Previous PageTable Of Contents

Recent Reports 
 Support HRW 
About HRW
Site Map

Human Rights Watch - Home Page


Putting a person behind bars is so common in the United States and so frequently imposed for minor conduct that it seems the country has lost sight of just how serious a punishment imprisonment is. Short of executions, imprisonment is the most severe exercise of a government's legitimate coercive and penal powers.

Imprisoned individuals lose their liberty, autonomy, and the unfettered exercise of most rights. Prisoners are deprived of their families, friends, jobs, and communities. Their days usually pass in unproductive idleness. Life in prison is all too often degrading, demoralizing, dehumanizing and dangerous. Inmates' health, safety, privacy, and even dignity are threatened by overcrowding and violence.

Imprisonment reduces ex-offenders' subsequent incomes and employability; it may increase the probability of future offending; and may result in their being denied the right to vote, to engage in certain occupations, and to receive various public benefits and services.109 Sending a parent or family bread-winner to prison can wreak havoc on the financial and social stability of prisoners' families, and the effects of a parent's imprisonment on children's development is likely to be substantial and deleterious.110 The consequences for communities of having large populations of former prisoners is unknown, but exposing millions of adults to the violence and racism of prison life scarcely bodes well. In inner city black neighborhoods, the high rate at which men are removed to prison may undermine those communities and contribute to perpetuating cycles of crime.

In short, prison is not a sanction that should be imposed needlessly or intemperately -- much less in a racially discriminatory fashion. Yet, in its zeal to tackle a tough social problem, the United States seems to have lost sight of the principles that should govern the use of incarceration. The extraordinary number of nonviolent drug offenders sent to prison bespeaks a nation determined to "send a message" about drugs and crime regardless of whether prison is ineffective, cruel, or unduly costly compared to other ways of responding to drugs. While drug abuse and drug trafficking warrant concerted national efforts, it may be that the human, social, and economic costs of the prison "cure" is worse than the "disease" itself. Certainly, for the black community, it would seem that the choice of penal sanctions to combat drug abuse has imposed inordinately high costs.

With the cocaine epidemic having run its natural course and the nation enjoying the fruits of a sustained period of prosperity, it is possible that the "moral panic" of the past decades will abate. The public and its political leaders may now be able to realize the need to move beyond the war on drugs and to begin to dismantle the racially unjust "drug gulag" it has spawned. Across the country there are encouraging signs of progress: some state legislatures are beginning to debate changes to their mandatory sentencing laws that would restore judicial discretion and flexibility and encourage the use of alternatives to incarceration; the number of drug courts with the ability to require substance abuse treatment for addicted offenders in lieu of a prison sentence is surging; and there is renewed attention to the pressing need for more substance abuse treatment.

Appendix: Methodology

This report uses several sources of data and presents both original as well as previously published statistics on prison admissions and incarceration.

There is no nationwide source of state-by-state data on the composition of prison populations by race and offense. State-by-state data on prison admissions is available from the National Corrections Reporting Program (NCRP) conducted annually by the U.S. Department of Justice. The NCRP annually collects individual-level data on all persons admitted to and released from state and federal prisons and its data sets are available from the National Archive of Criminal Justice Data at the University of Michigan. The explanatory guides that accompany the data provide a detailed explanation of the NCRP methodology and definitions.

For this report, we used the NCRP data for 1996, the most recent year for which they were available. We have provided admissions data from the 37 states who reported to the NCRP in 1996. (The number of participating states varies each year.) We did not obtain admissions data directly from the twelve states that did not participate in the NCRP because the states use different offense codes and we could not ensure consistency in the analysis of their admissions by offense.

We manipulated the NCRP data to identify all admissions that were new court commitments. New court commitments include all persons admitted to prison on a new sentence. That is, we excluded persons who had been admitted to prison previously, were released, and then returned because of parole revocations -- unless they returned with a new sentence. In that case, the new offense would be recorded, not the original offense. The NCRP does not use individual prisoners as a "unit" in its database, but uses each admission as a unit. It is theoretically possible, therefore, that a person can be in the database for 1996 more than once if he or she had been admitted more than once in that year. However, since offenders are rarely sent to prison unless they have a sentence of more than one year, we believe the possible number of double counted cases is insignificant.

Offenders admitted to prison are categorized in this report according to the most serious offense for which they were sent to prison. "Drug" offenders are prisoners admitted with a drug offense as their most serious offense. Anyone convicted of both a crime of violence and a drug offense would be categorized as a violent offender.

A number of states failed to provide complete data on their admissions. In Arkansas, for example, offense designations are missing for about fifty percent of the admitted population, and in Tennessee, offense values are missing for seventeen percent of the admissions. In a few other states there are much lower percentages of missing values.

There are significant numbers of missing values in the NCRP database in the category of type of drugs. We have therefore been cautious in our use of this data and only included the data for marijuana, which seemed consistent with other sources. Similarly, in the category of type of drug crime (e.g., possession vs. trafficking), some states did not provide complete records and our figures are best seen as approximations. The data from Arkansas, Hawaii, and Louisiana were so incomplete that we did not use them at all.

To calculate rates of admission relative to state populations, we have used population figures obtained from the Bureau of the Census for January 1997, which are an interpolation by the Bureau of the Census from the census estimates of July 1996 and July 1997. Reference to adults in this report includes individuals over the age of 18.

For data on incarceration rates and prison populations we used national prisoner statistics from the Bureau of Justice Statistics (BJS). BJS obtains year-end and mid-year counts of prisoners from departments of corrections in each of the fifty states. A distinction is made between prisoners in custody from those under jurisdiction. To have custody of a person, a state must hold that person in one of its facilities. To have jurisdiction means that a state has legal authority over the prisoner. Prisoners under a state's jurisdiction may be in the custody of a local jail, inanother state's prison, or in another correctional facility. In this report, unless otherwise indicated, we use figures of prisoners under state jurisdiction when we are citing prison population rates or prison population composition. However, for the following states, data are for custody rather than jurisdiction counts: Arizona, Florida, Georgia, Illinois, Iowa, Michigan, and Vermont.

For state incarceration rates we used the rates published in BJS, "Prison and Jail Inmates at Midyear 1999," published in April 2000. The racial breakdown of state prison populations was obtained from BJS, "Correctional Populations in the United States, 1996," published in April 1999. We used these state prison populations and census data to calculate incarceration rates by race.

Some states operate unified jail and prison systems, and their prison population figures include both jail and prison inmates. These are Alaska, Connecticut, Delaware, Hawaii, Rhode Island, Vermont, and the District of Columbia.

We have used NCRP and BJS designations of white and black inmates. These figures may include Hispanic whites and Hispanic blacks as some states did not disaggregate Hispanics. Nine states reporting to NCRP identified the race of a large number of prisoners as "unknown."


This report was written by Jamie Fellner, Human Rights Watch associate counsel. Sinead Walsh, a consultant to Human Rights Watch, performed the data analysis. Kenneth Roth, Human Rights Watch Executive Director, and Malcolm Smart, Program Director, edited the report. Production assistance was provided by Jacqueline Brandner and Maura Dundon.

Human Rights Watch would like to thank Allen Beck and James Stephan of the Bureau of Justice Statistics, Marc Mauer of The Sentencing Project, and Marie Pees from the Bureau of the Census for their assistance in the preparation of this report.

We are grateful to the Open Society Institute for the funding that supported this work.

Human Rights Watch

Human Rights Watch is dedicated to protecting the human rights of people around the world.

We stand with victims and activists to bring offenders to justice, to prevent discrimination, to uphold political freedom and to protect people from inhumane conduct in wartime.

We investigate and expose human rights violations and hold abusers accountable.

We challenge governments and those holding power to end abusive practices and respect international human rights law.

We enlist the public and the international community to support the cause of human rights for all.

The staff includes Kenneth Roth, executive director; Michele Alexander, development director; Reed Brody, advocacy director; Carroll Bogert, communications director; ; Barbara Guglielmo, finance director; Jeri Laber, special advisor; Lotte Leicht, Brussels office director; Patrick Minges, publications director; Susan Osnos, associate director; Maria Pignataro Nielsen, human resources director; Jemera Rone, counsel; Malcolm Smart,program director; Wilder Tayler, general counsel; and Joanna Weschler, United Nations representative. Jonathan Fanton is the chair of the board. Robert L. Bernstein is the founding chair.

Web Site Address:
Listserv address: To subscribe to the list, send an e-mail message to with "subscribe hrw-news" in the body of the message (leave the subject line blank).
109 Human Rights Watch and The Sentencing Project, Losing the Vote; Tonry and Petersilia, Prisons.110 More than 1.5 million children are estimated to have a parent who is incarcerated, and many more will have a parent incarcerated at some point during their lives. See Denise Johnson, "Effects of Parental Incarceration," in K. Gabel and D. Johnson, Children of Incarcerated Parents (New York: Lexington Books, 1995); Hagan and Dinovitzer, "Collateral Consequences of Imprisonment for Children," in Tonry and Petersilia, Prisons.

Previous PageTable Of Contents