MCLS Re-Classification
Testimony of Phillip Kassel to Joint Committee on Public Safety of the Massachusetts’ Legislature Regarding Overclassification of Prisoners in State Prisons
(October 25, 2003)
My name is Phillip Kassel. I have worked for the last 10 years as a staff attorney with Massachusetts Correctional Legal Services. I am here to testify regarding a topic that has been the focus of my work at MCLS: how the Department of Correction places prisoners in harsh and expensive maximum-security settings even when there is no need, a tendency that is commonly referred to as “overclassification.” I will discuss terms and the basics of classification, national standards governing how prisoner placement decisions are made, how it is done here in Massachusetts, what is wrong with practices here, and why it is important that changes be made.
Basics
The process for deciding where to confine prisoners is called “classification.” “External classification” governs how prisoners get placed in particular prisons. “Internal classification” governs how prisoners are placed in particular housing units within prisons. Assessing prisoners’ potential vulnerability to violence by other prisoners is part of the classification process, and those most likely to become victims are considered to need “protective custody.” When prisoners first come into the system, the vast majority is sent to MCI-Concord, where their security and custody needs are assessed and they are initially classified. They are later sent to another prison for longer-term housing, and are thereafter reclassified twice annually.
National Standards
There is broad agreement among corrections administrators, researchers, and experts nationally, that classifying prisoners properly is an essential element of effective prison management. Prisoners who are dangerous – to the public if they were to escape; to prison staff and other prisoners while locked up – need secure confinement. Drug treatment and vocational and educational program opportunities are mostly not available in high-security prisons. Prisoners who pose no or small threat to prison security should be maintained in relatively less secure and less expensive facilities where hopefully they can shed addictions that got them in trouble to begin with and learn skills that will help them to stay out of trouble when they get out.
Since this process of deciding which prisoners to house where is so crucial, almost every prison administration in the nation no longer leaves these decisions entirely in the discretion of individual correctional staff. “Subjective” decisionmaking of this sort will invariably result in inconsistent decisions in similar cases. Subjective decisions often rely on factors that may seem valid, but do not stand up to analysis. Therefore, every expert in this area will say that “objective” factors – that is, factors that have been proven to correlate with how prisoners actually behave behind bars – should govern decisions on how to place prisoners. Typical factors are: the prisoners’ criminal and escape history, age, and disciplinary record within prison.
Classification Practices in Massachusetts
Massachusetts, effectively, has a subjective classification process. In part, the failure to rely on objective factors in deciding where to house John Geoghan is what killed him. He was a frail, elderly man, and was placed in the most secure prison in the state in the same housing unit as a predatory killer. The decision to place Geoghan in this setting exemplifies the problem with DOC classification and protective custody practices.
Protective Custody
The Geoghan killing has correctional experts and average citizens around the country scratching their heads trying to understand how his killer was allowed to have access to him at all, let alone the opportunity to strangle and beat him to death. Certainly negligence is involved. The Geoghan killing, however, has at least as much to do with a systemic failure to provide adequate resources for prisoners at risk of violence. There are a total of approximately 70 cells (about half double-bunked) in protective custody units in two prisons for roughly 10 thousand prisoners who are confined in 18 facilities statewide. The American Correctional Association surveyed state prison systems in 1991, and reported that protective custody populations ranged from 4-17% and averaged 5.6% of jurisdictions’ total population. An expert in this area, the architect of the national model Arizona protective custody system, Steve J. Martin, states he becomes concerned regarding the adequacy of protective custody resources when less than 3-5% of total beds are reserved for this purpose. Massachusetts reserves about 1% of beds for protective custody.
Classification Practices Generally
John Geoghan did not belong in a maximum-security prison. While what resulted from his overclassification is extraordinary, his placement in unnecessarily secure confinement put him in the same boat as literally thousands of other prisoners.
The Department of Correction recognized the problem created by its subjective approach to classification as long ago as 1989, when its own research division found that half the prisoners in the system were overclassified, or placed in higher security confinement than could be justified given prison security and public safety needs, relative to national standards. The research caused the Department to launch a process for establishing an objective external classification system in state prisons.
The process conducted in the early 1990’s reflected much of what national authorities say should be part of establishing an objective system of classification. A broad array of DOC staff reviewed extensive data about what factors actually matter in predicting prison adjustment, and devised a weighted system of assigning points to particular demographic and historical indices, with the point totals indicating placement in particular prisons. See score sheets attached.
This work was left incomplete. No effort has ever been made to objectify internal (intra-prison) classification, which is not subject to any statewide standards. The external (inter-prison) system devised, however, could have gone far toward rationalizing prison placements. Unfortunately, this has not occurred, because the system has been undermined due to the lack of an element that national authorities consider essential: institutional “buy-in.” Prison staff and administrators must want to have objective standards corralling their decisionmaking. Without such “buy-in,” the system won’t be implemented properly – and it won’t work.
Despite the effort to develop the objective scheme, prison administrators and staff in the DOC never really came around to believing that giving up complete discretion in classification was a good idea. They loaded the policy with so-called “override” criteria, effectively allowing staff to ignore point scores for any reason at all (one override criterion is called “other”). When concern regarding “override” rates surfaced, the Department watered down the meaning of point scores, effectively gutting the system. Compare attached 1994 and 1996 score sheets. Currently, a prisoner who scores more than a very low point total can be placed anywhere in the system, leaving few decisions to “override.”
These moves reflect a lack of respect for objective classification in the DOC. Commissioner Maloney, in sworn testimony of March 28, 2001, stated, approvingly, that prison staff members are free to ignore point totals if they don’t agree with the results, even if the reasons they don’t agree have already been factored into the objective analysis (e.g. the prisoner’s disciplinary history). He also said he had never read the classification policy manual and didn’t know specifically what it said.
Present Status of Overclassification Problem
Massachusetts has retained a subjective classification system since overclassification was first formally diagnosed in 1989. Besides the failure to take adequate measures to address what all agreed then was an overclassified system, there are reasons to believe the problem has actually worsened. Since 1990, the number of minimum-security and pre-release facilities has gone from nineteen to, according to the DOC website, three. At the same time, a 1000 bed super-maximum prison came on line and was substantially filled with prisoners safely housed previously in medium security. Further, it is well known to all who work in the parole system that significant numbers of prisoners certified by a conservative parole board as ready for release after completing a period of residence in a minimum security prison never actually get paroled (and are therefore not supervised at all upon their release), because the DOC won’t place them in minimum security.
The determination of the state Supreme Judicial Court in Haverty v. Commissioner of Correction, 437 Mass. 737 (2002), provides even more evidence of continued overclassification. The Court held that the DOC illegally ignored regulations designed to prevent arbitrariness in the placement of prisoners into extremely harsh and psychologically damaging long-term solitary confinement – the most secure non-disciplinary confinement in the state – crediting unrebutted expert testimony that 50% of prisoners placed in such conditions not only didn’t deserve this treatment, but could be safely housed in lesser security prisons. In 200 of the 486 files selected for examination, the DOC had not even bothered to tabulate a “score” under the objective system. The correctional expert credited by the Supreme Judicial Court cited specific examples of arbitrary placement in long-term solitary confinement resulting from the failure to assert objective standards. Some prisoners placed in segregation had no history of even minor disciplinary infractions or none within the previous year and had objective classification scores that qualified them for minimum security. One prisoner was four days from release on parole, only to be moved into solitary confinement because he used his alleged gang nickname when he signed a Christmas card, with the message "Wishing you God's wonderful blessings for this Christmas Season." Another Haverty witness, convicted of politically motivated crimes from the 1970’s, was a model prisoner for 17 years in the federal system with excellent work and programming reports, administrator commendations, and virtually no disciplinary history. He was locked in solitary confinement upon his return to Massachusetts at MCI-Cedar Junction, stayed there for several years, and remains in maximum security today due to his past affiliation with a radical political organization defunct for more than 20 years.
Costs of Overclassification
Prison systems, like most, which have implemented objective classification schemes, do not treat prisoners such as those described above as harshly as they are treated here. Nor are such prisoners in other states denied rehabilitative experiences that benefit them as well as the communities to which 97% will return after their release. As discussed by MCLS Executive Director Leslie Walker in her testimony to this Committee, the unnecessary placement of prisoners in high security housing wastes millions of taxpayer dollars. The human costs of these practices are also significant. The Supreme Judicial Court in Haverty quoted expert testimony that:
[P]rolonged solitary confinement is highly toxic to psychological functioning. Inmates go into a kind of stupor, and some even become actively psychotic, agitated and paranoid. Difficulties with concentration and memory, and even overt confusional symptoms, are quite common. Intense anxiety, agitation, and panic attacks occur frequently. Many inmates become overtly paranoid--fearful and preoccupied with the ominous significance of every noise he hears and every shadow that passes his cell . . . [m]any inmates develop severe perceptual disturbances, including perceptual distortions and overt hallucinations.
The number of prisoners are released to the streets directly from maximum security prisons where they may suffer such trauma has more than tripled over the last 10 years. Many other prisoners are needlessly maintained in less onerous but not maximally rehabilitative conditions. Many prisoners are denied education and drug therapy that are proven to reduce recidivism. Overclassification is a serious detriment to public safety. It is a major part of criminal justice and imprisonment policies more and more observers are coming to see as tough on criminals but soft on crime.
What Needs to be Done
We understand that a review of Massachusetts classification practices is currently underway under the aegis of the National Institute of Correction. For the reasons stated above, we are confident that this study will conclude that the prison system remains overclassified. MCLS believes that the DOC must be mandated to properly implement an objective system of classification to correct this problem. We firmly believe that, otherwise, the findings of the NIC will be put to use in the same insincere manner that has marked the last 10 years, leaving us in 10 more years where we are now: with a prison system that is unnecessarily expensive and inadequately rehabilitative in significant part because prison placement decisions are subjective and often poorly-founded, such as in John Geoghan’s case. The Department of Correction, regardless of its current leadership, must be given maximum statutory incentive and continuing oversight to insure that those responsible “buy-in” to the use of objective standards in classification. There is too much history here and too much at stake to leave reform to wishful thinking.
Thank you.
Basics
The process for deciding where to confine prisoners is called “classification.” “External classification” governs how prisoners get placed in particular prisons. “Internal classification” governs how prisoners are placed in particular housing units within prisons. Assessing prisoners’ potential vulnerability to violence by other prisoners is part of the classification process, and those most likely to become victims are considered to need “protective custody.” When prisoners first come into the system, the vast majority is sent to MCI-Concord, where their security and custody needs are assessed and they are initially classified. They are later sent to another prison for longer-term housing, and are thereafter reclassified twice annually.
National Standards
There is broad agreement among corrections administrators, researchers, and experts nationally, that classifying prisoners properly is an essential element of effective prison management. Prisoners who are dangerous – to the public if they were to escape; to prison staff and other prisoners while locked up – need secure confinement. Drug treatment and vocational and educational program opportunities are mostly not available in high-security prisons. Prisoners who pose no or small threat to prison security should be maintained in relatively less secure and less expensive facilities where hopefully they can shed addictions that got them in trouble to begin with and learn skills that will help them to stay out of trouble when they get out.
Since this process of deciding which prisoners to house where is so crucial, almost every prison administration in the nation no longer leaves these decisions entirely in the discretion of individual correctional staff. “Subjective” decisionmaking of this sort will invariably result in inconsistent decisions in similar cases. Subjective decisions often rely on factors that may seem valid, but do not stand up to analysis. Therefore, every expert in this area will say that “objective” factors – that is, factors that have been proven to correlate with how prisoners actually behave behind bars – should govern decisions on how to place prisoners. Typical factors are: the prisoners’ criminal and escape history, age, and disciplinary record within prison.
Classification Practices in Massachusetts
Massachusetts, effectively, has a subjective classification process. In part, the failure to rely on objective factors in deciding where to house John Geoghan is what killed him. He was a frail, elderly man, and was placed in the most secure prison in the state in the same housing unit as a predatory killer. The decision to place Geoghan in this setting exemplifies the problem with DOC classification and protective custody practices.
Protective Custody
The Geoghan killing has correctional experts and average citizens around the country scratching their heads trying to understand how his killer was allowed to have access to him at all, let alone the opportunity to strangle and beat him to death. Certainly negligence is involved. The Geoghan killing, however, has at least as much to do with a systemic failure to provide adequate resources for prisoners at risk of violence. There are a total of approximately 70 cells (about half double-bunked) in protective custody units in two prisons for roughly 10 thousand prisoners who are confined in 18 facilities statewide. The American Correctional Association surveyed state prison systems in 1991, and reported that protective custody populations ranged from 4-17% and averaged 5.6% of jurisdictions’ total population. An expert in this area, the architect of the national model Arizona protective custody system, Steve J. Martin, states he becomes concerned regarding the adequacy of protective custody resources when less than 3-5% of total beds are reserved for this purpose. Massachusetts reserves about 1% of beds for protective custody.
Classification Practices Generally
John Geoghan did not belong in a maximum-security prison. While what resulted from his overclassification is extraordinary, his placement in unnecessarily secure confinement put him in the same boat as literally thousands of other prisoners.
The Department of Correction recognized the problem created by its subjective approach to classification as long ago as 1989, when its own research division found that half the prisoners in the system were overclassified, or placed in higher security confinement than could be justified given prison security and public safety needs, relative to national standards. The research caused the Department to launch a process for establishing an objective external classification system in state prisons.
The process conducted in the early 1990’s reflected much of what national authorities say should be part of establishing an objective system of classification. A broad array of DOC staff reviewed extensive data about what factors actually matter in predicting prison adjustment, and devised a weighted system of assigning points to particular demographic and historical indices, with the point totals indicating placement in particular prisons. See score sheets attached.
This work was left incomplete. No effort has ever been made to objectify internal (intra-prison) classification, which is not subject to any statewide standards. The external (inter-prison) system devised, however, could have gone far toward rationalizing prison placements. Unfortunately, this has not occurred, because the system has been undermined due to the lack of an element that national authorities consider essential: institutional “buy-in.” Prison staff and administrators must want to have objective standards corralling their decisionmaking. Without such “buy-in,” the system won’t be implemented properly – and it won’t work.
Despite the effort to develop the objective scheme, prison administrators and staff in the DOC never really came around to believing that giving up complete discretion in classification was a good idea. They loaded the policy with so-called “override” criteria, effectively allowing staff to ignore point scores for any reason at all (one override criterion is called “other”). When concern regarding “override” rates surfaced, the Department watered down the meaning of point scores, effectively gutting the system. Compare attached 1994 and 1996 score sheets. Currently, a prisoner who scores more than a very low point total can be placed anywhere in the system, leaving few decisions to “override.”
These moves reflect a lack of respect for objective classification in the DOC. Commissioner Maloney, in sworn testimony of March 28, 2001, stated, approvingly, that prison staff members are free to ignore point totals if they don’t agree with the results, even if the reasons they don’t agree have already been factored into the objective analysis (e.g. the prisoner’s disciplinary history). He also said he had never read the classification policy manual and didn’t know specifically what it said.
Present Status of Overclassification Problem
Massachusetts has retained a subjective classification system since overclassification was first formally diagnosed in 1989. Besides the failure to take adequate measures to address what all agreed then was an overclassified system, there are reasons to believe the problem has actually worsened. Since 1990, the number of minimum-security and pre-release facilities has gone from nineteen to, according to the DOC website, three. At the same time, a 1000 bed super-maximum prison came on line and was substantially filled with prisoners safely housed previously in medium security. Further, it is well known to all who work in the parole system that significant numbers of prisoners certified by a conservative parole board as ready for release after completing a period of residence in a minimum security prison never actually get paroled (and are therefore not supervised at all upon their release), because the DOC won’t place them in minimum security.
The determination of the state Supreme Judicial Court in Haverty v. Commissioner of Correction, 437 Mass. 737 (2002), provides even more evidence of continued overclassification. The Court held that the DOC illegally ignored regulations designed to prevent arbitrariness in the placement of prisoners into extremely harsh and psychologically damaging long-term solitary confinement – the most secure non-disciplinary confinement in the state – crediting unrebutted expert testimony that 50% of prisoners placed in such conditions not only didn’t deserve this treatment, but could be safely housed in lesser security prisons. In 200 of the 486 files selected for examination, the DOC had not even bothered to tabulate a “score” under the objective system. The correctional expert credited by the Supreme Judicial Court cited specific examples of arbitrary placement in long-term solitary confinement resulting from the failure to assert objective standards. Some prisoners placed in segregation had no history of even minor disciplinary infractions or none within the previous year and had objective classification scores that qualified them for minimum security. One prisoner was four days from release on parole, only to be moved into solitary confinement because he used his alleged gang nickname when he signed a Christmas card, with the message "Wishing you God's wonderful blessings for this Christmas Season." Another Haverty witness, convicted of politically motivated crimes from the 1970’s, was a model prisoner for 17 years in the federal system with excellent work and programming reports, administrator commendations, and virtually no disciplinary history. He was locked in solitary confinement upon his return to Massachusetts at MCI-Cedar Junction, stayed there for several years, and remains in maximum security today due to his past affiliation with a radical political organization defunct for more than 20 years.
Costs of Overclassification
Prison systems, like most, which have implemented objective classification schemes, do not treat prisoners such as those described above as harshly as they are treated here. Nor are such prisoners in other states denied rehabilitative experiences that benefit them as well as the communities to which 97% will return after their release. As discussed by MCLS Executive Director Leslie Walker in her testimony to this Committee, the unnecessary placement of prisoners in high security housing wastes millions of taxpayer dollars. The human costs of these practices are also significant. The Supreme Judicial Court in Haverty quoted expert testimony that:
[P]rolonged solitary confinement is highly toxic to psychological functioning. Inmates go into a kind of stupor, and some even become actively psychotic, agitated and paranoid. Difficulties with concentration and memory, and even overt confusional symptoms, are quite common. Intense anxiety, agitation, and panic attacks occur frequently. Many inmates become overtly paranoid--fearful and preoccupied with the ominous significance of every noise he hears and every shadow that passes his cell . . . [m]any inmates develop severe perceptual disturbances, including perceptual distortions and overt hallucinations.
The number of prisoners are released to the streets directly from maximum security prisons where they may suffer such trauma has more than tripled over the last 10 years. Many other prisoners are needlessly maintained in less onerous but not maximally rehabilitative conditions. Many prisoners are denied education and drug therapy that are proven to reduce recidivism. Overclassification is a serious detriment to public safety. It is a major part of criminal justice and imprisonment policies more and more observers are coming to see as tough on criminals but soft on crime.
What Needs to be Done
We understand that a review of Massachusetts classification practices is currently underway under the aegis of the National Institute of Correction. For the reasons stated above, we are confident that this study will conclude that the prison system remains overclassified. MCLS believes that the DOC must be mandated to properly implement an objective system of classification to correct this problem. We firmly believe that, otherwise, the findings of the NIC will be put to use in the same insincere manner that has marked the last 10 years, leaving us in 10 more years where we are now: with a prison system that is unnecessarily expensive and inadequately rehabilitative in significant part because prison placement decisions are subjective and often poorly-founded, such as in John Geoghan’s case. The Department of Correction, regardless of its current leadership, must be given maximum statutory incentive and continuing oversight to insure that those responsible “buy-in” to the use of objective standards in classification. There is too much history here and too much at stake to leave reform to wishful thinking.
Thank you.