From Alcatraz to Marion to Florence - Control Unit Prisons in the United States

by the Committee to End the Marion Lockdown [1]. This article was written in 1992. The situation as regards the proliferation of control units and the imprisonment binge and its racist nature has in each case become even worse by now, 1995. For the most recent imprisonment statistics see The Continuing Crime of Black Imprisonment


In this article we will trace the history of control unit prisons in the United States from their beginnings in the early 1970's to the situation in 1992, as a means to understanding their function within the prison system. However, any theory involving prisons is ultimately a theory of the society of which they are a part. Indeed, Dostoevsky wrote that if you want to understand a society you should look inside its prisons. Thus, we have attempted to analyze the development of control unit prisons within the context of society as a whole.

The term ``control unit'' was first coined at United States Penitentiary (USP) at Marion, Illinois in 1972 and has come to designate a prison or part of a prison that operates under a ``super-maximum security'' regime. Control unit prisons may differ from each other in some details but all share certain defining features:

  1. Prisoners in a control unit are kept in solitary confinement in tiny cells (six by eight feet is usual) for between twenty-two and twenty-three hours a day. There is no congregate dining, no congregate exercise, no work opportunities and no congregate religious services.
  2. These conditions exist permanently (temporary lockdowns occur at almost every prison) and as official policy.
  3. The conditions are officially justified not as punishment for prisoners but as an *administrative* measure. Prisoners are placed in control units in *administrative* moves and since there are no rules governing such moves (in contrast to *punitive* moves), prisoners are denied any due process and prison officials can incarcerate any prisoner in a control unit for as long as they choose, without having to give any reason [2].
The present article is structured as follows. Sections 2 and 3 will describe the history and ongoing conditions at USP Marion, which has been the focus of control unit development and, from 1983 until now, the premier example of a control unit prison. Section 4 deals with the imminent replacement of Marion by a new, purpose-built control unit prison in Florence, Colorado. Section 5 documents the proliferation of control units, modelled on Marion, in state prison systems across the country. In Section 6 we analyze the function of control units, contrasting the official claims with the facts. In Section 7 we broaden the analysis to look at imprisonment in the United States as a whole and draw conclusions as to the true purpose of prisons. Section 8 describes the state of public opinion on issues regarding prisons and the role of the media in shaping and maintaining that opinion. Section 9 is a brief summary.

Early History of the Marion Control Unit

USP Marion, located in rural southern Illinois, opened in 1963, the same year the federal prison at Alcatraz closed. Alcatraz had gained a reputation as the ``end of the line,'' the federal system's most repressive prison. Prominent gangsters, such as Al Capone, and Robert Stroud, the famous ``Bird-Man,'' were imprisoned there, as were celebrated political prisoners like Rafael Cancel Miranda, the Puerto Rican National Hero, and Morton Sobell, co-defendant with Ethel and Julius Rosenberg. However, the brutality of conditions at Alcatraz proved too controversial in an era when prisons were supposedly committed to the rehabilitation of prisoners.

Marion was constructed to hold 500 ``adult male felons who are difficult to control,'' according to Congressional testimony in 1971 by George Pickett, then superintendent of Marion (Mitford, 1973: 199). Nonetheless, Alcatraz's prisoners were not transferred directly to Marion. Presumably, the BOP considered that the move might spark off significant protest and they wanted to test out the new facility before sending the most politicized prisoners there. Not until the late 1960's were some of Alcatraz's former prisoners transferred to Marion (Breed and Ward, 1984: 10). At about the same time, Marion began its transformation into the new end of the line, a true heir to Alcatraz in its barbaric treatment of prisoners.

The transformation began with the prison's implementation in 1968 of a behavior modification program called Control and Rehabilitation Effort, or CARE. Prisoners in the program were put in solitary confinement and otherwise coerced into participating in group ``therapy,'' which consisted of intense psychological ``attack sessions.'' The purpose was to bring prisoners under the staff's control as totally as possible and turn them against other prisoners (Mitford, 1973: 134-5). 1972 marked a turning point in the program. In July, prisoners began a work stoppage to protest a guard's beating of a Mexican prisoner (Cancel Miranda, 1990). Officials confined all prisoners to their cells for six days, then put seven suspected strike leaders into segregation (solitary confinement). The strike abated briefly, then began again. Prisoners were then subjected to a mass reprisal to end the strike, with sixty men locked in segregation and enrolled in the CARE program, establishing the Control Unit. In 1973, H-Unit at Marion was officially designated the Long-Term Control Unit (Adams v. Carlson, 1973: 621-2; Anderson, 1975; Gruenberg, 1975).

The Control Unit was used to expand the CARE program to include prisoners from throughout the federal prison system ``whose behavior seriously disrupted the orderly operation of an institution,'' according to official federal Bureau of Prisons (BOP) policy (Breed and Ward, 1984: 10). This marked a return to a feature of BOP practice missing since Alcatraz's closing - the concentration in a single prison of those the BOP targeted for special punishment. Like some of the prisoners in Marion's traditional solitary confinement unit, the Disciplinary Segregation Unit (I-Unit), prisoners in the Control Unit were under ``administrative'' rather than disciplinary segregation. Officially, administrative segregation differed from disciplinary segregation in that it was not considered punishment, but rather an administrative response to the prison's purported inability to manage the prisoner by normal means (Adams v. Carlson, 1973: 606).

Prior to the establishment of the CARE program and Control Unit at Marion and similar behavior modification programs and facilities in other prisons, prison officials at least went to the trouble of setting guidelines on under what circumstances and for how long a prisoner should be subjected to solitary confinement. For example, the 1959 Manual of Standards of the American Correctional Association, noting that segregation could have a ``damaging effect upon some inmates'' and that ``[e]xcessively long periods [in segregation] for punishment defeat their own purpose by embittering and demoralizing the inmate,'' recommended ``a few days'' of punitive segregation for most infractions, and an additional thirty to ninety days of administrative segregation in extraordinary circumstances (Adams v. Carlson, 1973: 606). Evidently intending precisely to demoralize prisoners, Marion officials ignored these guidelines for prisoners in the Control Unit. Indeed, in 1975 the General Accounting Office reported that some prisoners in the CARE program had been in the Control Unit for its entire three year existence (Gruenberg, 1975).

In 1978, the BOP added a security-level six category to its prisoner classification system, and in 1979 Marion was designated the only level six prison. That same year a BOP report contemplated turning Marion into a ``closed-unit operation,'' and a 1981 report detailed plans to convert the entire prison into a control unit (Breed and Ward, 1984: 11, 22). Stiffer controls inspired prisoner hunger and work strikes throughout the early 1980's. The longest of these- ``reported to be the longest and most peaceful [strike] in U.S. prison history''- began in September 1980, when the warden, Harold Miller, refused to respond to the following list of concerns, which had been presented to him the preceding month:

  1. Allow Native Americans to practice purification rites.
  2. Allow religious services in segregation and the Control Unit.
  3. Allow Muslims to wear the fez and turban.
  4. Stop the use of boxcar cells (cells having solid doors).
  5. Stop guards from harassing and beating prisoners.
  6. Extend visiting and make the visiting room more comfortable.
  7. Improve medical care.
  8. Improve diet by using real meat.(Susler et al., 1984: 8)
The strike was never broken, but ended when the administration closed the prison factory in January 1981 (Lassiter, 1990: 76; Breed and Ward, 1984: 12). This extraordinary act of resistance was the result of the BOP's practice over the previous decade of sending its most dissident and politicized prisoners to Marion.

In two separate incidents on October 22, 1983, two guards were killed by prisoners in the Control Unit. Although no rebellion resulted and the prisoners responsible were identified, prison officials seized the opportunity to violently repress all prisoners and implement their 1981 plans. The ``lockdown'', or cell-confinement of all prisoners, was imposed on October 27, 1983. The next day, the warden declared a state of emergency. Sixty guards, including specially trained Special Operations Response Team (SORT) members, were transferred to Marion from other institutions to assist in the lockdown. In addition, eight BOP executive staff members and three senior wardens were sent to ``monitor'' the procedure (Carlson, 1984). A guard at the time, David Hale, recalls how a Marion official, evidently uninhibited by the team of outside monitors, set the tone for the ensuing shakedown:

``I seen them carry one inmate down the corridor with a guard on each leg and one on each arm. The assistant warden comes down the hall and grabs the inmate's testicles and starts yanking on them, saying, 'Who's doing it to who now, boy?' Well that was a signal for every guard in the place to do whatever the hell he wanted. I can't describe it to you- I never seen beatings like that. At least fifty guys got it, maybe more.'' (Lassiter, 1990: 76)
The guards were outfitted with helmets, plastic shields, bullet-proof vests and other special gear, and their name tags were removed, making it impossible to identify guards who were involved in abuses[3]. They administered severe beatings while conducting cell shakedowns and forcing cell transfers, using fists, boots, and three-foot riot bludgeons, each with a steel ball affixed to the end. These ``rib-spreaders,'' which have been part of regular equipment at the prison ever since, are designed to separate intercostal rib cartilage and inflict pain without breaking bones or leaving bruises. Prisoners were punched in the face, choked, knocked to the ground, and driven head-first into walls and metal doors. Four prisoners were beaten while in the prison hospital. In many cases, prisoners were handcuffed during the beatings.

Prisoners were subjected to illegal and excessive x-ray examination for contraband. In many cases, the guards ripped off prisoners' underclothing and conducted forced rectal searches. Several prisoners were confined to individual cells for up to four days, handcuffed behind their backs and wearing only underwear. One prisoner testified that he was injected with an unknown drug which caused him to lose consciousness for two days. Personal property was destroyed in the raid. Articles for religious worship, glasses and false teeth were destroyed or seized and never returned. For example, guards desecrated Alan Iron Moccasin's medicine bag and confiscated sacred articles of his Lokata religion. A minister's Koran was taken in the raids, and he was given a Bible in its place.

Prisoners were locked in their cells around the clock. Most privileges were curtailed or eliminated. Congregate religious worship was eliminated. Visits were restricted. After revelations about the beatings surfaced, attorneys were denied entry to meet with prisoner clients for a period of several days in November.

Within days, the Control Unit was expanded from its original seventy-two cells to include all 353 Marion prisoners. The entire population at Marion was collectively, severely and permanently punished in a calculated move by the BOP.

Ongoing Conditions of the Lockdown

USP Marion comprises nine living units, B through I and K. Conditions vary between units. The five general population units, B through F, are located on Marion's East Corridor. Among these units, D, E and F are the most restrictive. C-Unit is slightly less restrictive than these three units and holds prisoners who are being considered for transfer to B-Unit. B-Unit is a pre-release unit with conditions similar to those in most maximum security prisons. On the North Corridor are located the prisons' four ``special living units.'' I-Unit, the Disciplinary Segregation Unit, holds prisoners from the East Corridor units who are on disciplinary or administrative segregation, and those who are being considered for transfer to H-Unit. G-Unit is similar to I-Unit and also holds prisoners in protective custody. H-Unit, or the Control Unit, holds prisoners who are on long-term administrative segregation (Bruscino v. Carlson, 1985: 491-2). K-Unit, or the Director's Unit, holds prisoners assigned there specifically on the order of the Director of the Bureau of Prisons (Dickey, 1990: 69). The following description of conditions at Marion, unless otherwise noted, applies to units D, E and F. At the end of this section we will briefly describe some of the differences between these and the remaining units[4].

D, E and F-Unit prisoners are let out of their cells one and a half hours each day. By comparison, in the rest of the Federal prison system prisoners spend an average of thirteen hours per day out of their cells. The hour and a half of daily ``recreation'' is usually spent in the narrow hallway immediately outside the cell. This time provides little stimulation and no real exercise opportunity. One hour of outdoor recreation in a fenced area is offered once a week in winter and three times a week in summer. The only chance prisoners have to take showers is during the exercise period.

The cell itself measures six by eight feet. Meals are taken through the bars and eaten in the cell- there is no congregate dining. Beds are concrete slabs with pads laid on top of them. At each of the four corners of the bunk is a ring so that the men can be strapped down whenever prison authorities think that it is appropriate. Jackie Leyden from National Public Radio reports that ``guards have the power to chain a man spread-eagled and naked to a concrete bunk'' (Leyden, 1986). Prisoners have reported being chained like that for days at a time.

No one makes any pretensions about rehabilitation. The only jobs are barber and porter. Prisoners may take correspondence courses, but only one at a time. The prison feeds educational tapes into the cells via closed circuit TV, but no instruction, discussion or group classrooms exist. There are no large-group religious services.

Prison officials tamper with letters and legal mail. While it is illegal for prison officials to look into prisoners' legal documents, they do so with impunity. Moreover, they often withhold or send back personal correspondence [5].

Visitation rights are severely restricted and no contact visits are ever allowed. The men can never touch their children, wives or other loved ones who come to visit. Prisoners must conduct conversations through Plexiglas and over a phone, which is monitored. A guard remains present, watching and recording the entire affair. Few visitors venture so far to endure such painful conditions. Prisoners often ask loved ones not to visit while they are at Marion in order to avoid the humiliation that comes with this situation. As a result, one usually finds the visiting room virtually empty. Prisoners are allowed two ten-minute phone calls per month.

Despite the Plexiglas partition separating them from visitors, prisoners are strip-searched after every visit. Finger probes of the rectum may be conducted ``whenever there is reasonable suspicion'' that the prisoner is hiding contraband [6]. A general idea of what constitutes ``reasonable suspicion'' is given by the fact that every prisoner who leaves the prison for any reason is strip-searched and subjected to the ``finger wave'' on his return, despite being shackled and guarded the whole time. Whenever a prisoner is not separated from staff by bars, he is handcuffed behind his back and escorted by two guards equipped with rib-spreader bludgeons.

Only the vaguest and most arbitrary rules exist at Marion. These whimsical guidelines all revolve around pleasing the guards and warden. The guidelines allow for disciplinary actions at Marion for trivial matters, such as failing to replace salt and pepper packets on a food tray, or hanging wet clothes to dry on the bars of a cell. In particular, no rules govern graduation of prisoners to relatively less restricted status within Marion. Prisoner promotions and demotions are officially at the discretion of the assistant warden. However, the power to veto any promotion effectively resides in every guard in the form of the despised Incident Reports, or ``shots,'' citations for rule violations. One of the most common shots is ``disobeying the direct order of a guard,'' which can be used to cite any ``misbehavior'' a guard desires. A single shot wipes out all good-conduct time a prisoner has earned, and puts the prisoner back at the beginning of ``the program'' that ostensibly governs the progression out of Marion to a level five prison. (Leyden, 1986: 2-3)

B and C-Units are the stepping-stones to release from Marion. Prisoners in C- Unit spend slightly more time out of their cells, and are considered for transfer to B-Unit. B-Unit serves as an ``honor unit,'' the last stage before transfer out of Marion. Here prisoners can spend all day outside their cells, eat meals in the dining room, have lockers and suffer restraints only for legal visits. They also work seven hours a day in the prisons' factory for twenty-two cents an hour, which makes the monthly labor bill for the fifty prisoners less than $2000, during which period they produce $250,000 worth of electrical cable for the Department of Defense (Lehman, 1990: 30). However, placement in C and B-Units is completely arbitrary. A prisoner must have one continuous year of good conduct before being considered for C-Unit, a total of eighteen continuous months of good conduct for consideration for B-Unit, and a total of two continuous years of good conduct for consideration for release to a level five prison. There is no specification of what ``good conduct'' is, so even without the intercession of guards and their shots, no clear system for progressing from one unit to the next exists.

At the other end of the spectrum from B and C-Units are the North Corridor Units, G, H, and I-Units. In these units, what minute breathing space exists in D, E and F-Units is further restricted. Cells contain only a toilet, sink and concrete bed. Prisoners spend only one hour a day out of their cells, are strip-searched before and after exercise periods, are allowed only one phone call per month and three showers per week, and are put in both handcuffs and leg-irons and escorted by three guards, one holding their handcuffs, when they are out of their cells. Even this level of punishment is superseded in I and H-Units' ``boxcar'' cells, which have solid second front doors that cut off sound and air circulation.

Despite the gradations of repression, all of Marion functions as a control unit, where the men are under constant and total control of the guards. John Campbell, a prisoner at Marion, comments, ``No one belongs in an environment where he's being buried alive, where he's in a- like a tomb for the dead... And the police have total control over you, and they know they have total control, and they abuse that control frequently, either on a psychological level or a physical level.'' Another prisoner, Steve Layton, adds, ``They try to drag up the monster in you. It eats on a person, on a person's mind.'' (Leyden, 1986: 2)

Marion to be Replaced by Florence

Marion was not originally built as a control unit prison. It has thus been inadequate for the task of implementing the even tighter control of prisoners which BOP Director J. Michael Quinlan, in testimony before a Congressional subcommittee in the fall of 1989, said would constitute an improvement upon Marion's existing regimen (Lehman, 1990: 36-7). The BOP has decided to replace Marion with a control unit prison in Florence, Colorado specifically designed to achieve this goal. Scheduled to open in 1993, the prison's state-of-the-art technology will help to eliminate even the minimal levels of human contact prisoners have at Marion. It has proved very difficult to find out exact details of the new control unit prison to be built at Florence. When a Freedom of Information Act request for information on plans for Florence was submitted to the BOP, the BOP denied the request on the basis that the plans did not yet exist [7]. If that is the case, then the local newspapers appear to know more about the new prison than its designers. The following information comes from such newspapers (Miniclier, 1991: A1; Henson, 1990: B1; O'Keefe, 1991; Chronis, 1990: B1; Harmon, 1991: B2; Associated Press, 1990; Ritter, 1991: 13).

The Marion replacement is one of a complex of four federal prisons being built just south of Florence. The control unit will house 550 prisoners and is designed so that one guard will be able to control the movements of numerous prisoners in several cell-blocks by way of electronic doors, cameras and audio equipment. ``We'll be able to electronically open a cell door, shut it behind the inmate and move him through a series of sliding doors,'' according to Russ Martin, project manager for the Florence prison. Prisoners will be even more restricted than at Marion, according to the Pueblo, Colorado Chieftain: ``Inmates won't have to travel nearly as far in the new Florence prison.'' At Marion the prisoners can at least shout to each other through their bars. At Florence, solid cell doors will make that difficult or impossible [8] and there will be no windows in the cells.

Just five miles from the prison site, in Lincoln Park, is the Cotter Corporation, a uranium milling company owned by Commonwealth Edison of Chicago, Illinois. The area surrounding the mill and nearby railroad has been extensively radioactively contaminated. Uranium tailings dumped in unlined ponds have poisoned the underground aquifer and the nearby Arkansas River. Dried radioactive dust is carried for miles by the high winds. The contamination of the water alone has caused the Lincoln Park area to be on the Environmental Protection Agency's National Priorities List since 1984 and it has been designated a Superfund site for contamination clean-up (O'Keefe, 1991: 10).

The political landscape around Florence is equally bleak. Florence is in Fremont County where more than one in ten of the work force is employed by the Colorado Department of Corrections in the nine prisons clustered around Canon City (O'Keefe, 1991: 10). Prisoners constitute more than ten percent of the population of the county (Miniclier, 1991: A1).

Florence itself is an economically devastated community of 3,000 where unemployment stands at seventeen percent and the prospect of about 1000 temporary and 750-900 new permanent jobs has proved irresistible. Ninety-seven percent of respondents to one local mail-in poll were in favor of the building of the Florence complex. The citizens raised $160,000 to purchase the 600 acres for the site; 400 locals gathered for the ground-breaking; t- shirts bearing a map of the site were ``sold out'' at $7.99; a housewarming barbecue hosted by the BOP was attended by 1000 local residents. Now, Pueblo Community College is offering criminal justice courses customized to suit the needs of the federal prison.

Proliferation of Control Units

The model for the new control unit at Florence is the Security Housing Unit (SHU) at Pelican Bay State Prison in California (Wilson, 1991: 2). The SHU, which opened in December 1989, was designed as the ultimate facility for the implementation of Marion-style repression. Built to hold 1,056 prisoners in near-total isolation, it is already twenty percent over capacity (Smith, 1991: 1). Prisoners are confined to their eighty square foot cells with solid steel doors for twenty- two and one half hours a day. They are allowed out only for a ninety minute ``exercise'' period alone in an empty concrete yard the size of three cells with twenty foot high walls and metal screens overhead. Guards open the sliding doors by remote control and use loudspeakers to direct the prisoners in and out. Prisoners moved off the cell-block for any reason are shackled and flanked by two guards wielding truncheons. Except for the sound of a door slamming or a voice on a speaker, the SHU is silent. Prison officials, not the courts, ``sentence'' prisoners to SHU terms (Corwin, 1990: A1). Often, confidential tips from other prisoners serve as the basis for a disciplinary hearing to determine whether to send the prisoner to the SHU, and these hearings have few safeguards of due process. Many prisoners are sent there for filing grievances or lawsuits or for otherwise opposing prison injustices (Weinstein, 1990). SHU prisoners report the use of ``hog-tying'' (the intertwining of handcuffs and ankle-cuffs on a prisoner), ``cock-fights'' in which guards double-cell enemies or otherwise allow them to attack each other, and forced cell moves using Taser stun-guns, thirty-eight millimeter gas guns and batons [9].

Conditions such as those at the SHU and Marion are replicated in state control units throughout the country. Many of these prisons feature their own innovations in controlling and dehumanizing prisoners. At a second California control unit prison at Corcoran, armed guards patrol the Plexiglas ceilings over the cells and peer in at prisoners through Plexiglas cell walls (Wilson, 1991: 2). At Colorado's Centennial Prison in Canon City, the administrative segregation unit has been expanded to include the whole prison (Foster, 1990; Ruark, 1991). A priest hired by the prison delivers communion through a small, knee-high food slot in a solid steel cell door. ``If you ain't wrapped too tight, 23-hour lockdown can be enough to make you explode,'' says the priest. Guards are armed with ``nut- guns,'' wide-bore guns that fire wildly caroming, acorn-sized ``nuts'' at prisoners from close range. ``It's a miniature cannon,'' the priest explains. ``The recommended technique is to fire at the floor so that the acorn ricochets.'' Prisoners hit by the nuts can be maimed. ``One guy lost his eye, and since I arrived here three years ago, an acorn took off a guy's nose and plastered it to his cheek'' (Johnson, 1990: 12). A specially constructed, $44 million control unit prison, scheduled to be opened near Canon City in early 1993, will hold 500 prisoners, with an additional 250 capacity expansion part of the prison's design (Lemons, 1991).

At Lebanon, Ohio, prisoners under administrative control are held in eight by six foot isolation cells. Each cell has a second door so that prisoners can be locked in the extreme back, darkened portion of the cell. A prisoner describes being leg-shackled, having his arms cuffed to a belt about his waist and being escorted by three guards whenever he is moved from his cell. Other prisoners are forbidden to speak to him (Perotti, 1991). In Missouri, the state prison at Potosi is run by Warden Paul K. Delo, a Vietnam War veteran who, by Missouri law, doubles as the states' executioner since Death Row is at Potosi. Says Delo of his secondary duties, ``One of our officers had an analogy. He said it's just like at your own house. Nobody likes to take out the garbage, but somebody has to''(Uhlenbrock, 1989: 1). Perhaps inspired by Delo's army experience, prison officials apply the ``double-litter restraint'' to recalcitrant prisoners. The prisoner's hands are cuffed behind his back, his ankles are cuffed and he is forced to lie face-down on an Army-type cot, his head turned to the side. A second cot is then tightly strapped upside-down over the prisoner and the ends are strapped shut, totally enclosing and immobilizing him. Carl Swope, a 21 year-old sentenced to seven years for credit card fraud, filed suit after being held in the restraint for three hours (Bryant, 1991: A3).

Other state control unit prisons are at Ionia, Michigan (Detroit News, 1989: B3); Southport, New York; McAlester, Oklahoma; Baltimore, Maryland; Florence, Arizona (Jacobson, 1991): Starke, Florida; Walla Walla, Washington; Westville, Indiana (Associated Press, 1991a); and Trenton, New Jersey (Page,1991). A survey by the Federal Bureau of Prisons found that thirty-six states now operate some form of super-maximum security prison or unit within a prison (Lassiter, 1990: 80). The list continues to grow. Colorado (Lemons, 1991) and Connecticut (Cardaropoli, 1991) have control unit prisons under construction, and Indiana is building a second control unit prison at Sullivan.

Control unit technology is even trickling down to the local level. The Jefferson County Detention Center in Colorado holds each prisoner in an eighty square foot cell equipped with a concrete bed with a mattress on top, sink, toilet and concrete table. Everything from the lights to the locks on the doors is operated electronically by guards in control booths. The jail was designed to allow for a range of control measures, including nearly round-the- clock cell confinement (McGraw, 1986). New York City's Central Punitive Segregation Unit on Rikers Island holds 300 people under twenty one - twenty three hour a day lockdown with no television or radio. Most of those in the ``Bing,'' as the unit is informally known, are detainees awaiting trial. The city plans to expand the unit to hold 900 (Raab, 1991a: 12).

Control unit prisoners have resisted the brutality they are subjected to with the means at their disposal. Prisoners at the Pelican Bay SHU flooded the federal court with over three hundred civil rights petitions, forcing an unusual meeting between federal judges and the prison's warden to discuss prison conditions. Lawyers for the prisoners have since filed a class action lawsuit charging, among other things, that the extreme isolation violates constitutional safeguards against cruel and unusual punishment (Mintz, 1991). At Southport, New York, prisoners capped months of resistance by taking guards hostage and holding three of them for twenty-six hours until the prisoners' grievances were aired over local television (New York Times, 1991; Raab, 1991b).

Probably the most sustained resistance has occurred at the Maximum Control Complex [MCC] at Westville, Indiana, which opened in April, 1991. Sixteen of the thirty five prisoners in the MCC launched a hunger strike in September to expose conditions in the prison: twenty three and a half hour daily cell time, extremely cold temperatures, denial of mail, constant bright lighting of the cells, and severely restricted visitation. The announced minimum stay in the unit is three years. Four of the prisoners continued the strike for thirty-seven days, eating only after prison officials obtained a court order allowing them to force-feed the prisoners (Associated Press, 1991b). The hunger strikes continued intermittently. One prisoner severed off his fingertip with a razor, and a second tried unsuccessfully to do the same (South Bend Tribune, 1992). The protests garnered coverage in papers across the nation (Associated Press, 1991c; 1991d). Prison officials responded by having guards brutally beat prisoners, sometimes while they were in shackles, assigning some of the prisoners to isolation where they are clothed only in their underwear and socks, and obstructing attempts by lawyers to gain entry (Carmody, 1992).

The Function of Control Units

To understand the reasons for the spread of control units, we must determine what function they serve, what it is that they achieve. We will examine what is claimed about control units by prison officials and compare those statements with what is known. We will analyze three specific claims repeatedly made by prison officials all over the country and reported in any media coverage of control unit issues:
  1. Control units contain the most violent prisoners, the ``worst of the worst'', who have proved too violent to be held at other prisons.
  2. Control units reduce violence at other prisons by isolating the most violent prisoners.
  3. The reduction of violence allows security at these other prisons to be relaxed.
The first claim is the major one, on which the other two rest, so we will concentrate on it. The facts of Marion show that the claim is false. Federal prisons used to be given a security rating from one through six, one being the least secure and six being the most secure. In 1984, Marion was the only level six prison in the federal system and prisoners there were supposed to have a corresponding level six rating. However, a 1984 report by consultants hired by a Congressional oversight committee stated that eighty percent of prisoners at Marion did not deserve that level of security (Breed and Ward, 1984). In fact, prisoners are sent to Marion for a variety of reasons and sometimes for no reason at all. For example, the U.S. District Court ordered a cap on prison population and as a result, so many prisoners convicted of felonies in the District of Columbia have been moved to Marion to relieve overcrowding that they constituted seventeen per cent of Marion's population in 1990 (Lassiter, 1990: 80). Virtually all of these prisoners are Black.

There is, however, a trend to be seen. Prisoners have been transferred to Marion for writing ``too many'' lawsuits, for protesting the brutality of the prison system, or for angering prison officials in some other way. In addition, among the many political prisoners who have been in Marion, American Indian Movement leader Leonard Peltier, Sekou Odinga, member of the Black Liberation Army, Alan Berkman, Tim Blunk and Ray Levasseuer were sent directly to Marion from court (Can't Jail The Spirit, 1989; O'Keefe, 1991) thereby disproving the claim that prisoners at Marion have been violent at other prisons.

The Prison Discipline Study initiated in 1989 by the Prisoner Rights Union of Sacramento, California, investigated the question of which prisoners were most often disciplined and how (Prison Discipline Study, 1991). The report showed that solitary confinement was the most common disciplinary action. Included in this report were testimonies by prisoners that those of them exhibiting personal integrity are singled out for brutal treatment. Respondents to the survey described this group as: ``those with principles or intelligence''; ``those with dignity and self-respect''; ``authors of truthful articles''; ``motivated self-improvers''; those ``verbally expressing...[their] opinion'', ``wanting to be treated as a human being'' and/or ``reporting conditions to people on the outside.'' The study shows, therefore, that a practice such as sending prisoners to control units, which is based on arbitrary and subjective judgments by guards and other officials, will target prisoners who are most likely to be challenging the prison system.

In fact, the BOP's own rules for determining who gets sent to Marion are far broader than the ``violent at other prisons'' line given to the media. In the aforementioned ``one through six'' security rating system, prisoners were assigned their security rating on a number of factors: Type of Detainers, Severity of Current Offense, Projected Length of Incarceration, Types of Prior Commitments, History of Escapes or Attempts and History of Violence (Breed and Ward, 1984: 35). Although this rating system is obviously broader than the ``violent'' formula and open to a certain amount of interpretation, the finding that four out of five prisoners at Marion did not have the required level six rating meant the BOP had to find another, vaguer system. They have therefore revised their rules and now classify institutions as minimum, low, medium and high security. Prisoners must be ``high'' security to be sent to Marion, which is determined by pre-commitment factors such as severity of offense. In addition, prisoners at Marion should have a ``maximum'' custody rating, which is determined by post-commitment criteria such as ``disciplinary record'' (Dove, 1991). Having revised these rules, the BOP changed the classification of everyone at Marion to ``high-max'' (Dunne, 1991).

It is admitted at the highest level that a prisoner's political beliefs are basis for assigning that prisoner to a control unit. In a letter to Congressperson Kastenmeier, the then Chair of the Congressional subcommittee that oversees the BOP, Michael Quinlan, the Director of the BOP, stated:

``A prisoner's past or present affiliation, association or membership in an organization which has been documented as being involved in acts of violence [or] attempts to disrupt ... the government of the United States ... is a factor considered in assessing the security needs of an inmate'' (Quinlan, 1987).
We may ask what constitutes ``association'' with an organization, or what is meant by trying to ``disrupt'' the government. In a case brought by a prisoner in the Security Housing Unit (SHU) at the California State prison in Sacramento, Chief Justice Karlton made it clear that prisoners are sent to the SHU for reasons that have nothing to do with discipline. He noted that the plaintiff, who was challenging the prison's forbidding him to practise his Native American religion, was in the SHU for being ``an associate'' of a prison gang, the Mexican Mafia and that ``given that [he] is in the SHU by virtue of his status rather than as punishment for a particular act, there is no apparent way for him to work his way out'' (Sample v. Borg, 1987).

As a last point in our argument against the claim that Marion contains the ``worst of the worst'' we note that for this to be true, all or most prisoners who satisfy their criteria must be at Marion. For example, Oscar Lopez Rivera, a Puerto Rican Nationalist, is in Marion for ``conspiring to escape''. Since he is there, then other prisoners who ``conspire to escape'' should be there as well as all the prisoners who actually try to escape, as well as all the prisoners who actually *do* escape and are apprehended. Are they? There are prisoners at Marion who have assaulted guards (not in itself an indication of violence if the guard had been harrassing and abusing the prisoner). Are all prisoners who have assaulted guards, or even killed guards, at Marion? Obviously the answer is no.

Finally, let us address the two other claims made by officials about control units.

Prison officials claim that Marion, Pelican Bay and the other control units reduce violence in the rest of the prison system. Since we have shown that the control units do not hold the most violent prisoners, this cannot be true and there is no evidence that it has happened. Moreover, all the evidence points to the opposite being true. Most of the prisoners will be released at some stage either back into the general prison population or into society. It is known that control unit conditions produce feelings of resentment and rage and mental deterioration (Korn, 1988). Prisoners will have been so deprived of human contact that it will be hard for them to cope with social situations again. The inhumanity of control units cannot reduce violence, it can only increase it. Evidence includes the high level of violence at Marion during the period before the lockdown, when controls were being tightened but not yet to the extent of completely physically incapacitating prisoners. The tighter controls certainly did not have a calming effect on the prison. In addition, the guard deaths of 1983 occurred in the Control Unit itself.

The claim that control units allow security to be loosened at other prisons is also invalidated because of the truth about which prisoners go there. And again, there is no evidence that the situation in other prisons has improved. Furthermore, Marion has been the model for the numerous state control units [10]. A delegation of the U.S. House Subcommittee on Courts, Intellectual Property and the Administration of Justice which visited Marion in May, 1990, cited the need to ``develop a more humane approach to the incarceration of the maximum-security prison population. This is particularly true because the Federal Bureau of Prisons serves as a model for state prisons and for other countries in the world.'' (Lassiter, 1990: 80) Incredibly, similarity to Marion is now a defense against suits brought to contest inhuman conditions at other prisons (Reed, 1992). The existence of Marion has not improved conditions at other prisons; its example has dragged them downwards toward greater brutality.

Having disposed of the official claims regarding the purpose of control units, we turn to the true function. Ironically, this was clearly stated by Ralph Arons, a former warden at Marion, who testified in federal court:

``The purpose of the Marion Control Unit is to control revolutionary attitudes in the prison system and in the society at large'' (Whitman, 1988: 25).
(Notice ``revolutionary *attitudes*'' not ``actions.'')

This is born out by the large number of political prisoners who are, or have been at Marion and by the Prison Discipline Study. That control of dissent, protest and liberation movements is the true purpose of control units is also shown by history, most especially the history of the early seventies. In September 1971, the prisoners at the state prison at Attica in upstate New York rebelled against the inhuman and racist regime there, declaring their solidarity with all oppressed people and demanding their rights. The rebellion, and the consequent brutal murder of thirty-nine prisoners and hostages by New York State Troopers, under the orders of Governor Nelson Rockefeller, rocked the nation. The whole prison system was at boiling point. Despite the recommendations of the official report into the Attica rebellion that prison conditions be humanized, the response of the New York Department of Corrections was to plan a control unit in which to isolate prisoners such as those who lead the rebellion ( Kaufman, 1971). It was never built, due to resistance led by Martin Sostre, a Puerto Rican prisoner who had run a radical bookshop, groups supporting Puerto Rican political prisoners and POW's and a defense group headed by Angela Davis (Buhle et al., 1990). Even corrections experts judged the planned prison to be too brutal and to be counterproductive to the purported purpose of violence control (Tomasson, 1971). However, not long after, in 1972, the Control Unit at Marion was initiated.

Starting in the early seventies, around the time of the opening of the Control Unit at Marion and the Attica rebellion, the prison population in the U.S. started to increase rapidly. Concurrently, there has been an increase in the proportion of prisoners who are people of color. We will document these developments in the next section but mention them here since they lead us to interpret the proliferation of control units in the United States as an attempt to suppress the increased likelihood of protests and dissent.

Imprisonment as Control

On Dec. 31, 1990 there were well over 1,100,000 people behind bars in the United States, which makes the U.S. incarceration rate (the number of people per 100,000 of the population) the highest in the world at 455 (Mauer, 1992: 3). This is well above the next highest rate which was for South Africa under apartheid (311) and dwarfs rates in the Netherlands (46) and Australia (79) (Mauer, 1992: 5). The number of people in prison in the U.S. has more than doubled over the last decade and it is projected by the government that an additional 300,000 people will be incarcerated by 1995 (New Yorker, 1992: 27). Even in 1981 the situation was so bad that New York State Correction Commissioner Thomas Coughlin admitted that ``the department is no longer engaged in rehabilitative and programming efforts, but is rather forced to warehouse people and concentrate on finding the next cell'' (Day, 1988: 8).

These figures are even more striking when analyzed in terms of race. The incarceration rate for Black men is 3,370 per 100,000, more than seven times that for white men (Whitman, 1991). We do not have current data on the rates of incarceration for other non-white people, however through 1976-78, Indians were arrested at a rate more than ten times that of white people (U.S. Census of Population, 1976-1978). The U.S. incarcerates Black men at a rate five times higher than South Africa does (Mauer, 1992: 1). Just as control units suppress the prison population, so prisons act in our poor, Black, Latin and Native communities. It is no exaggeration to say that hardly anyone in these communities escapes the shadow of the ``criminal justice system.''

The devastation can be expressed in many ways. Black people are twelve percent of the U.S. population, forty-three percent of the prison population (Wicker, 1991). Using data based on a single day in mid-1989, a study by Marc Mauer for the Sentencing Project in Washington, D.C. found that about one in four Black men in their twenties were under some kind of control by the criminal justice system and about one in twelve were actually behind bars (Mauer, 1990). In 1985, the U.S. Bureau of Justice Statistics published the results of a 1979 survey that sought to determine the probability that a person in the U.S. would go to prison in his/her lifetime (Langan and Greenfeld, 1985). Using the data in this report, it can be calculated that in 1979 the probability that a Black man would go to prison sometime in his life was twenty-two percent. In 1992 we can be sure that this is higher so that probably one in four Black men will go to prison in his lifetime. What must this mean for the Black community? Families suffer financially and emotionally. Whatever few jobs might have been available to Black men will be further out of reach for an ``ex-con.'' Prisoners rejoin their communities from prisons which don't even pretend to rehabilitate and where conditions encourage violence and criminality.

Faced with the question ``Why do Black people go to prison at a rate seven times higher than white people?'' we can answer in three different ways. One is that Black people commit seven times as much crime and are genetically disposed to do so. The second is that Black people commit seven times as much crime and something about their disadvantaged social situation is responsible for this. The third is that Black people do not commit that much crime but the criminal justice system is racist enough to make sure they end up in prison that often.

Rejecting the first alternative, the truth must lie somewhere in between the last two answers, and, although it is difficult to determine how much weight to give to each, one cannot escape the conclusion that U.S. society is extremely racist. If the imprisonment rate accurately reflects the crime rate, one is lead to the conclusion that to effectively combat crime, poverty and racism must be eliminated (even if one is not interested in eliminating them for any other reason.)

The other alternative, that in fact Black people do not commit such a disproportionate amount of crime, is indicated by much evidence, though it is hard to calculate the exact degree of the disparity. For example, the number of crimes committed is so huge that actual imprisonments only account for a small fraction of the people who perpetrate them. The crime rate is difficult to determine and the two major national sources of crime data disagree significantly on both quantity and trends. They do, however, both show that the amount of crime is very large: in 1986, between thirteen [11] and thirty-four [12] million crimes were committed. Thus from a huge pool of potential prisoners, i.e., people who have committed crimes, the criminal justice system singles out those who will go to prison. This is done mainly via policing policy. One major example that shows how racist this is, is the ``War on Drugs,'' in which police target poor, Black neighborhoods even though the great majority of drug users are white. It is estimated by the government that by 1995, sixty-nine percent of people in prison will be drug offenders (Mauer, 1992: 7). A front page story in the Los Angeles Times said that while about eighty percent of the nation's drug users are white, the majority of those arrested for ``drug crimes'' are Black (Harris, 1990). Racism also explains why the 1986 Federal Anti-Drug Abuse Act equates selling five grams of crack cocaine worth about $100 with 500 grams of powdered cocaine worth about $50,000, both crimes drawing mandatory prison terms of five years. Black drug users often choose cheaper crack cocaine, while white drug users more often use the relatively expensive powder which is the real profit-maker for the drug trade (McPherson, 1992).

Crime is a problem that must be tackled. However, there is no evidence that high imprisonment rates are the answer to the problem of crime. Indeed, study after study shows that prisons do not deter crime (Blumstein et al., 1978; Visher, 1987: 513-543; Krajick and Gettinger, 1982) and, remarkably, we know of no research that indicates that they do. (The only slight reduction in the crime rate due to incarceration is by the incapacitation of those imprisoned, but the conclusion of the studies referenced above is that massive increases in the imprisonment rate have only a tiny effect on the crime rate.) Imprisoning large numbers of people in order to stop crime has been a spectacular and massively expensive failure. Academic research shows this and even prison officials sometimes admit to the reality of the situation. According to the Director of Corrections of Alabama, ``We're on a train that has to be turned around. It doesn't make any sense to pump millions and millions into corrections and have no effect on the crime rate'' (Ticer, 1989: 80).

Prisons do not reduce crime, so what do they do? They cause direct suffering to prisoners and their families. More subtly, though more significantly to our discussion, they are a major cause of the deterioration of communities of poor people and especially people of color. If one decides that the purpose of prisons cannot be to stop crime, because they do not and this has been known for many years, then one can conclude that this devastation is the real intention. The consequent suppression of active protest amongst people of color against the injustices of a society based on the maximization of profit is obviously a gain for those with a vested interest in such a society.

Control of Public Opinion

Control units seriously violate prisoners' rights. The facts about Marion shows that they serve to suppress dissent among the prison population. Imprisonment does not reduce crime but brutalizes entire communities. The ``War on Drugs'' has no effect on the problem of drug abuse but is a war on Black people. These truths never appear on our televisions or in our newspapers, even though crime and prisons are practically a media obsession.

The present system of mass incarceration with the accompanying specter of more and more control units can only be maintained with at least the tacit approval of society as a whole. So it is not surprising that those of the population least likely to experience the brutality of prison are also subjected to appropriate control procedures. We have already described how the media repeat the falsehoods concerning control units. Newpaper articles often do not even bother to attribute claims to prison spokespeople but make statements such as ``Florence will become the inheritor of the worst of the worst in the federal prison system '' (Pueblo Chieftan, 1990: 4A) as if they were facts.

We face a constant barrage of racist anti-crime and anti-drug hysteria from the establishment. Prisoners are portrayed as incorrigible and dangerous, undeserving of even the most basic human rights. Politicians and the mainstream media never even mention, let alone intelligently discuss, underlying problems of poverty, inequality and racism. Debate is thus limited to how to manage the ever-increasing flood of prisoners, the necessity of creating such a flood being taken as given.

The reality of the role of control units is carefully hidden from public view. Most control units and other newly constructed prisons are located in isolated, economically depressed, rural areas. This serves several purposes. The ardent support of local people, who rely on the prison for desperately needed jobs, is secured and prisoners are isolated from their families and friends.

Political figures support increased imprisonment since most of them thrive on ``tough-on-crime'' platforms. Nor can the courts be relied upon. In Bruscino v. Carlson, Marion prisoners sought compensation for the attacks which occurred during the October 1983 shakedown and relief from the ongoing conditions of the lockdown. A 1985 Magistrate's Report for this case was approved by the full U.S. District Court for Southern Illinois in 1987. The decision found that fifty prisoners who testified to beatings and other brutalities were not credible witnesses, and that only the single prisoner who testified that there were no beatings was believable (Bruscino v. Carlson, 1985). When the prisoners appealed the decision, the ruling of the Fifth Circuit Court of Appeals described conditions at Marion as ``ghastly,'' ``sordid and horrible'' and ``depressing in the extreme,'' but maintained that they were necessary for security reasons and did not violate prisoners' constitutional rights (Landis, 1988: A1).

Finally, there is no discussion of what should be considered a crime and who is to be considered a criminal. The Black drug addict who sells drugs to keep up his habit, the poor man who robs a drug store at gunpoint, the woman who kills her abusive husband: they are all sent to prison and considered dangerous. However, the violation of safety codes by slum landlords and mine owners, embezzlement and fraud by savings and loan executives, pollution of land, seas and atmosphere by oil and chemical company directors, the bombing of schools, hospitals and water purification plants by U.S. presidents, the aggressive marketing of cigarettes (the most deadly narcotic in the world, causing almost 200 times as many deaths as cocaine in the U.S. in 1988 according to C. Everett Koop, Reagan's surgeon general (Shalom, 1992: 15)) across the world by U.S. tobacco companies cause hugely more death, injury and impoverishment and yet are rarely punished by imprisonment. Crimes against humanity and the environment are not illegal if committed by the powerful.


We have described the development of control unit prisons in the U.S. and showed how this is an attempt by prison authorities to suppress protest and dissent within the prison system. The entire prison system is an attack on oppressed people, the poor and especially people of color, rendering them less able to organize and struggle for their rights and their liberation.


  1. The Committee to End the Marion Lockdown would like to thank all the prisoners who have provided information on control units. We welcome comments on this article as well as more information or inquiries concerning control units. Please write c/o CEML, P.O. Box 578172, Chicago, Illinois, 60657-8172.
  2. An example of how this works in practice is the case of an Ohio prisoner who was charged with ``inciting to riot'' for getting other prisoners at his medium security prison to sign a statement of grievances against the prison for bad conditions. The prison disciplinary committee found him guilty and recommended a punitive transfer to a control unit at another prison in Ohio. However, all transfers must be approved at Central Office in Columbus, which approval was denied since under the law, the prisoner could not be punished with such a transfer for what he had done. So the warden at his prison responded by requesting that Central Office approve an *administrative* transfer instead. The transfer was OK'ed without question (Reed, 1992).
  3. The following account is based on on the public report prepared by Cunningham and Susler (1984).
  4. This description was compiled through use of various sources, including: Bruscino v. Carlson (1985); Hart (1984); Gonzales (1986); Marion Penitentiary - 1985 (1985); Leyden (1986); Satchell (1987); Dickey (1990); Lassiter (1990); Isikoff (1991); Lehman (1990).
  5. For example, mail sent by the Committee to End the Marion Lockdown has been returned or ``lost.'' In one case, then warden Gary Henman, when challenged about a lost mailing, implied that twenty-seven prison addressees had lied about not receiving it. Henman wrote, ``You may want to look at their criminal/institutional records prior to putting your faith completely on their statements.'' (Letter from Gary Henman to Steven Whitman, Committee to End the Marion Lockdown, January 27, 1989.) Another mailing was returned because it included a general distribution leaflet for a demonstration with the phrase ``Join Us.''
  6. Among the ``Findings of Fact'' in the Magistrate's Report Bruscino v. Carlson, 1985: 493).
  7. Letter from Wallace H. Cheney, General Counsel for the Federal Bureau of Prisons, to Jan Susler, Attorney for the People's Law Office, Chicago, IL, December 31, 1990. The entire text of the letter read: ``This is in response to your request for information related to the Florence, Colorado project. The issues you enquired about have not yet been decided. Therefore, no records exist at this time pertaining to your request. I trust you you [sic] will find this information useful.''
  8. This assumes that this aspect of the model for Florence, the Pelican Bay State Prison Security Housing Unit, is copied. See e.g. (Wilson, 1991: 2) and Corwin (1990: A1).
  9. Letter from Pelican Bay SHU prisoner Thomas Fetters, to CEML, June 30, 1991. Fetters writes that he was transferred to the SHU for filing a lawsuit after being injured by a guard who assaulted him while he was in physical restraints. Letter from SHU prisoner Robert Lee Davenport, to CEML, September 28, 1990. Davenport reports being hog-tied and left on his cell-floor for ten hours and witnessing other prisoners left like that for twenty hours.
  10. See, e.g., Criminal Justice Newsletter (1990); and Smith (1991), a full-page plea, masquerading as a news report, for the Illinois prison system to build its own version of Marion.
  11. ``Uniform Crime Reports'' in Sourcebook of Criminal Justice Statistics - 1987, Washington D.C.: U.S. Government Printing Office, 1989.
  12. Bureau of Justice Statistics, ``National Crime Survey'' in Households Touched by Crime - 1988, Washington D.C.: U.S. Government Printing Office, 1989.

References Cited

BACK to the CEML Homepage