Disciplinary Responsibility in Prison

Moral and Legal Views

Whether it is fair to hold inmates accountable for misconduct born of mental illness turns on two factors: causal responsibility and moral responsibility. (This territory overlaps, of course, with moral and legal views of insanity.) Causal responsibility means that a person's action was the primary reason for some effect or harm⁴; it does not automatically confer moral responsibility. It is reasonable to judge people as morally responsible for their actions when they possess certain powers and abilities; the most basic of these might be autonomy (i.e., to act volitionally in the service of some intention) and rationality (i.e., to have rational desires, beliefs, and intentions).⁵ Thus, from a moral standpoint, it is unfair to punish individuals who do not possess the elements of moral responsibility, that is, who are not moral agents (e.g., children and, in some cases, people with severe mental illness).

Criminal jurisprudence acknowledges that it is unfair to punish people who cannot appreciate moral or legal injunctions.⁵ The insanity defense is the plainest expression of this moral sensibility. The same moral sensibility is generally absent, however, from the administrative laws that govern inmate conduct; standards of administrative insanity are rare. In a policy survey of 41 states in the United States and the Federal Bureau of Prisons, Krelstein² reported that, although many states permitted some degree of input from mental health professionals during the disciplinary process, only four states and the Federal Bureau of Prisons had policies that explicitly allowed prisoners to be exonerated when mental illness was found to have caused the alleged infraction.

Constitutional rights are at stake when inmates with mental illness are punished for symptomatic behavior. In Coleman v. Wilson,⁶ a class action lawsuit in California, the U.S. District Court found that the California Department of Corrections and Rehabilitation was deliberately indifferent to inmates' mental health needs. The court found, among other things, that when inmates with mental illness violated prison rules, custody officials did not consider the potential role of mental illness. As a consequence, inmates with mental illness were punished frequently, including placement in segregation, for symptomatic behavior. In conjunction with other rulings, the court found that the department violated plaintiffs' Eighth Amendment right to be free of cruel and unusual punishment. Other federal courts have overseen settlement agreements in which input from mental health professionals in disciplinary decisions was a stipulation.^7,–,9

Consideration of mental illness in determinations of criminal responsibility may also be a due process protection, although there is strong disagreement on the form this protection should take.^10,11 In Huggins v. Coughlin,¹² the New York Appellate Court upheld lower court decisions^13,14 when it concluded that “the Hearing Officer is required to consider the prisoner's mental condition in making the disciplinary disposition when the inmate's mental state is at issue” (Ref. 12, p 845). Other cases in New York have relied on Huggins¹⁵ and its related decisions in lower courts.¹⁶ In Powell v. Coughlin,¹⁷ the U.S. Court of Appeals ruled that a New York policy allowing hearing officers to consult with mental health staff outside of the presence of the accused inmate was consistent with the due process requirements set forth in Wolff v. McDonnell.¹⁸

Expert Views

Although there are moral and legal grounds for assessing disciplinary responsibility, not all experts agree that such assessments are appropriate. Metzner and Dvoskin,¹ for example, argued that because successful insanity pleas are rare in the community, inmates who could meet a standard of nonresponsibility would be so uncommon that limited clinical resources in correctional facilities are best directed elsewhere. Metzner¹⁹ expressed additional reservations: few psychologists have the necessary training to complete assessments of responsibility; the standard for nonresponsibility is hard to define; and a standard of nonresponsibility could affect relations with custody staff adversely (as in a case in which an inmate commits battery against an officer and is exonerated). Krelstein² expressed the concern that if prison mental health staff were to act in a forensic role, this would be “detrimental” to clinical care (presumably due to the dual relationship). Consistent with the position of these authors, the National Commission on Correctional Health Care does not endorse assessments of disciplinary responsibility specifically.²⁰ Rather, it encourages mental health staff to “offer consultation to disciplinary hearing officers that helps them recognize when mental illness may be a contributor to inmate misconduct” (Ref. 20, p 116).

Others take the opposite view. Knoll²¹ argued that assessments of disciplinary responsibility are necessary because the outcomes of disciplinary hearings are consequential. As a result of symptomatic behavior, inmates with mental illness may experience additional deprivations of liberty (e.g., extended prison time, denial of parole) and may be assessed penalties that threaten their mental health (e.g., placement in administrative segregation). The subsequent increases in disciplinary points result in placement at higher-security facilities, which are more restrictive and dangerous.²² The American Bar Association,²³ in its standards for the treatment of prisoners, states that when correctional officials determine mental illness to be the cause of an infraction and, when sanctions would be ineffective, discipline should be avoided. Similarly, Abramsky and Fellner,³ in their comprehensive report on mental illness in prison, recommended that mental health staff assess whether an inmate's behavior is “connected to or caused by mental illness” (Ref. 3, p 14). Finally, Maue²⁴ explained that a mental health evaluation that addresses whether the misconduct was “the result of mental illness” is a “key element” of optimal disciplinary practices (Ref. 24, p 48).

In the opinion of some experts, inmates would be unlikely to meet a nonresponsibility standard in correctional institutions that provide constitutionally adequate mental health systems.¹ Unavoidable realities do exist, however. Even when care is optimal, considerable time is often necessary to resolve serious symptoms, decompensation periodically occurs, residual symptoms are common, and some disorders, such as dementia and intellectual deficiencies, are chronic and unremitting. For some, the relationship between mental illness and disruptive behavior is so complex²⁵ that treatment cannot be expected to eliminate the role of symptoms in misconduct. Finally, the enforcement of prison rules is not suspended in mental health units. Thus, to include an assessment of disciplinary responsibility is to acknowledge the complexities of mental illness, treatment, and misconduct.

Research indicates a clear need for assessments of disciplinary responsibility. It is well established that inmates with mental illness are not only disproportionally represented in prison²⁶ but also are more likely to incur disciplinary violations than are inmates without mental illness. Steiner et al.²⁷ systematically reviewed studies of inmate misconduct and found that mental health problems were related to disciplinary problems. Compared with inmates with no mental illness, inmates with mental illness are more likely to have disciplinary problems,²⁸ more likely to complete their entire sentences, more likely to have sanctions imposed for longer durations,²⁹ and more likely to be placed in segregation.³⁰

Crafting a Test

Any test of disciplinary responsibility should provide for the appropriate legal protections. In the United States, this means that the test should protect inmates from Eighth Amendment harms (i.e., from cruel and unusual punishment). California has settled on two assessment questions that can prevent or mitigate such harms (Table 1). The first uses a “so strongly influenced” standard, and it provides for the possibility of dismissing disciplinary violations for inmates who are mentally ill. When this standard is not met, the second question permits evaluators to opine whether a mental illness contributed to the alleged misconduct, and this information may be used to mitigate the severity of penalties.

The “so strongly influenced” standard deserves further comment because its phrasing is, to my knowledge, uncommon. The prison regulation describing the standard reads: If the inmate's behavior was so strongly influenced by symptoms of mental illness or developmental disability/cognitive or adaptive functioning deficits at the time the rules violation occurred, mental health staff may recommend … that the inmate would be better served by having the behavior documented in an alternate manner (Ref. 31, § 3317.1).

In its proposal for the rule change, the California Department of Corrections³² explained “so strongly influenced” to mean “if not for the symptoms of the mental illness … the behavior likely would not have occurred, or would not have occurred with the severity it did” (Ref. 32, p 5). The phrase “if not for” indicates that “so strongly influenced” is a counterfactual test of responsibility: in the absence of a mental illness, the misconduct would not have happened. The proposal tasks clinicians with determining whether the mental illness was the “leading factor” in bringing about the alleged misconduct. (The phrase “or would not have occurred with the severity it did” is peculiar in that it is not a test of responsibility. It grants that mental illness did not cause the misconduct, only that it amplified some feature of the misconduct. Because the phrase cannot be reconciled with the counterfactual “so strongly influenced” test, I set it aside in the current discussion.)

Although the phrase “so strongly influenced” implies degrees of connection, in practice it effectively demarcates the line at which the role of mental illness is substantial enough to be regarded as an exonerating cause of the offense. The upshot is that “so strongly influenced” has the same meaning as “but for” or “without which.” Thus, “so strongly influenced” is the administrative equivalent, in meaning and function, of the Durham rule or product rule, a legal test of insanity that raised considerable controversy during its use in the District of Columbia from 1954 to 1972.^33,34

The Durham rule stated that a defendant was not responsible for a criminal act if that act was the product of a mental illness. In a later ruling, “product of” was clarified to mean “cause” in the counterfactual “but for” sense.³⁵ The relationship between the act and mental illness, “whatever it may be in degree, must be … critical in its effect in respect to the act” (Ref. 35, p 617). Furthermore, according to the ruling, the illness must have been “the decisive difference between doing and not doing the act” (Ref. 35, p 617).

Other notable similarities exist between California's “so strongly influenced” test and the Durham rule. California's prison regulations do not require any specific symptom or capability; evaluators may use any aspect of mental functioning to explain how mental illness caused the offense in question. Similarly, Judge Bazelon, who wrote the majority opinion in Durham, argued that tests of responsibility should not rely narrowly on any one symptom or capability (e.g., knowledge of right and wrong).³⁶ Rather, he sought a “broader test” that permitted psychiatrists to inform the trier of fact fully about the nature of a defendant's mental illness. Judge Bazelon's aim of enlarging the role of psychiatrists is comparable with the aim of judicial intervention in Coleman v. Brown (i.e., to amplify clinical input and its impact on disciplinary outcomes). Finally, the Durham rule and the “so strongly influenced” regulation link a positive finding with exoneration.

Although I have argued that “so strongly influenced” is a synonym for “cause” and is thus comparable with the Durham rule, this novel phrasing poses a challenge to evaluators because no court opinions, legal scholarship, or settled literature exist to guide them on the meaning and application of “so strongly influenced.” On the other hand, forensic mental health professionals are accustomed to making causal analyses regarding past mental states (in cases of criminal responsibility, negligence, etc.).³⁷

Finally, it is noteworthy that California now has two legal tests of responsibility: “so strongly influenced” for disciplinary responsibility and the M'Naughten rule for criminal responsibility. Some could argue that California uses a test of responsibility that is more generous for incarcerated felons than for people in the community who commit the same type of offense. It is possible that a forensic evaluator, given the same set of facts, could find evidence for the role of mental illness using the “so strongly influenced” test but not for the capabilities specified by the M'Naughten rule. On the other hand, perhaps a more generous legal test of disciplinary responsibility is a reasonable remedy to the constitutional violations found in Coleman v. Wilson and to the limited rights afforded to inmates in disciplinary hearings.²¹

Identifying Eligible Inmates

California proactively identifies which inmates might benefit from a mental health evaluation prior to the disciplinary hearing. The screening criteria, presented in Table 2, have two significant features. First, the criteria are an attempt to balance constitutional protections with limited staff resources; not every inmate with mental illness has access to clinical input, but the criteria cast a wide net by using a combination of mental health indicators (e.g., level of care) and types of misconduct that carry the most severe penalties. The intent is to capture not only inmates who have a severe and persistent mental illness but also inmates in the mental health system charged with misconduct that carries serious penalties. Second, the criteria were designed to be objective because reliance on the judgment of correctional officers to recognize signs of mental illness had proven to be inadequate.³⁸

As for the assessment of disciplinary responsibility itself, California's prison regulations do not exclude any mental disorder from consideration. In contrast, the California Penal Code prohibits insanity defenses based on “personality or adjustment disorder, a seizure disorder, or an addiction to, or abuse of, intoxicating substances.”³⁹ Similar exclusions exist in California's Mentally Disordered Offender Act,⁴⁰ a law that permits the involuntary hospitalization of inmates with mental illness upon their release from prison.⁴¹ One criterion for involuntary hospitalization in the act requires that mental illness must have been “one of the causes or was an aggravating factor in” the commission of the offense that resulted in incarceration (Ref. 40, § 2962(d)(1)). The act's definition of mental illness bars “a personality or adjustment disorder, epilepsy, intellectual disability or other developmental disabilities, or addiction to or abuse of intoxicating substances” (Ref. 40, § 2962(a)(2)).

By opting not to specify which disorders should be excluded, California's prison regulations permit prison evaluators to make causal conclusions based on disorders that are not otherwise valid in California for insanity defenses or for qualification as a mentally disordered offender. These include disorders ordinarily not considered to interfere grossly with logical reasoning or behavioral controls (e.g., depressive disorders, anxiety disorders). In the case of substance use problems, the danger exists that inexperienced evaluators may use the science of addiction to negate unintentionally the concepts of free will and responsibility embedded in the law.⁴²

Addressing Professional Dilemmas

Little has been written about how to conduct assessments of disciplinary responsibility or the professional concerns involved. In this section, I discuss three common professional dilemmas: ultimate issue opinions, forensic training, and ethics concerns.

As shown in Table 1, a positive finding of “so strongly influenced” is tied to the recommendation that correctional officials reduce or dismiss the charge. The recommendation is a not-so-covert declaration by evaluators that, in their opinion, the inmate is not responsible for the alleged offense. In other words, California requires ultimate issue opinions on disciplinary responsibility. Although some controversy about ultimate issue opinions exists,^43,44 many jurisdictions permit ultimate issue opinions, but none, so far as I know, require them. On the other hand, an ultimate issue opinion does supply unambiguous information to custody officials, which makes it consistent with California's goal of ensuring that clinical input carries greater weight in disciplinary decisions.

Some have characterized mental health evaluations for disciplinary matters as quasi-forensic, thereby diminishing the importance of forensic expertise.⁴⁵ Broadly speaking, clinicians are acting in a forensic capacity when they apply their expertise to matters of law. California regulations provide for disciplinary hearings and clinical input into these hearings.³¹ Furthermore, the results of mental health evaluations generally, and assessments of disciplinary responsibility specifically, are relied upon in findings of guilt or innocence and inform sanctions. Thus, the evaluations are plainly forensic tasks.

Because assessments of disciplinary responsibility involve many of the same complexities as insanity evaluations, specialized knowledge and training are necessary.⁴⁶ While forensic specialists would be ideal, prospective evaluators, at a minimum, should be familiar with ethics guidelines for forensic practice,^47,48 forensic principles,⁴⁹ and the assessment process,⁵⁰ as well as a basic understanding of the law, related case law,⁵¹ and prison regulations. In California, findings in Coleman v. Wilson highlight the need for specialized training. The special master for the case found that clinicians had difficulty articulating the causal connections and writing in a way that was accessible to hearing officers. Clinicians sometimes provided only diagnostic information, relied on the inmate's version of the rule violation, used clinical jargon, or did not make useful recommendations regarding penalty mitigation.⁵²

In California, the evaluator cannot also be the treating clinician. Although this solves the problem of dual relationships, it does not solve the concern of dual agency: the conflict between advancing the needs of patients versus the needs of the institution.⁵³ Metzner¹⁹ described a hypothetical scenario in which a correctional clinician completes an assessment of an inmate's disciplinary responsibility for assaulting a correctional officer. The dismissal of the offense, he reasoned, has the potential to strain the working relationship between the clinician and the officer and to distort the reputation of mental health staff in general (e.g., viewed as “thug huggers”).

On the other hand, dual agency is not an insoluable problem. For example, the scenario described by Metzner¹⁹ can be overcome in a variety of ways. Knoll²¹ suggested hiring forensic psychologists to conduct all evaluations in-house. Another option is to refer all evaluations to an external forensic team. Even without such resources, there are workarounds. Evaluators should recuse themselves if they have a close relationship with the victim of the misconduct. When this is not possible (e.g., in small facilities), evaluators should seek consultation. They also should articulate carefully the causal nexus between the mental illness and the offense to insulate themselves against perceptions of soft-heartedness for prisoners. Finally, evaluators should be mindful of how urges to protect colleagues from the misconduct of an inmate may impinge upon their objectivity. The institution, not the evaluator, is responsible for protecting staff who are victims of misconduct and, as a result, fear for their safety. A willingness to engage in ethical problem-solving is the only realistic bulwark against the problem of dual agency.