Describing the standard interrogation techniques for Iraqis detained at Abu Ghraib, Mr. Womack, the lawyer for Specialist Charles A. Graner, said “a certain amount of violence was to be expected,” adding, “Striking doesn’t mean a lot. . . . Breaking a rib or bone—that would be excessive.” Mr. Volzer, the lawyer for Specialist Megan M. Ambuhl, juggled his terms, arguing that it was intimidation, not torture: “I wouldn’t term it abuse.” Mr. Bergrin, the lawyer for Sergeant Javal S. Davis, argued that the prisoner was not harmed when Davis stomped on his fingers. “He may have stepped on the hands, but there was no stomping, no broken bones.”

After the revelation of abuses at Abu Ghraib, Secretary of Defense Donald Rumsfeld found time to draw comparably subtle distinctions: “I’m not a lawyer, but I know it’s not torture—probably abuse.” Rumsfeld’s own blurring of the distinction between obvious torture and possible abuse has a real legal history. The now-famous documents written by lawyers for the White House and the Departments of Defense and Justice—an August 1, 2002, memorandum prepared by Judge Jay S. Bybee and a March 6, 2003, memorandum entitled “Working Group Report on Detainee Interrogations in the Global War on Terrorism” (authorized by the Pentagon’s general counsel, William J. Haynes II)—redefined the meaning of torture and extended the limits of permissible pain.

It might seem at first that the rules for the treatment of Iraqi prisoners were founded on standards of political legitimacy suited to war or emergencies; based on what Carl Schmitt called the urgency of the “exception,” they were meant to remain secret as necessary “war measures” and to be exempt from traditional legal ideals and the courts associated with them. But the ominous discretionary powers used to justify this conduct are entirely familiar to those who follow the everyday treatment of prisoners in the United States—not only their treatment by prison guards but their treatment by the courts in sentencing, corrections, and prisoners’ rights. The torture memoranda, as unprecedented as they appear in presenting “legal doctrines . . . that could render specific conduct, otherwise criminal, not unlawful,” refer to U.S. prison cases in the last 30 years that have turned on the legal meaning of the Eighth Amendment’s language prohibiting “cruel and unusual punishment.”

What is the history of this phrase? How has it been interpreted? And how has its content been so eviscerated?


In the appellate case Turnipseed v. State (1844), Chief Justice Henry W. Collier of the Alabama Supreme Court reversed a lower court’s conviction of the defendant for his “cruel” beating of his slave, Rachel. In overturning the indictment because of its “general terms,” Collier spent a great deal of time interpreting the phrase “cruel and unusual punishment” as it applied to the treatment of slaves in the sixth chapter of the penal code of the state of Alabama. “Cruel, as indicating the infliction of pain of either mind or body, is a word of most extensive application; yet every cruel punishment is not, perhaps, unusual; nor, perhaps[,] can it be assumed that every uncommon infliction is cruel.” Using the phrase’s conceptual uncertainty to evade actual harm done, he denies that a crime has been committed. “We must hold the scales of justice in equipoise, and however odious the offence, we must admeasure right to every one according to law.”

Since the 18th century “cruel” and “unusual” have been coupled in lasting intimacy in our legal language and courts, yet they have been vexed by a persistent rhetorical ambiguity that has been alternately used to protect prisoners and legitimize violence against them. Unlike due process, the business of cruel and unusual punishment does not have a history so much as a kind of compulsive repetition; the history of its jurisprudence cycles interminably between these two poles: safeguarding rights and justifying their revocation.

First appearing in the English Bill of Rights of 1689, drafted by Parliament at the accession of William and Mary, the phrase “cruel and unusual punishment” seems to have been directed against punishments unauthorized by statute, beyond the jurisdiction of the sentencing court, or disproportionate to the offense committed. The American colonists included the principle in some colonial legislation, and after much debate the formula was incorporated into most of the original state constitutions. It became part of the Bill of Rights in 1791 as the Eighth Amendment to the U.S. Constitution: “Excessive bail shall not be required, nor excessive fines imposed, nor cruel and unusual punishments inflicted.” The American draftsman intended that the phrase apply to “tortures” and other “barbarous” methods of punishment, such as pillorying, disemboweling, decapitation, and drawing and quartering. In other words, what mattered in the American context was unusual cruelty in the method of punishment, not the prohibition of excessive punishments.1

As perhaps the least understood of all parts of the Constitution, the Eighth Amendment has no landmark ruling, although it received its most expansive interpretation during the prisoners’ rights movement. Jackson v. Bishop (1968, Eighth Circuit) concerned the corporal punishment of convicts. Justice Harry Blackmun—writing his opinion during the height of the prisoners’ rights era and two years before his appointment to the Supreme Court—recognized the need to give substance to the Eighth Amendment and recognized how distinctions between degrees of brutality, or between the meanings of words, remained a pretext for continued excess. “We choose to draw no significant distinction between the word ‘cruel’ and the word ‘unusual’ in the Eighth Amendment.” Authorization of whipping with the strap as punishment for not picking enough cotton or leaving cucumbers on the vine—whether ten lashes, in the fields, or within 24 hours of any earlier whipping—prompts Blackmun to ask, “How does one, or any court, ascertain the point which would distinguish the permissible from that which is cruel and unusual?”2

In Laaman v. Helgemoe (1977), the federal district court held that confinement at New Hampshire State Prison constituted cruel and unusual punishment in violation of the Eighth Amendment. The court’s far-reaching relief order constituted the broadest application ever of the Eighth Amendment to prison conditions, condemning “the cold storage of human beings” and “enforced idleness” as a “numbing violence against the spirit.”

And in the most famous Eighth Amendment case, Furman v. Georgia (1972), the Supreme Court declared capital punishment cruel and unusual, and therefore unconstitutional. In Justice William Brennan’s words, the system of capital punishment was “irrational and arbitrary,” was “degrading to human dignity,” and deprived the criminal of “human status.”3 The Court voted 5-4 to strike down every capital punishment law in the United States.4 In a lengthy concurring opinion that ranged from an analysis of the three cases under consideration to the English and American legal history of the term “cruel” to an assessment of the necessity or usefulness of such an extreme punishment in contemporary society, Thurgood Marshall wisely complained that “the use of the word ‘unusual’ in the English Bill of Rights of 1689 was inadvertent, and there is nothing in the history of the Eighth Amendment to give flesh to its intended meaning.”


While Brennan and Marshall sought to make the Eighth Amendment a prohibition against degrading and inhuman punishment, Chief Justice Burger’s dissent in Furman (joined by Blackmum, Powell, and Rehnquist) has set the tone for its recent interpretation. Burger explained that “of all our fundamental guarantees, the ban on ‘cruel and unusual punishments’ is one of the most difficult to translate into judicially manageable terms.” This unmanageability, what Burger described as “the haze that surrounds this constitutional command,” invites rhetorical slippage in defining the limits of torture, and, at its extreme, allows the complete evasion of actual harm done.

The shift away from more expansive Eighth Amendment protections began in earnest in the 1980s with a series of cases challenging inadequacies in medical care for prisoners, use of force, and conditions of confinement.5 In struggling to create a framework for recently emergent prison jurisprudence, the court sought to give meaning to words such as “cruelty,” “pain,” “injury,” and “punishment.” But the new legal analyses of the Eighth Amendment weakened existing standards and increasingly shifted attention from the type and degree of injury or indignity suffered by prisoners to prison officials’ subjective motivations.

In Rhodes v. Chapman (1981), the Supreme Court, concerned about the new federal judicial activism, sought to clarify the federal role in the operation of state prisons. By the mid-1980s, federal law had turned away from rehabilitation in the direction of a radically retributive penology that emphasized incapacitation. Words such as “degradation,” “degeneration,” “imposed dependency,” and “unnecessary suffering” would never again be applied to that class of persons called “prisoners.”

Writing for the majority in Rhodes, Justice Powell asserted that prison overcrowding does not fall within the scope of “serious deprivations of basic human needs” by contemporary standards. Referring in particular to the horrific conditions of confinement in the two Arkansas prisons of Hutto v. Finney (1978), he argued that discomforts such as the double celling of inmates in Rhodes were not serious enough to violate the constitutional standard. “To the extent that such conditions are restrictive, and even harsh, they are part of the penalty that criminal offenders pay for their offences against society. . . . Therefore, short of causing unnecessary and wanton pain, deprivations . . . simply are not punishments.” He did not specify the degree of severity that would violate the Eighth Amendment, and he suggested a policy of deference to the penal philosophy of prison officials.

It was only, however, when William Rehnquist became chief justice that the court fully revealed its talent for defining away the substance of an Eighth Amendment violation. Its decisions have, in turn, had a profound impact on the recent “torture memos,” which provide the legal basis for routinizing exceptional treatment. But they echo a much older history. If there is something unique about contemporary punishment in the United States—practices anomalous in the so-called civilized world (the death penalty, prolonged and indefinite solitary confinement, use of excessive force and other kinds of psychological torture)—that special thing can be found in a colonial history of legal stigma and obligatory deprivation. The argument of legitimacy, security, and necessity has its most crucial, if most concealed, history, in the Code Noir (1685) of the French Antilles, the British West Indian slave laws of the 18th century, and the black codes of the American South. The Court’s recent Eighth Amendment decisions summon in new places and under new guises this older genealogy of slavery and civil incapacity.

In The History, Civil and Commercial, of the British West Indies (1793), Bryan Edwards explained the logic of containment necessary in countries where slavery is allowed: “the leading principle on which the government is supported is fear: or a sense of that absolute coercive necessity which, leaving no choice of action, supersedes all questions of right.” In her groundbreaking but out-of-print The West Indian Slave Laws of the Eighteenth Century (1970), the Jamaican historian Elsa V. Goveia emphasizes how the codification of already existing laws depended on a policy of both protection and disabling. Though “the provisions safeguarding the slave as a person were either laxly enforced or neglected,” she wrote, “the part of the law which provided for his control and submission continued in vigor.”

Let us not forget that the minimal needs of slaves—their “necessary wants” and their “personal security”—were defined in great detail in legislation and case law. Black codes and slave courts in the North American colonies, like those in the Caribbean, focused intensely on protecting the bodies of slaves while masking the extremities of mutilation. And in many legal restrictions, the chance to exceed what might be considered “humane” lay in the unsaid—in those places where the law falls silent—or where the language is deliberately unclear or hypothetical.6 This spurious generality, operating under cover of excessive legalism, is perhaps nowhere so pronounced as in laws that made violence against slaves a “necessary” or “ordinary” incident of slavery. In John Haywood’s A Manual of the Laws of North Carolina (1808), a person would be judged “guilty of willfully and maliciously killing a slave” except when the slave died resisting his master or when “dying under moderate correction.” To style the “correction” of a slave that causes death “moderate,” is to assure that old abuses and arbitrary acts would continue to be masked by vague standards and apparent legitimacy.

In the Black Code of Georgia (1732-1899), assembled by W.E.B. Du Bois for the Negro exhibit of the American section of the Exposition Universelle in Paris in 1900, the state penal codes are compiled, replete with their detailed adjustments over time. In the penal code amended and approved in January 1851, if an overseer or employer inflicted “unusual” or “inhuman” punishments, the question remained of whether this treatment was “cruel.” The particular acts of cruelty were listed: “unnecessary and excessive whipping, beating, cutting or wounding or . . . cruelly and unnecessarily biting or tearing with dogs . . . withholding proper food and sustenance.” In the very act of curbing gratuitous and extreme cruelty, the meaning of “human” is held in suspension for the slave for whom the use of whips, cudgels, and dogs was not only possible but to be expected. This commitment to protection thus became a guarantee of tyranny, and the attempt to set limits to brutality, to curb tortures, not only allowed masters to hide behind the law but ensured that the guise of care would remain a “humane” fiction.

The ghost of slavery is built into our legal language and holds our prison system in its grip. To the extent that slaves were allowed personalities before the law, they were regarded chiefly—almost solely—as potential criminals. During the second session of the 39th Congress (December 12, 1866-January 8, 1867) debates raged on the meaning of the exemption in the 13th Amendment to the Constitution. It abolished slavery “except as punishment of crime whereof the party shall have been duly convicted.” The parenthetical expression guaranteed enclosure, a bracketing of servitude that revived slavery under cover of removing it. Those who were once slaves were now criminals, and forced labor in the form of the convict lease system ensured continued degradation. As Charles Sumner warned, the locale for enslavement would move from the auction block to the courts of the United States.


The Rehnquist Court has formulated new rules for determining the limits of cruelty for those restrained in their liberty. It is no accident that in cases in which the majority opinion delivered increasingly harsh decisions regarding the punishment, transfer, confinement, and segregation of prisoners, dissenting justices recalled Ruffin v. Commonwealth (1871). In that case justices were called on to define the condition of a convict and consider the implications of civil death for the applicability of the Bill of Rights to the case of Woody Ruffin, a convict charged with murder on a chain gang. Justice Christian decided, “The Bill of Rights is a declaration of general principles to govern a society of freemen, and not of convicted felons and men civilly dead. Such men have some rights it is true, such as the law in its benignity accords them, but not the rights of freemen. They are the slaves of the State undergoing punishment for heinous crimes committed against the laws of the land.” In using Ruffin to condemn the prisoners’ status as dead in law in the 20th century, Justices Marshall, Brennan, and Stevens exhumed the state-sanctioned bondage the court did not name.7

Lewis v. Casey (1996) recalled the civil death of penal servitude. The Supreme Court overturned the district court’s decision and reversed the court of appeals, which had ruled as inadequate the law libraries and legal assistance programs in the Arizona state prison system. Justice Antonin Scalia, delivering the majority opinion, disclaimed the rights or needs of prisoners to discover grievances and to litigate effectively once in court—to “transform themselves,” as he put it, “into litigating engines.” By the time the majority—which ranged from five to nine justices depending on the issue being considered—decided the case, plans were in place to gut already existing libraries in the Arizona prison system and to substitute forms, pencils, and bilingual paralegals for law libraries, legal assistants, and the right to conduct research or ask the state for assistance.8 In his dissent, Stevens summoned the ghost of Ruffin without naming it: “While at least one 19th-century court characterized the inmate as a mere ‘slave of the State,’ . . . in recent decades this Court has repeatedly held that the convicted felon’s loss of liberty is not total. The ‘well-established’ access to the courts . . . is one of these aspects of liberty that States must affirmatively protect.”9

The creation of a new class of civil slaves has enabled a mobile, endlessly adaptable strategy of domination and control. What is at stake here is the translation of this strategy into legal terms. Through an often ingenious technical legalism, the court has paved the way for cruelty that passes for the necessary incidents of prison life. Extreme verbal qualifications make deprivation or injury matter only when “sufficiently serious,” when involving “more than ordinary lack of due care,” or inflicting “substantial pain.” Conditions such as indefinite solitary confinement are unconstitutional only when they pose a “substantial risk of serious harm.”

In Hudson v. McMillan (1992), Keith Hudson, an inmate at the state penitentiary in Angola, Louisiana, sued three corrections security officers for punching him in the mouth, eyes, chest, and stomach. One of the officers held Hudson in place. The third officer, the supervisor on duty, watched the beating, saying, “Don’t have too much fun.” By the time Hudson v. McMillan reached the Supreme Court, Justice O’Connor decided that the use of excessive physical force might constitute cruel and unusual punishment, even if no “serious injury” resulted.10 What does “serious” mean? The bruises were only minor, though there was swelling of face, mouth, and lip. The officers cracked his dental plate and loosened his teeth.

Justice Clarence Thomas wrote the dissenting opinion, and Justice Scalia joined. He argued that the judgment had not only ignored the “significant injury” requirement, but had loosened the Eighth Amendment “from its historical moorings.” What, after all, did the framers mean by “barbarous” punishment? The rack, the thumbscrew, drawing and quartering. How, then, could a mere beating be “sufficiently serious”? Turning to dictionary definitions from 1771 to 1828, the dissent narrowed the meaning of “punishment” to “the penalty imposed for commission of a crime”—not the methods used for controlling prisoners. Thomas concluded, “A use of force that causes only insignificant harm to a prisoner may be immoral, it may be tortuous, it may be criminal, and it may even be remediable under other provisions of the Federal Constitution, but it is not ‘cruel and unusual punishment.’ ”

What is the threshold of suffering necessary to trigger a violation? In Madrid v. Gomez—a class action suit against California’s Pelican Bay state prison in 1993 heard by the federal District Court of California—prisoners incarcerated at Pelican Bay challenged the constitutionality of a broad range of conditions and practices to which they were subjected. Chief Judge Thelton Henderson condemned the habit of caging inmates naked outdoors in freezing temperatures “like animals in a zoo”; the unnecessary and sometimes lethal force used in cell extractions; and the scalding of a mentally disabled inmate, burned so badly that “from just below the buttocks down, his skin peeled off.” An officer said, mockingly, “Looks like we’re going to have white boy before this is through.”

Finding in the plaintiff’s favor, Henderson wrote that “defendants have unmistakably crossed the constitutional line with respect to some of the claims raised by this action,” citing failure to provide adequate medical and mental-health care and condoning a pattern of excessive force. He reserved his greatest condemnation for conditions in the Special Housing Unit. Yet even though indefinite solitary confinement in the SHU, the separate, self-contained super-maximum security complex, might cause “psychological trauma,” it did not cross over into the realm of “psychological torture.” Although Henderson acknowledged that conditions in the SHU “may well hover on the edge of what is humanly tolerable for those with normal resilience,” such conditions remain within the limits of permissible pain. They do not violate “exacting Eighth Amendment standards.” Henderson explained that these conditions of “extreme social isolation and reduced environmental stimulation” did not violate the Eighth Amendment in regard to all inmates, but only when imposed on those inmates “at risk of developing an injury to mental health of sufficiently serious magnitude.”11


In Louisiana ex rel. Francis v. Resweber (1947), Willie Francis, “a colored citizen,” was sentenced to death by a Louisiana court. The attempted electrocution failed due to mechanical difficulties, and Francis petitioned to the Supreme Court, arguing that a second attempt to execute him would be unconstitutionally cruel. Justice Stanley Reed, writing for the majority, ruled against Francis. Even though he had already suffered the effects of an electrical current, that does not “make his subsequent execution any more cruel in the constitutional sense than any other execution. The cruelty against which the Constitution protects a convicted man is cruelty inherent in the method of punishment, not the necessary suffering involved in any method employed to extinguish life humanely.” The dissenting justices understood Francis’s experience to be akin to “torture culminating in death,” and asked, “How many deliberate and intentional reapplications of electric current does it take to produce a cruel, unusual and unconstitutional punishment?”

How does punishment, no matter how insufferable, become legal? While granting that the Eighth Amendment prohibited “the wanton infliction of pain” and admitting that Francis would now be forced again to undergo the mental anguish of preparing for death, Reed concluded by turning to the intent of the one who pulls the switch: “There is no purpose to inflict unnecessary pain nor any unnecessary pain involved in the proposed execution. The situation of the unfortunate victim of this accident is just as though he had suffered the identical amount of mental anguish and physical pain in any other occurrence, such as, for example, a fire in the cellblock.”

The intent requirement of Louisiana ex rel. Francis, would become the controlling precedent for later cases that analyzed how the “cruel and unusual punishments” standard applied to the conditions of a prisoner’s confinement. In Franzen v. Duckworth (1985), Judge Richard Posner, writing for the U.S. Court of Appeals for the Seventh Circuit, concluded that although shackled prisoners were injured during transport when their bus caught fire, there was no Eighth Amendment violation. The intent requirement was not met, since the officers had not intended “maliciously” to cause harm. “Negligence, perhaps; gross negligence . . . perhaps; but not cruel and unusual punishment.” What happened was nothing more than “if the guard accidentally stepped on the prisoner’s toe and broke it.” Invoking Samuel Johnson’s A Dictionary of the English Language, Posner defines “punishment” as “ ‘Any infliction or pain imposed in vengeance of a crime.’ ” In other words, punishment is decreed by the sentencing judge and has nothing to do with what happens afterward, whether deprivations within or accidents outside a prison. Only “malicious intent,” not what Justice Frankfurter called “an innocent misadventure”—he was referring to the accidental character of the failed execution in Louisiana ex rel. Francis v. Resweber—could make unconstitutional what prisoners suffered after incarceration, no matter how harmful to their minds and bodies.

The joint opinion in Gregg v. Georgia (1976), which overturned Furman v. Georgia, coined the phrase “unnecessary and wanton infliction of pain.” But it was the Rehnquist court, and especially Scalia’s precedent-setting opinion in Wilson v. Seiter (1991) that gave the intent requirement and the word wantonness their fiercest play. It would prove to be the most crucial Eighth Amendment prison case in a decade. Pearly Wilson, an inmate at the Hocking Correctional Facility in Ohio, brought a pro se lawsuit alleging that conditions in the prison, including overcrowding, excessive noise, inadequate heating and ventilation, unsanitary dining facilities, and lack of protection from communicable disease, violated the Eighth Amendment. Scalia, upholding the decision of the lower court, focused on the meaning and extent of punishment. Writing for the five-member majority in this sharply divided decision—joined by Rehnquist, O’Connor, Kennedy, and Souter—he relied on Posner’s definition of punishment in Duckworth v. Franzen: “a deliberate act intended to chastise or deter.” Adopting the “subjective component” standard of Estelle v. Gamble, which concerned “deliberate indifference to serious medical needs,” Scalia argued that the deliberate-indifference standard should apply to all conditions of confinement claims.

Estelle had presented two alternatives for the validation of Eighth Amendment violations: either those that are incompatible with the “evolving standards of decency that mark the progress of a maturing society” or those that “involve the unnecessary and wanton infliction of pain.” Wilson made no allowance for such a choice. In establishing an Eighth Amendment violation, the Wilson Court recognized as prohibitive only “obduracy and wantonness” and “not inadvertence or error in good faith.” The preoccupation is with the knowledge, deliberation, or intent of those in control. If not a specific part of the prisoner’s sentence, then, deprivations are not really punishment unless they are imposed by officers with “a sufficiently culpable state of mind.” In other words, no matter how much actual suffering is experienced by a prisoner it is not a subject for judicial review unless the intent requirement is met.

Elizabeth Alexander, of the ACLU National Prison Project, argued for the plaintiffs that no inquiry into state of mind should mitigate unconstitutional conditions, which are often the result of accumulated actions over time. She explained that the “government has an affirmative duty to supply the basic necessities of life to those whom it has deprived of the ability to supply those necessities on their own.” What is at issue here is not the structure of the two-pronged objective and subjective approach, but the complete replacement of objective conditions with concerns about the thoughts of prison officials. “Once continuing conditions of confinement in a prison are bad enough to violate the Constitution by denying the basic necessities of life, the point of injunctive relief is to end the suffering, not to fix the blame.”

But the Wilson court was not listening to concerns about suffering. Instead, they emphasized that prison conditions cannot reach the level of cruel and unusual punishment unless they produce “the deprivation of a single, identifiable human need such as food, warmth, or exercise.” In arguing that anything “so amorphous as ‘overall conditions’ “ cannot constitute an Eighth Amendment claim, the Court dismantled the traditional “totality of circumstances” test that had been so crucial to lower federal court cases in the ’70s.

The indifference to objective conditions is carried forward in the torture memos, which display the same obsessive concern with intent.12 Whether the interrogator has maimed, blinded, or killed the suspect does not matter, unless the interrogator intended to maim, blind, or kill. In the section titled “Specifically Intended,” it is explained that violation of the “Torture Statute” “requires that severe pain and suffering must be inflicted with specific intent” and that the defendant “must have expressly intended to achieve the forbidden act.” If we follow the legal logic, the full force of the mental (it gets to be wanton, malicious, obdurate, willfull), is transferred to the person of the government official. He even gets another loophole: “a good faith belief” that whatever he did would not result in mutilation or death. Did he intend to harm? Did he act in good faith? The possibilities are endless. The results—the presence of a mutilated, blind, or dead body—get defined away by the vain search for intent, while the defendant who committed the act is vindicated.

What becomes of the legal personality of the detainee, once tethered to these recognized acts of will or agency? Subjectivity emerges as the privilege of those in control, while something vicious is being done to the object of harm, now reduced to a mere body controlled by administrative power.


The Rehnquist court’s Eighth Amendment cases prepared the ground for the verbal quibbles, fastidious distinctions, and parsing of definitions that characterize the recent memoranda prepared for the “war on terror.” The legal decimation of personhood that began with slavery has been perfected in the logic of the courtroom and adjusted to prisoners. This reasoning—so long ignored, except by corrections officials who learned how to manipulate legal language—was carefully studied by the White House lawyers.13

The third section of the March 6 memorandum, called “Domestic Law,” while adopting much of the legal language and logic of the August 2002 memo, makes explicit the use of recent Eighth Amendment cases and presents a chilling recapitulation of the cases I have been discussing. In fractured, powerfully condensed form (in just four pages), the Eighth Amendment standards analyzed—whether the determination of mental culpability or significant injury, or the question of what renders unconstitutional either conditions of confinement or excessive force—distill from pages of legal opinions the precise legal foundation for evading the character of punishment.

Whether the Eighth Amendment has been violated or not does not depend on the cruel and inhumane treatment of prisoners: it depends on the motivation or intention of prison officials. The memo repeats Wilson v. Seiter: all claims about unacceptable prison conditions must show “deliberate indifference” to the conditions of prisoners. And it reiterates the Wilson majority’s refusal to recognize anything “so amorphous as ‘overall conditions’ of confinement,” focusing instead on the actual, physical, “specific deprivation of a single human need.”

Verbal qualifiers gut the substance of suffering in favor of increasingly rarified rituals of definition. The imprecision of such terms not only neutralize the obvious but trivialize abuse. Consider this: in the March 6 memorandum, the legal analysis relies particularly on the definition of the term “severe” in the Torture Statute in Federal Criminal Law (18 U.S.C. 2340). Torture is defined as any “act committed by a person acting under the color of law specifically intended to inflict severe physical or mental pain.” Since the word is not defined in the statute, the Pentagon lawyers decided to “construe a statutory term in accordance with its ordinary or natural meaning.”

Their turn to the dictionary definition is a crucial gesture repeated throughout the memo. The pile-up of references—to Webster’s New International Dictionary, The American Heritage Dictionary of the English Language, and The Oxford English Dictionary—introduces the next section on “Severe mental pain or suffering,” which then leads to a list of what constitutes “prolonged mental harm,” which ushers in the further equivocation of “prolonged” or “lasting” but not “necessarily permanent damage.” And once schooled in rituals of redefinition, the lawyers turned to their dictionaries in order to apply them to the new situation (to “terrorists,” not “criminals”). Then they parsed words to a degree that went far beyond the practice of the courts in order to derive legal standards of interrogation during this “war without end.”

The March 6 memorandum also discusses at length why the United States had imposed reservations to Article 16 of the United Nations Convention Against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment” (proposed by the UN in 1984 and ratified by the United States in 1994). The lawyers explain it was “primarily because the meaning of the term ‘degrading treatment’ was vague and ambiguous.” According to Amnesty International’s “Briefing for the UN Committee Against Torture” (May 2000), the United States’s reservation to Article 16 “has far-reaching implications and can apply to any US laws or practices which may breach international standards for humane treatment but are allowed under the US Constitution, for example, prolonged isolation or the use of electro-shock weapons.”

In his memorandum to the president in January 25, White House Counsel Alberto Gonzales argued that the Geneva Convention rules governing treatment and interrogation of prisoners had been rendered “obsolete” and “quaint.” What, then, was the legal foundation for the humiliation, degrading treatment, and physical abuses of prisoners in Iraq, Afghanistan, and Guantànamo Bay, Cuba?14 It was that other law, the law for prisoners—and the language of corrections—in the United States, which has reshaped itself in accord with the situational and discretionary legal system of slavery.


In Furman v. Georgia, Justice Brennan argued that there could be cruelty far worse than bodily pain or mutilation. It was not just “the presence of pain” that proved the significance of the Eighth Amendment but the treatment of “members of the human race as nonhumans, as objects to be toyed with and discarded.” The ominous leeway of American legal rules—from slave codes, to prison cases, to the Bush administration’s torture memos—redefines these persons in law. That redefinition—the creation of a new class of condemned—sustains a metaphysics that goes beyond the mere logic of punishment. Once you create the category of the stigmatized, whether they are called “terrorists,” “security threat groups” (gangs in our prisons), or “security detainees” (prisoners in Iraq), the use of torture can be calibrated to the necessities of continuously evolving and aggressive security measures.

What is most striking about current legal redefinitions of punishment is how much they have in common with statute and case law that reconstructed the identity of the slave. In terms of social conditioning, this re-animation of servile or subhuman status as the criminal “type” guarantees the stigma that justifies dehumanization.

We must fix with precision this characterization of servility. Slave codes in the southern United States demanded that slaves receive clothing, food, and lodging sufficient to their basic needs. In Creswell’s Executor v. Walker (1861), slaves, although dead to civil rights and responsibilities, must be provided with “a sufficiency of healthy food or necessary clothing . . . and the master cannot relieve himself of the legal obligation to supply the slave’s necessary wants.” Like the slave whose brute body had yet to be protected against unnecessary mutilation or torture, the criminal is reduced to nothing but the physical person.

The Prisoner’s Litigation Reform Act (tacked onto a spending bill by Congress), which Clinton signed into law on April 26, 1996, dramatically curtailed prisoner litigation into the next century. Designed to limit what was said to be a massive increase in “frivolous” inmate litigation, the PLRA permits injunctive relief related to prison conditions, but it erects substantial hurdles that must be negotiated before such relief can be given. What are the contours of injury? In order to get an injunction, a plaintiff must prove that every plaintiff or member of the proposed class has suffered actual, physical injury, thus prohibiting damages for mental injury. The prisoner must prove that the request for relief is narrowly focused, extends no further than necessary to correct the injury, and is the least intrusive means necessary to correct or prevent the harm. As an inmate wrote to me, “Only prisoners are excluded from relief or damages stemming from mental pain or suffering (as if such pains are rightfully reserved for us alone).”

This realm of constitutional minima—alternating between mere need and bare survival—set the stage for Guantànamo Bay and Abu Ghraib. I recall the words of Marine Brigader General Michael R. Lehner at Guantànamo Bay in 2002: “There is no torture, no whips, no bright lights, no drugging. . . . We are a nation of laws.” But what kind of laws? Laws that permit indefinite solitary confinement in state-of-the-art, high-tech units, with cell doors, unit doors, and shower doors operated remotely from a control center. Physical contact is limited to being touched through a security door by a correctional officer while being placed in restraints or having restraints removed. Inmates have described life in the massive, windowless supermax as akin to “living in a tomb,” “circling in space,” or “being freeze-dried.”

When does an emotional scar become visible? To make it visible is to stigmatize, yet only certain kinds of stigmatization are recognized: those that accord with the substandard of what prisoners are assumed to be. They are all bodies. Only some are granted minds. And who is to decide? The unspoken assumption remains: prisoners are not persons. Or, at best, they are a different kind of human: so dehumanized that the Eighth Amendment no longer applies. The naked pyramid of flesh in Abu Ghraib, the kneeling and shackled bodies, blindfolded by blacked-out goggles and hooded in Guantànamo, sanction degradation. Such inhuman treatment, however, is made lawful when our government refuses to recognize that “cruel, inhuman, and degrading” treatment has a precise meaning, when our current courts continue to ignore obvious violations of human dignity and worth. In a penal system that has become instrumental in managing the dispossessed, the unfit, and the dishonored, such phrases as “minimal civilized measure of life’s necessities” or “the “basic necessities of human life” prompt us to reconsider the meaning of “human.”


1) For the still unsurpassed analysis of the strange adaptation of the “cruel and unusual” clause to the exigencies of colonial America, see Anthony F. Granucci’s “’Nor Cruel and Unusual Punishments Inflicted’: The Original Meaning,” California Law Review 57(1969): 839-865.

2) For Blackmun’s gradual turn against the death penalty and a bracing discussion of his “abolitionist conversion,” see Austin Sarat, When the State Kills: Capital Punishment and the American Condition (Princeton University Press, 2001), 251-254.

3) At the start of the 20th century, in Weems v. United States (1910), the Supreme Court for the first time turned away from the mere ban on “torture” and other “barbarous” punishments and considered also punishments disproportionate to the offense committed. But it was the decision of Chief Justice Earl Warren in Trop v. Dulles (1958) that articulated the “evolving standards of decency” argument that would be crucial to Brennan’s opinion in Furman. Not only did Warren emphasize a flexible interpretation of the Eighth Amendment that would conform to changing social conditions and enlightened public opinion, but he posited “the dignity of man” as the lynchpin of the Eighth Amendment. More powerfully still, he emphasized mental suffering or anguish as central to the meaning of cruel and unusual punishment. For Warren, nothing could be more cruel than the demolition of the capacity to exercise the rights of persons in the civil community—“no physical mistreatment, no primitive torture,” but “instead the total destruction of the individual’s status in organized society,” having “lost the right to have rights.”

4) In just two years, 28 state legislatures enacted new death penalty statues. In 1976, a majority in Gregg v. Georgia reinstated the death penalty, since adequate procedural safeguards had been adopted. As I write, the Supreme Court is preparing to hear Roper v. Simmons, revisiting the constitutionality of executing minors aged 16 years or older. The United States is one of only five countries that execute juvenile offenders. Two years ago, in Atkins v. Virginia, the court declared unconstitutional capital punishment for mentally retarded offenders.

5Estelle v. Gamble (1976), Rhodes v. Chapman (1981), Whitley v. Alpers (1986), Wilson v. Seiter (1991), Hudson v. McMillian (1992), Helling v. McKinney (1993) and Farmer v. Brennan (1994).

6) For a detailed analysis of this calculated evasiveness in terms of the Code Noir, see Joan Dayan, Haiti, History, and the Gods (Berkeley and London: University of California Press, 1995, 1998): 199-212.

7) See the dissents in Jones v. North Carolina Prisoners Labor Union, Inc. (1977) (Marshall dissenting, joined by Brennan); Meachum v. Fano (1976) (Stevens dissenting, joined by Brennan and Marshall); and Lewis v. Casey (1996) (Stevens dissenting).

8) The cynical play with de minima legality helps us to understand how Washington, by finding alternatives to constitutionally protected legal procedures, now aims to skirt the Supreme Court decisions regarding the Guantànamo detainees. After the Lewis v. Casey decision, the Arizona Department of Corrections banned “all formal use of Inmate Legal Assistants and Law Clerks,” substituting instead paralegals under contract to the Arizona state prison system. After the decisions in Hamdi et al. v. Rumsfeld, Secretary of Defense, et al.; Rumsfeld, Secretary of Defense v. Padilla et al.; Rasul et al. v. Bush, President of the United States,, the U.S. Department of Defense has announced its intent to set up what it calls “Combatant Status Review Tribunals.” Under this plan, detainees would not be allowed legal assistance. Instead they would be provided what the Bush administration calls a “personal representative,” whose job it would be to explain the legal process to the detainees and help them gather evidence.

9) For a fuller discussion of this case, as well as the Supreme Court’s return to an anachronistic and mandatory deprivation of rights, privileges, and capacities, see Joan Dayan, “Held in the Body of the State,” History, Memory, and the Law, eds. Austin Sarat and Thomas R. Kearns (Ann Arbor: The University of Michigan Press, 1999): 183-249 and “Legal Slaves and Civil Bodies,” Materializing Democracy, eds. Russ Castronovo and Dana Nelson (Durham and London: Duke University Press, 2002).

10) Even though the majority opinion broke with the “significant injury” requirement (e.g. leaving permanent marks or requiring medical attention), in his concurring opinion, Justice Blackmun stressed that mental as well as physical harm mattered. Aware of what the court left unsaid, he made explicit the inclusion of “psychological” harm: “As the Court makes clear, the Eighth Amendment prohibits the unnecessary and wanton infliction of ‘pain,’ rather than ‘injury.’ ‘Pain’ in its ordinary meaning surely includes a notion of psychological harm.”

11) After Madrid, Henderson installed a special master at Pelican Bay, and prison abuses declined. In July 2004, Judge Thelton Henderson renewed his threat to take over California’s entire penal system and condemned Governor Arnold Schwarzenegger’s deal with the prison guards’ union, arguing that the union had already seized too much control of corrections department operations.

12) Torture is defined in the Convention Against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment as “any act by which severe pain or suffering, whether physical or mental, is intentionally inflicted on a person” in order to extract information or a confession.” Terms such as “torture,” “inhuman,” and “degrading treatment” or “punishment,” though often considered to be placed in a hierarchy, are interlinked in Article 3 of the European Convention for the Protection of Human Rights and Fundamental Freedoms. Inhuman or degrading treatment—the very terms of suffering that the torture memos and our courts have refused to recognize—remain the focus of the European Convention. There is, moreover, no need for intention to cause degrading treatment or punishment.

13) The more the law got involved in prisoners’ rights, the more prison regulations carried the power of redefining these very rights. In my work on the Arizona prison system and the courts’ rationalization of custody and control, I do not imply a direct relation between the language of court decisions and actual practice inside the prisons. Instead, I demonstrate how the legal manipulation of terms, by both lawyers and wardens, can evade an obvious Eighth Amendment violation. For example, if you can claim that classification is not punitive, not disciplinary, but merely administrative, then something called “administrative segregation”—even if it means indefinite isolation in solitary—is not subject to judicial review. By engaging language in legality within the courts and inside the prisons, the function of labeling the criminal “type”—the manipulation of “status”—has kept correctional actions and legal opinions in dialogue.

14) The continued internationalization of our prison project under cover of ensuring democracy is now the focus of “reform” in Haiti, where U.S. prison administrators direct the redefinition of civil life. Not only are Jean-Bertrand Aristide’s supporters torture, killed, and imprisoned, but on August 17, 2004, a trial in Port-au-Prince acquitted Jackson Joanis and Jodel Chamblain of the 1993 murder of businessman Antoine Izmery. Chablain, the co-founder and chief of operations of FRAPH (Front Revolutionnaire pour ‘Avancement et le Progres haitiens), Haiti’s notorious death squad, is supported by the American-backed government, which is not only reforming the Haitian judiciary but supervising the reconstruction of the Haitian prison system. See DominiqueEsser and Kim Ives, “Haiti and Iraq: The Abu Ghraib and Haiti prison connection, Haiti Progres (June 17, 2004) at; Dan Frosch, “Exporting America’s Prison Problems” (May 12, 2004) at; “Uncle Sam Wants You Anyway,” Alternet, May 24, 2004; and especially the ongoing reports of the Institute for Justice and Democracy in Haiti (Brian Concannon Jr., Director) at

Now available from Boston Review Books