Homicides among Chicago families: 1870-1930.

• Author: Chilton, Roland

There are at least two exciting aspects of the files available to the authors of this volume. One is the age and range of the data--the fact that these accounts of Chicago homicides cover a sixty-year period that starts in 1870. (1) In addition, these historical accounts of deaths reported to the Chicago police contain a surprising amount of detail. Because of the amount of detail and number of cases available, we can focus on family homicides--homicides where the victim and offender are related by birth or marriage, by more or less permanent living arrangements, or by some apparent degree of emotional attachment combined with a desire or plan to create a family. One obvious question about these cases concerns the extent to which family homicides have changed over the sixty-one year period for which information is available. Another obvious question concerns the ways in which these earlier family homicides are different from or similar to contemporary accounts of family homicide.

Any comparison of these accounts with contemporary studies will be complicated by the definitions of family homicides that are used. One contemporary approach to the use of very similar information focuses on intimate partner violence or violence by intimates. "Intimates" in these studies have generally referred to people who are husbands, wives, boyfriends, or girlfriends. (2) Other family members are ignored in this focus on intimates. In part, this approach probably reflects a concern for male violence against women. Although the intimate partner studies include discussions of homicides committed by women, much of the emphasis in them is on the victimization of women. Another reason some researchers have limited the discussion to "violence by intimates" is that a major source of information on violent crimes other than homicide is the National Crime Victimization Survey, which does not collect information on victims under twelve-years-old. (3) Moreover, the only real possibilities for looking at the family relationships of all of the victims and offenders in a set of criminal cases are to use something approximating incident-based data. (4)

The files available for this analysis are essentially incident-based in that information about suspects can be linked to victim information and both types of information can be checked against a set of case summaries. Using them, I was able to combine homicides by spouses, lovers, parents, children, and other family members. The complete list of categories is shown in Table 1. In this family homicide approach, spouses may be "common law" husbands and wives, lovers may be heterosexual or homosexual, parents may be stepparents or grandparents, and children may be stepchildren, foster children, grown children, or sons and daughters-in-law. Former husbands and wives are included as are former boyfriends and girlfriends. Also included are the friends and associates of family members who were or were thought of as romantic or sexual rivals of other family members.

As indicated in the top part of Table 1, were we to focus on violence by "intimates," as the term is sometimes used, there would be only 906 cases of spouses and lovers to examine. By including parents and children and other family members who are reported as homicide victims or defendants, we expand our focus to "family homicide," which adds another 475 cases to the set. This produces a total of 1381 cases of homicide involving family members or people who might eventually form families.

If we expand the "lovers" classification to include rejected or jealous suitors, thirty-five more homicides need to be added. Finally, if we include as family-related homicides those committed by or against persons outside the family who are closely involved with a family member, we must look at another seventy-one homicides. Examples of such "triangle" homicides are those where a husband kills his wife's lover or where his wife's lover kills him. Any one of the parties in these situations can be a victim or a defendant. With these situations treated as family homicides, there were 1487 homicides among Chicago families from 1870 to 1930. (5) Were we to include sixty-six "unintentional" and presumably accidental family deaths, this number is 1553. In the analyses that follow, however, these sixty-six deaths and 1731 unintentional non-family deaths, primarily motor vehicle accidents, are not included.

All 1553 family homicides can be viewed in their larger context by noting that they constitute only fourteen percent of all of the homicides in the file. Moreover, they are unlike the vast majority of homicides in the file in some important ways. For example, when family involvement is ignored and the sex of suspects is tallied by the sex of the victim for all cases, we find that ninety percent of the known homicide defendants were men or boys and that seventy-nine percent of the victims were male. As a consequence, seventy-two percent of percent of the cases in which the sex of the victim and offender was available involved a male suspect and a male victim.

However, when the total set of cases is divided into family and non-family homicides and unintentional deaths are excluded, we find sharply contrasting pictures. In fifty percent of the family homicides the victim is female and the suspect is male. This figure is nine percent for non-family homicides. For family homicide cases, male suspects with male victims make up only twenty-three percent of all cases, but for non-family homicides this figure is eighty-four percent. The details for this comparison of the patterns for non-family homicides and cases involving family members are shown in Table 2. In the top part of the table, the sex of the defendant is tabulated for male and female victims for cases where the sex of both offenders and victims were reported and where there was no indication of a family relationship between victims and offenders. In the bottom part of the table, the same information is shown for cases where both the offender and the victim were family members with different degrees of closeness. Accidental deaths have been excluded from both tabulations.

The pattern shown for all 1428 cases in Table 2 is important, but another important fact is that the family homicides shown in Table 2 were not evenly distributed over the sixty-one year period. About 87% of all family homicides occurred after the year 1900. As shown by the bottom line in Figure 1A, family homicides increased from one in 1870 to eighty-four in 1930. Moreover, the sharpest increase in family homicides occurred after 1918 and the increase in non-family homicides was much more pronounced. The top line in Figure 1A shows...