a first-time prisoner sentenced to 20 years, for [page 37] example, could reasonably expect to serve no more than eight years (see further Cooney 1991: 158-162).
The informal sanctions typically levied against those who kill are less well-known. As the following discussion indicates, these fall into two major categories: avoidance and self-help.
Avoidance - the reduction or elimination of contact following a conflict - is common in settings in which social ties are fluid and easily severed (Baumgartner 1988: 60-66; Black 1990: 50-51). Thus, avoidance has been found to be frequent among, for example, hunter-gatherers (see, e.g., Woodburn 1979), shifting horticulturalists (see, e.g., Stauder 1972), gypsies (Gronfors 1986: 119-121), travellers on the overland trail in 19th century America (Reid 1980), and modern suburbanites (Baumgartner 1988).
To this list should be added those in prison. Prisoners are in a unique relational environment, one in which their ability to maintain their social ties is severely limited. Their ties are not so much loose as coercively distant, but the effect is the same: prisoners are easily avoided. Forty-five of the 75 defendants (60 percent) in the present sample experienced some avoidance, making it the most common informal sanction for homicide in the sample.
Avoidance of the defendant can occur at different times: before defendant is arrested, while defendant is out on bond, while defendant is incarcerated at the local jail, or after defendant is sent to prison. Note that although people who kill and are incarcerated typically experience a sharp contraction in their social relationships, not all of this can be considered `avoidance.' Some ties attenuate simply because the prisoner is no longer in circulation. By contrast, avoidance is a normative phenomenon, involving some expression of disapproval of the killing. In order to distinguish avoidance from mere relational mortality, the defendants in the sample were asked who had reduced contact with them following the incident and what they felt the reason for the reduction was. Only when the reason was related to the homicide would the behavior be considered avoidance.
Case 76 provides an example. The defendant is a 19 year old woman convicted of killing her five month old son:
My boyfriend left me three days before all this happened. Except for the trial, I haven't seen him since. I called him from the jail, but he didn't want to talk to me. Then I wrote to [page 38] him a few times, but he never wrote back. He's definitely mad at me; I know that from his Mom. It was his son I killed. He hasn't forgiven me, although I asked him to in my letters.
A lot of my street friends came and visited me in jail. But they were just being nosy. After they'd asked their questions they never came back again.
In this case, the defendant's boyfriend's behavior would be considered avoidance, but that of her friends would not, since it is not clear that their lack of contact implied condemnation of her actions. Like all subjective procedures, this measure is only as reliable as the perceptions on which it is based. This limitation should be kept in mind.
When measured by the defendant's awareness of it, avoidance is relatively common. Of the 45 defendants in the sample who experienced some diminution of social contact as a result of being convicted of the homicide, most (24) were avoided by one or two people, but as many as nine were avoided by more than ten people.
Avoidance is not limited to the defendant. His or her intimates may also be avoided. A study of prisoners' wives found that people will sometimes refuse to have any dealings with them: acquaintances may withdraw, landlords decline to rent to them, and shopkeepers refuse to serve them (Fishman 1990: 113-126). Defendants' intimates interviewed for the present study reported similar experiences. Generally, though, it was only those with a close tie to the victim who made it plain that they wanted to have nothing more to do with an individual closely connected to a murderer. In Case 70, for instance, the defendant's mother said:
They [the victim's family] stopped speaking to us. I was hurt. We'd been friends for years. They'd moved away before this happened. But we'd see them when they came back to visit. I don't know if they've been back recently. But I know they're not talking to us because they wouldn't do so at the trial. When we looked at them or tried to talk, they turned away.
The sample also contains a small number of examples of more generalized avoidance. In Case 35, for example, the killing took place in a small town. The defendant's sister said:
A few people from high school I hadn't met for a while treated me differently. When I bumped into them on the street or at the high school reunion they stared at me. They would have absolutely nothing to do with me. Also, at my sister-in-law's wedding I approached a few people and said "Hi". They said nothing; just turned away.
This surprised me. I hadn't known these people that well although they had been at our house when we were kids. I was in the school band with some of them. It hurt.
All told, 23 of the 75 defendants' intimates were the objects of some form of avoidance. This constitutes less than one-third of the sample. But it also represents more than half of the 40 families interviewed.
In many societies, those who kill or their family members are subject to aggression or `self-help'. Self-help is the typical way of handling homicides outside family groups in many societies that are either formally or effectively stateless (e.g., Thoden van Velzen and van Wetering 1960; Otterbein and Otterbein 1965). But self-help is sometimes found in state societies as well, as the legal and informal order vie for control over the conflict (e.g., Gronfors 1986; Ginat 1987). The mere existence of law does not preclude recourse to self-help (see Senechal 1992: 13).
In the present sample, defendants and their family members were sometimes subject to self-help. However, this took the form of threats of violence rather than violence itself. In twenty-two out of 69 cases (i.e. almost one-third) in which information was obtained, the defendant or a member of the defendant's family was threatened with violence following the homicide. These threats were made at various stages: while the defendant was suspected but not charged, during the period he or she was in jail, or after he or she was released from custody on bail.
Threats occur under two scenarios. In the more common, one or more of the victim's close relatives threaten the defendant or a member of his family. Here threats represent an inchoate form of private vengeance, or feuding. Threats made by the victim's kin appear to be most common among young men resident in lower-income and minority communities. In Case 75, for instance, the defendant, a 24 year old black man had, together with a friend (`George'), [page 40] hunted down and shot the ringleader of a group of men who had gotten the better of them in a public fight earlier that day. The victim died a few days later. Defendant described what happened next:
The day he died two car loads of guys rolled by my Mom's house looking for me. My sister answered the door. She told the guy I was not at home. He pulled back his coat and showed her a big gun. He told her he was [the victim's] brother and she was to tell me he was looking for me. The following night they came around again. When George and I heard about it, we got our hardware [i.e., guns] and went to their neighborhood looking for them. We did this a few times but we never found them.
This category of self-help appears to depend largely on the existence of strong ties of partisan support. This variable can also itself determine the outcome of a prosecution. For example, partisan support may permit the defence to shape the evidence presented to legal officials (Cooney 1992).
In the second scenario, the source of the threats is not the victim's kin but a self-appointed representative of the community. Threats of this kind are found in settings with dense social ties, such as low-income neighborhoods and rural communities. Unlike the first category, they tend to be made in cases that attract severe sentences. These threats seem to express a collective anger, to represent an inchoate form of vigilantism. Often, they are anonymous. Seven defendants in the present sample were the objects of anonymous phone calls or letters conveying insults or ill wishes (e.g. "I hope they fry [i.e., electrocute] you"). But communal threats are not invariably anonymous and, especially when the probability of retaliation is low, the person making the communal threats may be known to the person threatened. This was the situation in Case 32. The defendant and three other young men went to rob a local store. When they arrived, the defendant and another man lured the store-keeper into a back room. The defendant held the store-keeper while his companion stabbed him repeatedly. The four men were arrested for murder a few days later. The defendant's mother describes what happened next:
Rick was arrested on the 2nd. We moved out - we had to move out - on the 14th. Teenagers were vandalizing the house, throwing things through the window. When we went out on the street they made remarks. We were threatened a few times. I was told "you're time is coming" - that kind of thing. Nobody would have anything to do with us. It was terrible. I have two daughters who were 19 and 22 at the time. I was afraid for them and for myself. We called the police a few times but they couldn't do much. So I decided we had to leave. Nobody knows where we are now and everything is much better.
This case represents the greatest amount of informal social control found in the sample. As such, it serves to illustrate a central feature of the informal response to homicide: its overall mildness. While the defendant's family were made to feel unwelcome and to fear for their safety, neither they nor anybody else in the sample suffered any injury. This is not to say that people do not exact vengeance in modern settings. Conflict between gangs and groups involved in organized crime may involve protracted, feud-like vengeance (see, e.g., Bing 1991). But, despite the attention these receive in the mass media, they appear to be relatively uncommon. If the present data are representative, in the typical homicide case the violence never escalates beyond the verbal level.
The relative mildness of the informal response is consistent with the thesis that legal and informal sanctions tend to vary inversely (Schwartz 1954; Black 1976: 107). Thus, the lengthy prison sentences and death penalties imposed by the courts render severe informal sanctions superfluous. But, as the next section shows, those homicides which attract the most severe sentences also attract the most avoidance, suggesting that the relationship of law and informal social control may require further specification.
A common morality
Law and avoidance
Avoidance, then, is the most common type of informal social control found in these homicide cases. At first glance, avoidance appears to be very different from law. Law is founded on the organizational capacity of the state, avoidance is a product of decentralized networks whose members act according to shared norms but not in concert. Legal sanctions are based on published and publicized rules enforced by specialized institutions and personnel (Griffiths 1984). Avoidance is not governed by authoritatively formulated norms and is carried out by ordinary individuals acting in the course of their everyday affairs. Legal sanctions for homicide mean the loss of liberty or even life; avoidance results merely in the loss of contact. Nothing inherent in the nature of these sanctions would lead one to expect their imposition to exhibit the same pattern. But it does. Table 1 crosstabulates the number of people avoiding the defendant by the severity of the defendant's sentence (divided into two categories: terms of imprisonment ranging from five to 70 years and sentences of life imprisonment and death) in cases with a single victim.
|Table 1. Severity of sentence by number of people avoiding defendant (in percentages)|
|Number of avoiders||Severity of sentence|
The results clearly indicate that the legal and primary informal sanction for homicide vary directly. It must be said, however, that these results are not wholly unexpected. Black (1976: 10) suggests that characteristics of the parties which increase the severity of law also tend to increase the severity of the informal social control found in that particular setting, and, as the remainder of the present section demonstrates, the data bear this suggestion out.
A variable that consistently appears in the research literature as a predictor of the amount of law applied to a case is the social status of the parties (see Black, 1976:16-29; Baumgartner, 1992b: 142-148). Status has also been used to explain variation in the severity of informal social control (e.g., Morrill 1989; Black 1990; 1993, Chapter 8; Tucker 1992). In some societies, for instance, the killing of a high status individual is less likely to be compensated and more likely to be avenged than the killing of another person (see, e.g., Barth 1959: 85; Bourdieu 1966: 216 cited in Black 1987: 569). In others, high status victims command more compensation than people of less elevated rank (see, e.g., Ekvall 1968: 77; Barton 1919: 60; Lewis 1961: 164).
In the present data, the only reliable indicator of status for both parties is
the indirect one of race. Race has, however, been shown to be of considerable
Table 2. Severity of sentence by race of defendant and victim(in percentages)
|Sentence||Race of defendant-victim|
Table 3. Number of people avoiding defendant by race of defen-dant and victim (in percentages)
|Number of avoiders||Race of Defendant-Victim|
importance in determining outcomes in the contemporary United States, at the arrest (Smith, Visher, and Davidson 1984), prosecutorial (see e.g., Jacoby and Paternoster 1982), conviction (Bernard 1979), and sentencing (see, e.g., Bowers and Pierce 1980) stages of the criminal justice process. Contrary to a popularly held belief, black defendants are not necessarily treated more harshly than their white counterparts. The race of the victim must also be considered. Generally, blacks who offend against whites are treated most severely, followed, in order, by whites who offend other whites, blacks who offend other blacks, and whites who offend blacks (Black 1976: 28-29). There are not enough inter-racial cases in the present sample to permit a complete analysis of these effects, but a comparison of the 60 intra-racial cases in which a single victim was killed reveals the expected pattern (Table 2). Dividing sentence severity into two categories - prison sentences of five to 70 years, and sentences of life imprisonment or capital punishment - reveals that white-on-white killings are ten times more likely to attract the most severe category of sentence than black-on-black killings.
Avoidance exhibits the same pattern. There are 56 cases in which a single victim was killed for which there is information about avoidance. As Table 3 demonstrates, white-white cases generate more extensive avoidance.
Since the vast majority of the people avoiding the defendants are of the same race as the defendants, the results reported in Tables 2 and 3 are a testament to the remarkable power of social stratification. Not only does law treat the killing of blacks as less serious than the killing of whites, so do blacks themsleves. But social status is not the only variable that operates in this way.
Another characteristic of people which helps to seal their fate in the criminal justice system is their respectability, or the amount of social control to which they have been subject in the past (Black 1976: 111). A victim with a criminal record, for instance, is likely to attract less support from the law than one with an unblemished record. As one recent review of the literature puts it, "officials at every stage of the criminal process proceed more vigorously against those [page 45] who offend respectable victims than against those accused by the morally compromised" (Baumgartner 1993b: 137). Thus, in a case from Georgia known to the author, a prisoner wanted to bring charges of assault against another prisoner. The case appeared to be strong, since the victim had sustained relatively severe injuries and had a number of eyewitnesses. But when the Assistant District Attorney discovered that the victim had a number of prior convictions for aggravated assault and armed robbery he immediately dismissed the case (author's field notes). What is true of victims is also true of offenders. A defendant with a prior criminal record can expect to be pursued more vigorously by the police (e.g., Werthman and Piliavin 1967: 72-73), be prosecuted more fully (e.g., Myers and Hagan 1979), and receive less leniency from judge and jury than a first offender (e.g., Feeley 1979: 132, 142). In the present sample, a person with an arrest record was twice as likely to receive a sentence of life imprisonment or death as one without a history of arrest.
Respectability is also critical to informal social control. In many societies, killers who have no record of deviance will be protected by their kin but those who do will be abandoned or even killed by their intimates (e.g., Warner 1958: 162; Meggitt 1977: 79-80; Ginat 1987: 90-99). For instance, persistent deviants among the !Kung San of the Kalahari desert, especially those who have taken the lives of others, are sometimes executed without any intervention by their kin (Lee 1979: 392-395). Victims who have no history of deviant behavior will be avenged but those who do (e.g., witches) will not (e.g., Knauft 1985: Chapter 5). Albanian highlanders are quick to avenge the killing of their kinfolk, but make an exception when the victim was suprised in adultery (Hasluck 1954: 212).
Avoidance of modern killers also appears to increase with their unrespectability. This is difficult to demonstrate quantitatively with the present data, however, since there is a complicating factor for which it was impossible to obtain systematic information: the respectability of the avoiding third party. Individuals who themselves have a criminal record would appear to be less likely to avoid a person accused of homicide than those who have no record. However this may [page 46] be, examination of individual cases tends to support the idea that the parties' own unrespectability affects the probability of his or her being avoided.
Take the following two cases, for instance. In Case 3, the defendant had a long series of arrests for misdemeanors as a juvenile and had served four years in prison for armed robbery. When he was accused of homicide, he was avoided by all his relatives and friends. Only his nuclear family members continued to have contact with him, and only they believed him when he protested that he had been falsely convicted. Prior violations of informal norms can have similar effects. In Case 36, the defendant was convicted of hiring her co-defendant's brother to kill her co-defendant's boyfriend who had been violent toward both of their children. The defendant had always had a fast lifestyle. She gave birth to a child when she was only 15; she frequently disappeared for weeks on end without telling anybody where she was, leaving her child with her mother; and she was a regular cocaine user. Her aunts strongly disapproved of her behavior, and when she was charged and convicted of the killing they declined to have anything more to do with her.
Two other cases illustrate the effect of victim respectability on avoidance. In Case 46, the victim was a widow in her 80s with a good reputation in the community. After the defendant ‑ a 26 year old man ‑ was charged with her murder, his friends ceased all contact with him. In Case 41, by contrast, people sought the defendant out. The defendant had been married to her husband, the victim, for a short time. They lived in a small town. After the marriage, the victim had given up his job and lived off his wife's income. When the victim began to run up bills at the local liquor store, his wife barred him from her house. The victim retaliated by slashing the tires of her car, breaking her bedroom window, and throwing rocks at her house while he shouted insulting and threatening remarks at her and her son. This lasted for three nights. The couple then had an argument during which the defendant shot the victim. The defendant did not experience any avoidance. Indeed, she was sent letters of support from quite a number of people, especially women. In addition, she received gifts of flowers from her church and money from her mother's church.
Social integration, the degree to which people participate in social life, is another variable that helps to explain the amount of legal sanction an offender attract (Black 1976: 48-49). Individuals with extensive social ties are more integrated than those with a restricted set of contacts. A lack of integration (marginality) tends to attract social control, legal and otherwise. For instance, during the European witch-hunts of the 15th and 16th centuries, marginal people, such as elderly widows, were disproportionately named and punished as witches (see, [page 47] e.g., Levack 1987: 124-134). Today, holding the offense committed constant, unemployed defendants are more likely to be incarcerated before and after trial than employed defendants (Chiricos 1991). An analysis of homicides committed in the course of another crime in a large northern industrial city during 1972, reported that the strongest predictor in a multiple regression analysis of whether the defendant would be prosecuted was the employment status of the parties. Unemployment on the part of the defendant increased the probability that the case would be prosecuted while unemployment on the part of the victim reduced it (Boris 1979: 152-153). A California study found that defendants convicted of first degree murder who were unemployed at the time of the killing or had an unstable employment history faced an elevated risk of being sentenced to death (Baldus et al. 1980: 27-28; see also Judson et al. 1969: 1337).
In America today one category of highly integrated individuals are the owners and operators of small stores. Although only four of the victims in the sample were store-keepers, it is striking that those convicted of killing them experienced a lot of social control, both formal and informal. All four defendants received sentences of life imprisonment. In three of these cases there was also extensive avoidance of the defendant. In Case 65, for instance, the defendant was a high school student with a wide circle of friends. They all avoided him. In addition, his family found their social ties sharply curtailed. His grandparents had been members of a church all their lives and had held positions of responsibility in it. They were made to feel so uncomfortable that they eventually stopped attending for a while. Similarly, in Case 32, the defendant killed a neighborhood store-keeper. He was partially avoided by his siblings and grandmother and totally avoided by his father (who lived elsewhere) and by his friends. His family received even harsher treatment, being forced, as described earlier, to move out of the neighborhood to escape the threats and insults of the local teenagers.
Were the present sample larger, it might be possible to isolate further variables associated with both law and avoidance. Even with this small number of cases, however, it seems clear that the inverse relationship between law and other social control does not preclude law and avoidance responding to similar characteristics of victims and offenders. Thus, cases involving white and integrated victims and those with unrespectable defendants seem to arouse more avoidance in the network of the defendant and evoke longer sentences from the criminal justice system.
These findings have wider implications. People commenting on the results of empirical research on law are often critical of the way the legal system works in practice. The fact that offenses against high status victims attract heavier penalties, for example, has seemed to many to be a kind of unjustifiable discrimination. But the more scholars learn about informal social control, the [page 48] clearer it becomes that such discrimination is not peculiar to law (see also, e.g., Baumgartner 1984; Morrill 1989). As these cases show, the sanctions of ordinary citizens are similarly attentive to who the parties are. Both law and avoidance respond to at least some of the same characteristics of deviants and victims. Both can therefore be seen as manifestations of a single underlying morality that treats some cases as more serious than others.
Sharing a common morality does not mean that there is perfect overlap between law and avoidance, a point that emerges from an analysis of the effects of relational distance on both forms of social control.
Relational distance refers to the strength of the prior relationship between the parties. Black (1976: 41) proposes that the relationship between law and relational distance is curvilinear. Thus, law rarely intrudes when parties are intimate or when they live in different worlds; between these extremes, the amount of legal authority increases as the strength of the relationship between the parties weakens. A considerable body of evidence, contemporary, historical, and cross-cultural, supports this proposition (Black 1976: 42-46; Baumgartner 1992: 131-136). The evidence includes homicide cases. Lundsgaarde (1977) analyzed all 268 homicides reported to the Houston police in 1969 and found relational distance to have a strong effect on the amount of legal authority imposed on the killer. "Thus, 61 percent of killers of relatives escaped any form of legal penalty; 53 percent of killers of friends or associates similarly escaped any form of legal penalty, and 36 percent of killers of strangers escaped legal punishment" (1977: 16). Among those who were punished, the average prison sentence given to people who killed intimates was 8 years; for those convicted of killing acquaintances it was 10 years; and for those who killed strangers it was 28 years. In addition, five cases resulted in a death sentence: all were homicides of strangers (Lundsgaarde 1977: 16, 232).
The tendency for the death penalty to be reserved for killings of strangers is confirmed by other studies. Gross and Mauro (1989) analyzed all homicides reported to the F.B.I. over a five year period (1976 through 1980) from Georgia, Florida, and Illinois (over 8,000 cases in all). They found that "those who killed strangers were far more likely to be sentenced to death than those who killed family members, friends, or acquaintances: ten times as likely in Georgia, four times as likely in Florida, and over six times as likely in Illinois" (1989: 47-48). An analysis of Georgia death penalty cases found that after controlling for some 40 variables relating to the characteristics of the parties and the crime killing a stranger increased the defendant's odds of being sentenced to death by a factor of 2.8 (Baldus, Woodworth, and Pulaski 1990: 319-320).
Although the present sample is confined to those defendants actually imprisoned for homicide, the same pattern can be discerned. Table 4 reports the sentences given cross-classified by the relational distance between the parties when a single victim was killed. Sentences are once again divided into two categories: terms of imprisonment ranging from five to 70 years, and sentences of life imprisonment and capital sentences. Relational distance is divided into three categories: where the parties are intimates (i.e., family members, spouses, lovers), acquaintances (i.e., friends, neighbors, familiar faces) or strangers (i.e., people previously unknown to each other).
Table 4. Severity of sentence by relational distance of defendantand victim (in percentages)
|Sentence||Relational distance of defendant and victim|
Consistent with earlier studies, the results indicate that killers of intimates are treated more leniently than killers of acquaintances or strangers. None of the 20 people convicted of killing an intimate received the most severe category of sentence. But one out of every four people convicted of killing an acquaintance and one out of three convicted of killing a stranger did.
Intimacy also reduces the severity of informal social control. In many societies homicides between strangers are avenged while those between intimates merely give rise to avoidance, or even toleration (e.g., Oberg 1934: 148; Peters 1967: 263-265; Kiefer 1972: 74-75). When a person kills an intimate among the Lugbara of Uganda, for instance, his kin "fear and avoid him as unnatural" but [page 50] they do not kill him as they would a stranger (Middleton 1965: 51). Similarly, on the Micronesian island of Yap fratricide is not avenged: people believe that it "would compound the evil should the surviving brother punish the offending brother" (Schneider 1957: 794).
Given this background, it is something of a surprise that the present data indicate that avoidance does not increase with the relational distance of the parties. People who kill strangers do not experience more avoidance than those who take the lives of their intimates. If anything, they experience less, as Table 5 indicates:
Table 5. Number of people avoiding defendant by relational dis-tance of defendant and victim (in percentages)
|Number of avoiders||Relational distance of defendant and victim|
Thus, while the relationship between the law and relational distance is strong, relational distance has only a weak impact on third-party avoidance. Relational distance therefore forms an exception to the conclusion reached so far, that law and avoidance vary directly with each other. To explain this it is necessary to examine the social characteristics of those who enforce both sets of sanctions. The following section addresses this topic.
Assuming that avoidance, like other types of informal social control, increases with the relational distance between the principals (Black 1993: 145-149), the [page 51] most likely explanation of the results in Table 5 lies in the different jurisdiction - the ability to sanction deviance - of law and informal social control.
Social control generally requires that the party exercising it has access to the deviant. Avoidance, for instance, presupposes that there is an existing or potential relationship to reduce or eliminate. Similarly, self-help depends upon the aggressor's ability to communicate or interact, at least indirectly, with the other party. As a result, informal social control tends to be highly restricted in social space. When people are separated by a lot of relational distance - when they live in different parts of the country, for instance - offenses between them rarely generate retaliatory action. Nor do they produce much third-party social control: when the principal parties are distant from each other, their third party supporters will probably also be. Hence, as Table 5 shows, third-party avoidance is not likely to be found when the principals are relationally distant.
Law differs from informal social control in not being limited to personal networks. Even when there are no third parties with ties to the defendant law may still be found. Law rests on an organizational network which connects all four corners of the nation. This network significantly extends the jurisdiction of law. Through it, people never or no longer connected by personal ties can still have access to one another. A person who kills a social isolate or who flees following a homicide still runs a significant risk of being sanctioned. By substituting a bureaucratic for a personal network the state extends its jurisdiction across a much broader area of relational space.
Ultimately, then, both law and informal social control appear to have a curvilinear relationship with relational distance. Both forms tend to increase with relational distance. However, after a point, further increases in relational distance bring a decline in the frequency and severity of law and informal social control. The difference is that the turning point is wider for law than for informal social control. Law's organizational network gives it wider jurisdiction than the personal networks of the informal system.
Figure 1 illustrates the curvilinear relationship of both law and informal social control and the wider jurisdiction of the former.
|Figure 1. Relationship between relational distance and quantity of two types of social control|
Describing the legal and informal control of homicide in one state in the United States, this paper makes three principal points. First, the state subjects those who kill to a lot more social control than society does. The law imposes lengthy prison sentences but the informal order mainly imposes avoidance. Some killers and their families have their lives threatened but actual violence against them is relatively rare. Second, despite their procedural differences, there is a [page 53] fundamental similarity between law and avoidance: they respond to a number of the same sociological variables. Thus, the race, respectability, and social integration of one or both of the parties explain variation in the amount of law and avoidance that killings attract. Third, an important difference between law and informal social control is that the latter is considerably more local in character than the former. The greater relational distance between those who enforce legal as compared to informal sanctions distributes law more extensively in social space.
Taken together, the three points outlined yield a tentative but specific view of the relationship between law and avoidance. Responding to similar characteristics of offenders and victims, law and avoidance are found in different regions of relational space. Avoidance is enforced by people who know each other; law is enforced by strangers against parties who are themselves more often strangers.
How far these findings extend beyond avoidance in homicide cases can only be determined empirically. Avoidance may or may not be typical of all informal social control (although, most likely, it is both). Even if it is not, the fact that it varies positively, rather than negatively, with law suggests that the overall relationship between the two systems of social control is complex.
Perhaps the most significant implication of the present work is that law and informal social control sanction and protect at least some of the same kinds of individuals. This suggests that notions of the inherent moral superiority of law are suspect. If equality of treatment is a criterion for judging the morality of normative regimes, there may be little to choose between law and informal social control. From the perspective of this paper, the distinctiveness of law lies in its wider jurisdiction. Informal social control has a limited ability to exert authority over the wrongdoing of strangers. Law rectifies this, allowing strangers to pursue grievances against those who victimize them. Law's large organizational base provides it with a degree of coordinating capacity that the personal networks of the informal system lack. The resulting ability to provide a common morality for a civilization built upon the interaction of strangers helps to explain why despite the many criticisms of lawyers and the system they serve, the expansion of law continues unabated throughout the modern world. But that expansion reproduces some of the principal characteristics of the system it replaces. Law's empire, with its panoply of rules, institutions, and personnel, reproduces rather than replaces the inequality found within the domain of informal social control.
BALDUS, David C., et al.
1980 `Identifying comparatively excessive sentences of death: a quantitative approach.' Stanford Law Review 33: 1-174.
BALDUS, David C., George WOODWORTH, and Charles A. PULASKI Jr.
1990 Equal Justice and the Death Penalty: A Legal and Empirical Analysis. Boston: Northeastern University Press.
1959 Political Leadership among Swat Pathans. London: Athlone Press.
BARTON, R. F.
1919 Ifugao Law. Berkeley: University of California Press, 1969.
1988 The Moral Order of a Suburb. New York: Oxford University Press.
1992a `War and peace in early childhood.' Pp. 1-38 in James Yucker, ed., Virginia Review of Sociology, Volume 1: Law and Conflict Management. Greenwich: JAI Press.
1992b `The myth of discretion.' Pages 129-162 in Keith Hawkins, ed., The Uses and Misuses of Discretion. London: Oxford University Press.
1993 `Violent networks: the origins and management of domestic conflict.' Pages 209-231 in Richard B. Felson and James Tedeschi, eds., Violence and Aggression: The Social Interactionist Approach. Washington D.C.: American Psychological Association.
1979 `Interaction between the race of the defendant and that of jurors in determining verdicts.' Law and Psychology Review 5: 103-11.
1991 Do or Die. New York: Harper Perennial.
1976 The Behavior of Law. New York: Academic Press.
1984 `Social control as a dependent variable.' Pp. 1-36 in Donald Black, ed., Toward A General Theory of Social Control, Volume 1: Fundamentals. Orlando: Academic Press.
1987 `Compensation and the social structure of misfortune.' Law and Society Review 21: 563-584.
1989 Sociological Justice. New York: Oxford University Press.
1990 `The elementary forms of conflict management.' Pp. 43-69 in School of Justice Studies, Arizona State University, eds., New Directions in the Study of Justice, Law, and Social Control. New York: Plenum.
1993 The Social Structure of Right and Wrong. San Diego: Academic Press.
BLACK, Donald and M.P. BAUMGARTNER
1983 `Toward a theory of the third party.' Pp. 84-114 in Keith O. Boyum and Lynn Mather, eds., Empirical Theories about Courts. New York: Longman.
BORIS, Steven Barnet
1979 `Stereotypes and dispositions for criminal homicide.' Criminology 17: 139-158.
BOSHIER, Roger and Derek JOHNSON
1974 `Does conviction affect employment opportunities?' British Journal of Criminology 14: 264-268.
1966 `The sentiment of honour in Kabyle society.' Pages 193-241 in J.G. Peristiany, ed., Honour and Shame: The Values of Mediterranean Society. Chicago: University of Chicago Press.
BOWERS, William J. and Glenn L. PIERCE
1980 `Arbitrariness and discrimination under post-Furman capital statutes.' Crime and Delinquency 26: 563-635.
BUIKHUISEN, Wouter and Fokke P.H. DIJKSTERHUIS
1971 `Delinquency and stigmatisation.' British Journal of Criminology 11: 185-187.
1991 `Unemployment and punishment: an empirical assessment.' Criminology 29: 701-724.
1988 The Social Control of Homicide: A Cross-Cultural Study. Unpublished S.J.D. dissertation, Harvard Law School.
1991 Law, Morality, and Conscience: The Social Control of Homicide in Modern America. Unpublished doctoral dissertation, Department of Sociology, University of Virginia.
1992 `Evidence as partisanship.' Paper presented at the Law and Society Asssociation annual meeting, Philadelphia, Pennsylvania.
COSTELLO, John L.
1991 Virginia Criminal Law and Procedure. Charlottesville: Michie.
DAWSON, John M. and Barbara BOLAND
1993 Murder in Large Urban Counties, 1988. Washington D.C.: United States Department of Justice.
1971 `The rule of law versus the order of custom.' Social Research 38: 42-72.
EDGERTON, Robert B
1979 Alone Together: Social Order on an Urban Beach. Berkeley: University of California Press.
EKVALL, Robert B.
1964 `Peace and war among the Tibetan nomads.' American Anthropologist 66: 1119-1148.
1968 Fields on the Hoof: Nexus of Tibetan Nomadic Pastoralism. New York: Holt, Rinehart, and Winston.
ELLICKSON, Robert C.
1991 Order without Law: How Neighbors Settle Disputes. Cambridge: Harvard University Press.
ERICSON, Richard V.
1977 `Social distance and reaction to criminality.' British Journal of Criminology 17: 16-29.
FEDERAL BUREAU OF INVESTIGATION
1988 Uniform Crime Reports. Washington D.C.: U.S. Department of Justice.
FEELEY, Malcolm M.
1979 The Process is the Punishment: Handling Cases in a Lower Criminal Court. New York: Russell Sage Foundation.
FISHMAN, Laura T.
1990 Women at the Wall: A Study of Prisoners Wives Doing Time on the Outside. Albany: State University of New York Press.
1981 `Justice in many rooms: courts, private ordering, and indigenous law.' Journal of Legal Pluralism 19: 1-47.
GIBBS, James L. Jr.
1963 `The Kpelle moot: a therapeutic model for the informal settlement of disputes.' Africa 33: 1-10.
1987 Blood Disputes among Bedouin and Rural Arabs in Israel: Revenge, Mediation, Outcasting, and Family Honor. London: University of Pittsburgh Press.
1984 `The division of labor in social control.' Pages 37 -70 in Donald Black, ed., Toward a General Theory of Social Control, Volume 1: Fundamentals. Orlando: Academic Press.
1986 `What Is Legal Pluralism?' Journal Of Legal Pluralism 24: 1-55.
1986 `Social control and law in the Finnish gypsy community.' Journal of Legal Pluralism 24: 101-25.
GROSS, Samuel R. and Robert MAURO
1989 Death and Discrimination: Racial Disparities in Capital Sentencing. Boston: Northeastern University Press.
1979 Disputes and Negotiations: A Cross-Cultural Perspective. New York: Academic Press.
1954 The Unwritten Law in Albania. Cambridge: Cambridge University Press.
1985 `Community justice, capitalist society, and human agency: the dialectics of collective law in the cooperative.' Law and Society Review 19: 303-327.
HORWITZ, Allan V.
1990 The Logic of Social Control. New York: Plenum.
JACOBY, Joseph E. and Raymond PATERNOSTER
1982 `Sentencing disparity and jury packing: further challenges to the death penalty.' Journal of Criminal Law and Criminology 73: 379-387.
JUDSON, Charles, et al.
1969 `A study of the California penalty jury in first-degree murder cases.' Stanford Law Review 21: 1297-1497.
KIEFER, Thomas M.
1972 The Tausug: Violence and Law in a Philippine Moslem Society. New York: Holt, Rinehart, and Winston.
KNAUFT, Bruce M.
1985 Good Company: Sorcery and Social Action in a Lowland New Guinea Society. Berkeley: University of California Press.
1974 War and Peace in Jalémó: The Management of Conflict in Highland New Guinea. Cambridge: Harvard University Press.
LEE, Richard Borshay
1979 The !Kung San: Men, Women, and Work in a Foraging Society. Cambridge: Cambridge University Press.
LEVACK, Brian P.
1987 The Witch-Hunt in Early Modern Europe. London: Longman.
1961 A Pastoral Democracy: A Study of Pastoralism and Politics among the Northern Somali of the Horn of Africa. London: Oxford University Press.
LUNDSGAARDE, Henry P.
1977 Murder in Space City: A Cultural Analysis of Houston Homicide Patterns. New York: Oxford University Press.
1977 Blood is their Argument: Warfare among the Mae Enga Tribesmen of the New Guinea Highlands. Palo Alto: Mayfield.
MERRY, Sally Engle
1984 `Rethinking gossip and scandal.' Pages 271-302 in Donald Black, ed., Toward A General Theory of Social Control, Volume 1: Fundamentals. Orlando: Academic Press.
1988 `Legal pluralism.' Law and Society Review 22: 869-896.
1965 The Lugbara of Uganda. New York: Holt, Rinehart, and Winston.
1989 `The management of managers: disputing in an executive hierarchy.' Sociological Forum 4: 387-407.
1994 The Executive Way: Conflict Management in Corporations. Chicago: University of Chicago Press, forthcoming.
MYERS, Martha A. and John HAGAN
1979 `Private and public trouble: prosecutors and the allocation of court resources.' Social Problems 26: 439-451.
1934 `Crime and punishment in Tlingit society.' American Anthropologist 36: 145-156.
OTTERBEIN, Keith F. and Charlotte Swanson OTTERBEIN
1965 `An eye for an eye, a tooth for a tooth: a cross-cultural study of feuding.' American Anthropologist 67: 1470-1482.
1967 `Some structural aspects of the feud among the camel-herding Bedouin of Cyrenaica.' Africa 38: 261-282.
REID, John Phillip
1980 Law for the Elephant: Property and Social Behavior on the Overland Trail. San Marino: The Huntington Library.
1984 `Social control in illegal markets.' Pages. 29-58 in Donald Black, ed., Toward A General Theory of Social Control, Volume 2: Selected Problems. Orlando: Academic Press.
SCHNEIDER, David M.
1957 `Political organization, supernatural sanctions, and the punishment for incest on Yap.' American Anthropologist 59: 791-800.
SCHWARTZ, Richard D.
1954 `Social factors in the development of legal control: a case study of two Israeli settlements.' Yale Law Journal 63: 471-491.
SCHWARTZ, Richard and Jerome K. SKOLNICK
1962 `Two studies of legal stigma.' Social Problems 10: 133-142.
SENECHAL DE LA ROCHE, Roberta
1992 `Collective violence as social control.' Presented at the American Sociological Association annual meeting, Pittsburgh, Pennsylvania.
SHIM, Kelley H. and Marshall DeBERRY
1988 Criminal Victimization in the United States, 1986. Washington D.C.: United States Department of Justice.
1991 Homicide: A year on the Killing Streets. New York: Fawcett Columbine.
SMITH, Douglas A., Christy A. Visher, and Laura Davidson
1984 `Equity and discretionary justice: the influence of race on police arrest decisions.' Social Problems 29: 167-177.
1972 `Anarchy and ecology: political society among the Majangir.' Southwestern Journal of Anthropology 28: 153-168.
THODEN VAN VELZEN, H. U. E. and W. VAN WETERING
1960 `Residence, power groups and intra-societal aggression: an enquiry into the conditions leading to peacefulness within non-stratified societies.' International Archives of Ethnography 49: 169-200.
THOMAS-BUCKLE, Suzanne R. and Leonard G. BUCKLE
1982 `Doing unto others: disputes and dispute processing in an urban American neighborhood.' Pp. 78-90 in Roman Tomasic and Malcolm M. Feeley, eds., Neighborhood Justice: Assessment of an Emerging Idea. New York: Longman.
TRUBEK, David M., et al.
1983 Civil Litigation Research Project: Final Report. Part A. Madison: University of Wisconsin Law School.
1992 The Therapeutic Corporation: Social Control in a Post-Bureaucratic Organization. Unpublished doctoral dissertation, Department of Sociology, University of Virginia.
WARNER, W. Lloyd
1958 A Black Civilization: A Social Study of an Australian Tribe. New York: Harper and Brothers (second edition, first edition, 1937).
WERTHMAN, Carl and Irving PILIAVIN
1967 `Gang members and police.' Pages 56-98 in David Bordua, ed., The Police: Six Sociological Essays. New York: John Wiley and Sons.
1984 Murder in Miami: An Analysis of Homicide Patterns and Trends in Dade County (Miami) Florida, 1917-1983. Lanham: University Press of America.
WINANS, Edgar V. and Robert B. EDGERTON
1964 `Hehe magical justice.' American Anthropologist 66: 745-764.
1979 `Minimal politics: the political organization of the Hadza of North Tanzania.' Pp. 244-66 in William A. Shack and Perry S. Cohen, eds., Politics in Leadership: A Comparative Perspective. Oxford: Clarendon Press.
 The jurisdiction of informal social control varies fromsetting to setting depending on the social relationships out of which it emerges. Where people are collectively organized into gangs, clans, or corporate organizations for example, informal social control is more relationally extensive than it is when people pursue their conflicts individually. Perhaps the most extensive informal jurisdiction is found within transnational corporations (a point suggested Calvin Morrill). Transcending national boundaries, these corporate bodies can exert social control against large numbers of people in distant places. Even so, their jurisdiction is limited, within and beyond their home borders, to particular individuals and to particular types of behavior.