The "crack epidemic" and an ensuing wave of violence began in New York City in late 1984. Criminal justice researchers, practitioners and private citizens argued that crack was different from other drugs because it caused a significant increase in the homicide rate. In New York, the total number of homicides committed in 1988 reached 1,896. This was an increase of more than 13 percent over the 1987 total of 1,672 and an apparent single-year record for the city.
We examined the relationship between use and trafficking of crack and homicide in New York City. The analysis is based on data directly collected from the New York City Police Department during the peak homicide year of 1988. The data includes 414 homicides that occurred March 1-October 31. These 414 homicide events created a database including 490 perpetrators and 434 victims.
Who are the Killers?
With regard to location, 45 percent of the 414 homicide events occurred in the street. An additional 35 percent occurred in residences, primarily the residence of the victim. The remainder of the homicides were distributed between bars, abandoned buildings, the transit system, and commercial or other public areas.
Sixty-eight percent of the homicides involved the use of firearms. The vast majority of these were handguns, with .38 caliber and 9 mm. weapons being the most prevalent. About 20 percent of the homicides were committed with knives or other cutting instruments. Physical force, such as strangulation, was the means used in 7 percent of the homicides.
In those homicides where data were available, 95 percent of the perpetrators were male, 83 percent were black, and the mean age was about 27 years. About 23 percent of the perpetrators were also identified as Hispanics. Eighty-four percent of the homicide victims were male, 83 percent were black, and the mean age was about 30 years.
A total of 140 perpetrators (29 percent) were identified by the police as being drug traffickers. A somewhat higher proportion (34 percent) of the victims were classified as drug traffickers (148). Police detectives who completed the forms were asked to distinguish between "high-level" and "low-level" dealers. The vast majority of perpetrators and victims were labeled as low-level traffickers. Only 16 perpetrators and 15 victims were considered to be high-level traffickers.
Why Did They Kill?
All homicides were classified in a tripartite conceptual framework. The three categories were psychopharmacological, economic, compulsive and systemic. The psychopharmacological classification involved those homicides in which a person, as a result of ingesting a specific substance, became excitable or irrational. The economic compulsive classification involves homicides that occurred during economic crimes to finance their drug use. The systemic classification relates to homicides arising from drug distribution or the black market drug trade.
A total of 218 (52.7 percent) of the 414 homicide events were classified as being primarily drug-related. Cases were defined as "drug-related" only when it was believed by both the police and the researchers that drugs contributed to the outcome in an important and causal manner. Homicides were only classified as drug-related in those cases where sufficient information was available to clearly make that determination. Thus, the finding that 52.7 percent of the homicides were drug-related must be viewed as a conservative approximation. The majority (162)"of the drug-related homicides, about 74 percent, were classified as systemic.
The specific drug or drugs associated with each primary classification were also recorded. Crack was the leader in all homicide events and led in each category except for psychopharmacological, where alcohol was the leader. All 21 alcohol-related homicides were psychopharmacological. Alcohol-related murders constituted about 68 percent of the 31 psychopharmacological homicides.
In-Depth Analysis of Crack Homicides
Below is a more detailed review of the 118 "crack-only" homicides and the 13 "crack-in-combination" homicides.
Psychopharmacological. A total of 31(7.5 percent) of the 414 homicide events were classified as being psychopharmacological. Only five of these homicides involved the use of crack. Of these five, one involved crack/alcohol combination and one involved crack/cocaine/alcohol combination. The other three involved crack only. Two of the three crack-only homicides were considered to be victim-precipitated. The single crack-only psychopharmacological homicide that was not victim precipitated involved a 22-y6ar-old black male who, while high on crack, beat his infant daughter to death.
The crack/alcohol homicide involved a 26-yearold black male and a 56-year-old Hispanic woman who was babysitting for the perpetrator. While high on both alcohol and crack he tried to rape her. She resisted, and he struck her numerous times in the head with a blunt instrument, causing death. The perpetrator reported to the police that his alcohol use was the primary motivating factor in his violent behavior.
The crack/cocaine/alcohol homicide event involved a 47-year-old black male who stabbed his 21-year-old black girlfriend to death during a dispute of unknown origin. The perpetrator was high on all three substances. The victim was a reported user of crack, but it was unknown whether she was using at the time that she was murdered.
While these events were tragedies, they are hardly the basis for claims that crack induces violent behavior.
Economic Compulsive. Only about 2 percent of the homicides that were studied were economic compulsively motivated. However, all eight of these economically motivated murders were classified as crack-related. Six of the eight cases involved the murder of elderly persons ranging from 64 to 98 years old, during robberies or burglaries by crack users seeking money to finance their crack use. Of the two economically motivated murders not involving robbery of an elderly person, one occurred between crack users. Police report the perpetrator knew the victim possessed drugs and money, and killed him during a robbery undertaken to finance his personal drug use. The other incident appears to have been victim-precipitated. The police believe that an automobile owner surprised the victim while he was committing theft of his automobile. The owner chased the victim/auto thief into the park, and hit him with the automobile jack handle, killing him.
Systemic Homicides. About 39 percent (162) of all the homicide events studied were classified as systemic. Sixty-five percent (106) of these systemic homicides were classified as being primarily crack-related. These included 100 crack-only cases, with four cases involving crack/marijuana combinations. Twenty-eight percent (45) of the systemic cases were primarily cocaine-related. Only 7 percent (11) of the systemic cases involved drugs other than cocaine or crack.
Systemic homicides included a variety of black market drug business-related incidents. The most common were territorial disputes (44 percent of crack systemic homicides) followed by robberies of dealers (18 percent). Other incidents included assaults to collect debts, punishment of workers, disputes over drug thefts and the dealer selling bad drugs.
Is Prohibition More Dangerous Than Crack?
Several findings emerged rather clearly. First, the majority, 52.7 percent, of the homicide events that occurred during an eight-month study period were classified as drug-related. Second, most of the drug-related homicides, 60 percent, involved the use or trafficking of crack. Cocaine, in all its forms, including crack, was the primary drug involved in about 84 percent of the drug-related homicides. Third, most of the drug-related homicides, 74 percent, were classified as systemic — related to
the drug trade — not drug use. This was true with all drugs except alcohol.
A substantial proportion, 26 percent, of the homicides that took place in New York City in 1988 may be projected from this study to have been crack-related systemic events. The most common sort of crack-systemic homicide was shown to be territorial disputes between rival dealers. Clearly, the emergent popularity of crack, combined with an unstable distribution system, produces a high level of violence.
Crack may be produced with relative ease, thereby bringing into the marketplace many small-scale entrepreneurs. These persons, with comparatively small amounts of cocaine, are able to begin their own businesses. Some of these small dealers are independent of established organizations and the normative controls evidenced in more traditional dealing hierarchies do not exist.
The entry of many small dealers into the crack marketplace has created a number of boundary disputes leading to violence. In an area as small as an apartment house, a tenement stoop, or a street corner, two or more crack dealers may be competing for the same customers. Dealers and customers interact in a highly volatile environment in which disputes and conflicts are routinely settled by a resort to physical force in which one or both of the parties tend to be carrying firearms. There appear to be efforts toward consolidation of these independents into larger organizations, and this trend toward consolidation may involve considerable violence also.
The crack distribution scene is undoubtedly violent. However, the crack phenomenon does not appear to have greatly influenced the overall homicide rate in New York City as reported in the Uniform Crime Reports. This rate did rise steeply in 1988 by about 13 percent. Yet crack first appeared in late 1984, and the homicide rates be- tween 1984 and 1987 were still below the then peak years of 1979-81. In both nature and volume, crack-related homicides largely appear to be replacing other sorts of homicides rather than just adding to the existing homicide rate. The fact that only three of the 414 homicide events surveyed in this report were primarily related to the use or trafficking of heroin is certainly indicative of a decline in heroin-related homicides.
The foregoing data should clearly focus attention on the black market crack distribution "system" as the primary source of crack-related homicide violence. There were few cases of psychopharmacological or economic compulsive homicides involving crack. These data support earlier findings that drug users are more likely to finance their drug use by working in the drug business than by engaging in violent predatory theft.
Paul J Goldstein, Henry H. Brownstein, Patrick J. Ryan and Patricia A. Bellucci, "Most Drug-Related Murders Result From Crack Sales, Not Use," The Drug Policy Letter, March/April 1990, P. 6.