drugs/violence relationships over time within specific localities, or
to validly compare one city to another.
Law enforcement officials from different cities frequendy claim that
20 percent, or 80 percent, or whatever percent, of their violence
is drug related. They seldom define what they mean by drug related
violence; nor do they specify how they determined a proportion of
violence that is drug related. Did they utilize forensic toxicologies
by medical examiners? Did they utilize investigative reports of
detectives? When there was variation between cities; for example,
one city claiming that 20 p ercent of its violence was drug -related
and another city claiming that fiQ percent of its violence was drug -
related, were there real differences in the nature and amoimt of
drug -related violence between the two cities? Or were the two
cities just defining drug-related violence differentiy? Or counting it
differentiy? It is usually impossible to tell.
One of the major problems in this regard has been the lack of a
consensually agreed upon definition of what is "drug-related
violence." In 1985, I first published my definition of drug-related
I argued that drugs and violence were related in three different
ways: psychopharmacologically, economic-compulsively, or
The psychopharmacological model suggests that some persons,
as a result of ingesting particular substances, may become excitable
and/or irrational, and may act out in a violent fashion.
Psychopharmacological violence may also result from the irritability
associated with withdrawal syndromes or "crashes" from particular
substances. Psychopharmacological violence may involve substance
use by either perpetrators, or victims, of violence. In other words,
substance use may contribute to a person behaving violently, or it
may alter a person's behavior in such a manner as to bring about
that person's violent victimization. Finally, some persons may ingest
substances purposively in order to reduce nervousness or boost
courage and thereby facilitate the commission of previously
intended violent crimes.
The economic compulsive model suggests that some persons feel
compelled to engage in economic crimes in order to finance costly
drug use. Sometimes these economic crimes are inherendy violent,
as in the case of robbery. Sometimes the violence results from an
unintended or extraneous factor in the social context in which the
economic crime is perpetrated. Such factors include the
perpetrator's nervousness, the victim's reaction, the presence or
absence of weapons carried by either victim or perpetrator, the
intercession of bystanders, and so on.
The systemic model refers to the normally aggressive patterns of
interaction within systems of illicit drug distribution. Examples of
systemic violence include territorial disputes between rival dealers.
assaults and homicides committed within particular drug dealing
operations in order to enforce normative codes, punishment for
selling adulterated or bogus drugs, assaults to collect drug related
debts, and so on.
So, having formulated this tripartite conceptual framework, I
undertook a series of studies to validate it, and to elaborate upon it.
I have been studying drugs-violence relationships since 1984. NIDA
supported two field studies on the lower east side of New York City.
The first examined drugs-violence relationships among male drug
users and distributors. The second studied drugs-violence
relationships among female drug users and distributors.
NU supported two homicide studies. The first utilized existing
police records to study the drug relatedness of all homicides that
occurred in New York State in 1984. The second involved working
with New York City police detectives during active investigations of a
sample of homicides that occurred in New York City in 1988.
Data available from these four studies enable us to examine
the nature and scope of drug related violence. While research
findings have been presented in far more detail and depth in a
variety of publications, the following siunmary of key findings
should prove useful for the purposes of this hearing.
About one - half of all violence in all studies was drug-related.
The following findings emerged from the two street studies.
1. Relatively high proportions of violence engaged in by male
and female street drug users and distributors were unrelated to drug
use or trafficking. If half of their violence was drug related, then
the other half of their violence was not drug related.
2. Psychopharmacological and systemic violence were the
most common forms of drug-related violence reported by both
males and females. Economic compulsive violence was rare.
Specifically, male subjects reported that only 5 % of their violent
participations were economic compulsive. Female subjects reported
that only 2 % of their violent participations were economic
3. For both men and women, alcohol was the substance most
likely to be associated with psychopharmacological violence.
Heroin and cocaine were the substances most likely to be
associated with systemic violence.
4. Both the frequency and volume of cocaine use, in either the
powder or rock form, were related to involvement in violence, but
the nature of this involvement was quite different for men and
women. Higher frequencies and volumes of cocaine use were
associated among men with being a perpetrator of violence. Higher
frequencies and volumes of cocaine use among women were
associated with being a victim of violence.
The following findings emerged from the two homicide
1. The two most common types of drug-related homicide were
psychopharmacological and systemic. Very few drug-related
homicides were economic compulsive. That is, very few were
motivated by the compulsive need of a drug user to get money for
drugs. In the New York City sample, in 1988, about 39% of all
homicides, and about 74 % of all drug-related homicides, were
systemic, that is, related to drug trafficking.
2. Psychopharmacological homicides were most often alcohol
3. Systemic cases were most often cocaine (predominandy
crack) related. In New York City, in 1988, 93% of the systemic
homicides involved cocaine.
These findings provide evidence that certain common
assimiptions about drug-related violence are incorrect or
exaggerated. For example, it is commonly believed that an
important threat to public safety by drug users is their violent
predatory acts to obtain money for drugs. Our data indicate that
very few homicide victims were killed by drug users during property
crimes to get money to buy drugs.
Drug users typically try to avoid violent predatory offenses.
This fact is also reflected in the small proportions of economic
compulsive violence that we found in our two street studies. Use of
drugs is often financed by working in a variety of roles in the illicit
drug business. Violence is most likely to arise in the context of the
illicit drug marketplace, and to involve others who are similarly
Another common assumption is that the public safety is
endangered by persons who are "crazed killers" due to their use of
illicit substances. Data indicate that various forms of violence,
including homicide, do occur as a result of perpetrator and/or
victim inebriation. But generally these cases involve people under
the influence of alcohol, a legally obtainable substance.
If we examine trends in the national homicide rate over the course
of the twentieth century, we find the following. Homicide rates
began to dimb when prohibition of alcohol was instituted by
constitutional amendment. Homicide rates peaked in 1933, the
year that prohibition was repealecL Rates then declined and
remained relatively stable through the 1960's. Since 1969,
homicide rates have fluctuated within a fairly narrow range,
roughly 8-to-lO per 100,000 population. The absolute nimiber of
homicides has increased, but so has our population. The rate of
homicide, the number per 100,000 population, has changed very
littie over the last 25 years. There were peaks in 1974, and from
1979-1981, where the rate climbed over 10 homicides per 100,000
population. In 1993, the homicide rate was 9.5 per 100,000
population. This was lower than the rates from 1973-1975, and
lower than the rates from 1979-1981.
The 1979-1981 peak was a result of what I call Cocaine War I (CWI).
Crack had not yet appeared on the scene, and the illicit market
being fought over involved powdered cocaine. Miami was the
murder capital of the United States then. The national homicide
rate then declined in the early and mid-1980s. In the mid-to-late
1980s the homicide rate began to climb again, heralding the arrival
of crack and CWn. New York City and Washington, DC, replaced
Miami as the nation's murder capitals.
The New York experience suggests that in the embryonic stages of
the crack market, a steadily increasing number of new users
provided distributors with sufficient business. Rates of violence at
this preliminary stage of market development was low. As the
market matured, and the number of users began to stabilize,
competition between distributors for "market share" grew.
Organized gangs tried to consolidate turf and bring independent
dealers under their control. For example, some gangs tried to
create a monopoly by forcing small dealers to buy raw products
from them exclusively, and eliminated those dealers who refused.
3 9999 05984 032
Intensified law enforcement efforts probably contributed to
increased levels of violence. Street sweeps, neighborhood
saturation, buy-bust operations, and the like lead to increased
violence in a number of ways. For example, removing dealers from
their established territory by arresting them creates a vacuimi that
other dealers fight to fill. By the time these hostilities have ended,
convicted dealers may have returned from prison and attempted to
reassert their authority, resulting in a new round of violence.
Systemic violence fluctuates with phases of the iUicit market
economy. Rates of homicidal violence were high when a new market
was being forged for powdered cocaine. When those wars were
over, even though there was plenty of cocaine on the streets,
homicide rates declined in the mid-1980's. The peak level of
homicidal violence caused by the crack wars is similar to the peak
caused by the powdered cocaine wars, and also similar to the peak
caused by the alcohol wars during prohibition.
The nature of a particular drug, as a contributor to homicide, is
less important than the nature of the illicit marketplace in which
those drugs are being sold. Crack is currently a major contributor
to drug related violence. But this has little to do with the
psychopharmacological properties of crack. It has more to do with
the fact that crack is currently the most prominent product in
violent, anarchic, street markets for illicit substances. The only
substance clearly associated with a particular form of drug-related
violence is the association between alcohol and
We are finally beginning to understand the multi-faceted
relationships between drugs and violence. However, additional
research is needed to provide the best possible information to
1. A national data collection system should be established that will
enable us to assess trends in drug-related violence over time and
2. Increased information must be provided to judges regarding
which defendants are likely candidates for court diversion
programs, and what sorts of treatment particular defendants, at
particular stages of their lives, are most likely to benefit from.
3. We must better evaluate the impact on drug-related violence of
patterns of law enforcement, judicial and legislative policies,
school-based violence prevention curricula, and other intervention
In conclusion, let me state clearly that in examining drug
relationships to violence, I can find no justification for differential
sentencing policies for rock and powder cocaine.