Psychiatric Conditions and the Framework of Responsibility
When should or should not people be held responsible for things they do under the influence of psychiatric disorder?
Questions about responsibility are often legal. Should this person be punished for an illegal act? Were there factors that diminish his responsibility or even absolve him altogether? The questions are often moral, often linked to praise or blame: If her terrible record of not meeting deadlines reflects her psychiatric problem, isn’t it unfair to blame her?
Other questions about responsibility do not reflect the external standpoint of the law or the moral critic. People wonder about themselves. Should I feel bad about what I did? How far am I a free agent? How far am I in charge of my own life? For some people with psychiatric problems these questions can be even harder to answer than for the rest of us. And the answers often matter. When coping with some psychiatric disorders, it may be crucial not to take a fatalist view of your own life.
It is extraordinary that, extended but otherwise almost unchanged, Aristotle’s analysis of voluntary action still forms the central framework of thought about the conditions of moral and legal responsibility. The framework’s persistence for more than two millennia suggests that it reflects something deeply rooted in our assessments of people and what they do.
Aristotle was giving an account of what makes actions voluntary, though he wisely approached it mainly by saying what makes an action involuntary. He said that voluntary actions attract praise and blame, while involuntary ones attract pardon and sometimes pity. So his account of involuntary actions is about how to decide when someone is not responsible for an action. The framework is simple. Involuntary actions are those done either in ignorance or under compulsion.1
To excuse an action, ignorance must be specific and relevant. Aristotle cites not knowing that the thing blurted out was a secret, mistaking a son for an enemy, or thinking the poison was lifesaving medicine. He rejects many kinds of ignorance as excuses. We do not excuse normal people because they do not know who they are, or because they do not know they are killing someone with a knife. Of these things “no one could be ignorant unless he were mad.” Aristotle leaves dangling how we should respond to mad people who are deluded in these ways. But later development of the framework rightly allows these delusions as excuses.
Aristotle cites as central cases of compulsion being blown off course by a gale, or being carried off by men who have you in their power. He says that an act under compulsion is one “of which the moving principle is outside” and where “nothing is contributed by the person who is acting.”
This account of the totally passive person has to be stretched a bit to fit being forced to do things. Being forced may be coercion (the tyrant who threatens to kill your family) or being forced by a storm to throw goods overboard in order to survive. A forced act is in one way voluntary: you chose to do it. But in another way it is not: “no one would choose any such act in itself.” Aristotle thinks some acts are so terrible that no coercion excuses them: you should accept death or any other catastrophe rather than kill your mother. But according to how the circumstances and choices are assessed, being forced is a candidate for the compulsion that excuses from responsibility.
It is striking how external Aristotle’s version of compulsion is. His view of madness seems limited to delusions (though his later discussion of weakness of will is highly relevant to addiction). His account of voluntariness leaves out the “inner” constraints of modern psychiatry: the addictive, the obsessive-compulsive, and so on. These are the main problem for his view that all acts whose “moving principle” comes from within are entirely voluntary.
He does consider the suggestion that we can be compulsively attracted, for instance to pleasure. But he rejects it, as pleasure is such a common motive that this could make all acts involuntary. And he objects that pleasure as a moving principle is internal: “It is absurd to make external circumstances responsible, and not oneself, as being easily caught by such attractions.” He concludes that compelled actions are those “whose moving principle is outside, the person compelled contributing nothing.”
The conflict between Aristotle and the modern view of drug addiction or compulsive handwashing is about whether inner constraints exist. The desire to take heroin or to repeatedly wash one’s hands is internal to the person in a way that the gale at sea is not. We either have to hold the addict and the obsessive-compulsive fully responsible or else depart from Aristotle. The majority modern view departs from his view. It allows internal constraints to excuse people, either partly or entirely. But as Aristotle suggests, we need some limit to this loophole if it is not to excuse all actions.
Apart from the problem of internal constraints, Aristotle’s broad outline is still the central framework of responsibility. The M’Naghten rules, drawn up to guide English law in the mid-nineteenth century, reflected the “ignorance” side of the framework, but not yet the “compulsion” side. In this, the jury in M’Naghten’s case were ahead of the law.
Daniel M’Naghten was paranoid about the Tory party and the police. Sir Robert Peel, Tory prime minister and founder of the Metropolitan Police, was an obvious focus for him. In 1843 M’Naghten fatally shot in the back the prime minister’s private secretary, probably in mistake for Peel. He said, “The Tories in my native city have compelled me to do this. They follow and persecute me wherever I go, and have entirely destroyed my peace of mind. They followed me to France, into Scotland, and all over England … They have accused me of crimes of which I am not guilty; in fact they wish to murder me. It can be proved by evidence. That’s all I have to say.”2
At his trial witnesses testified that M’Naghten was deluded. The defense argued that this took away “all power of self-control.” The judge’s summing-up centered not on self-control but on knowledge. Had M’Naghten at the time known that what he did was “a violation of the law of God or of man”? If he had been capable of distinguishing between right and wrong, he was a responsible agent and liable to all legal penalties. The jury, guided perhaps more by instincts of fairness and humanity than by the letter of the law, brought in a special verdict. M’Naghten went to Bethlem Royal Hospital and then to Broadmoor, where he spent the rest of his life.
The House of Lords, concerned about the lack of a “guilty” verdict, posed questions to the judges about the criteria to be applied in considering the responsibility of people acting under insane delusions. The resulting “M’Naghten rules” are essentially summed up in this passage from the majority response: “To establish a defence on the ground of insanity, it must be clearly proved that, at the time of committing the act, the party accused was labouring under such a defect of reason, from disease of the mind, as not to know the nature and quality of the act he was doing; or, if he did know it, he did not know he was doing what was wrong.” Like Aristotle, the judges saw relevant ignorance as an excuse but, again like Aristotle, did not see the possibility of psychiatric conditions creating inner compulsions.
Some problems lurking in the M’Naghten rules were about “knowing” an act is wrong. Is wrongness the same as illegality, as some judges seemed hopefully to think? If not, there are also questions about resting the law on the debatable belief in objective truths about right and wrong. Are moral truths what “the reasonable man” believes? Or are they tied to religious belief, as the phrase “the law of God or of man” seems to suggest?
Another problem about “wrong” was raised in 1863 by the case of Townley, who killed his fiancée when she broke off their engagement. He was paranoid, saying six people were conspiring to kill him. He showed no remorse. He said he had as much right to deal with his fiancée as with any other property of his, and that he recognized no man’s right to judge him. He seemed to know his act was illegal. The M’Naghten question about “knowing what he was doing was wrong” did not anticipate that someone might know it was both illegal and widely morally condemned, yet sincerely think he had a moral right to do it.
The major defect of the M’Naghten rules was not seeing that mental disorder might make some things very hard or even impossible to resist. This was unfair to accused people who acted under the influence of such disorders. But allowing irresistible impulses raised a large problem, as James Fitzjames Stephen pointed out in 1855: “I fear there is a disposition to confound them with unresisted impulses.” Over a century later the social scientist Barbara Wootton echoed this: “It must not be inferred that an impulse was irresistible merely because it was not resisted.”3
One way of giving greater flexibility (though without solving the hard question anticipated by Aristotle and pressed by Stephen and Wootton) was the introduction in English law of the concept of “diminished responsibility.” But this was not until the 1957 Homicide Act, more than a century after the M’Naghten rules. During that previous century there was pressure on both legislators and judges to find ways to mitigate the injustices coming from the rules’ rigidity.
The rigidity was challenged by cases where a mother killed her baby under the influence of what would now in Britain be called postnatal psychosis (in the United States, postpartum psychosis). After childbirth women are at most risk of mental disorder. There are huge hormonal changes in pregnancy, some hormones increasing two-hundred-fold, followed by a rapid decline over the twenty-four hours after birth. According to modern studies, postnatal psychosis is characterized by agitation, huge mood shifts between mania and depression, a delirious state, delusions, confusion, and thought disorder. Dr. Margaret Spinelli points out that neuropsychiatric tests indicate cognitive impairments.4 There is a 5 percent risk of suicide and a 4 percent risk of infanticide.5
In England the dark ages of the legal response to infanticide started to close after 1849, the year Rebecca Smith was hanged despite the jury having recommended mercy. She was the last mother to be executed in England for killing her baby. At that time, without epidemiological studies or neuroscience, many people already had quite a good idea of the pressures of postnatal psychiatric problems. Some personally knew women who had these problems, or heard what emerged in highly publicized trials. Where juries, with the support of the judge, recommended mercy toward a woman they convicted, she was not executed. Eventually all mothers committing infanticide were reprieved. Attempts were made to fit the law to this practice. For a long time they failed.
Success came (thirty years after most of the rest of Europe) with the passing in 1922 of the Infanticide Act. This allowed that a woman who had killed her newly born child before she had “fully recovered from the effect of giving birth to such child” could be convicted of the noncapital crime of infanticide rather than murder. This gave judges wide scope in sentencing. The punishment could be as little as a conditional discharge. Soon the standard sentence was either committal to a mental hospital or else probation on condition of accepting psychiatric treatment.
The Infanticide Act was recognition of diminished responsibility without the phrase. Yes, she did kill her baby and sometimes knew she was doing so. But her condition, with its delusions and desperation, was either an excuse or at the very least a strong mitigation. One kind of injustice and cruelty was starting to be defeated. The central framework of responsibility remained in place: only relevant ignorance or appropriately strong compulsion excuse someone from responsibility for an act. But the framework was extended to allow for degrees of inner compulsion. The boundary problems this change leaves us with are a price well worth paying for its humanity.
Caroline Beale was 30 in 1994, living in London with her boyfriend Paul and working in the Department of Health. She was very distressed by the death from cancer of her close friend Alison, the girlfriend of one of Paul’s brothers. Alison had always wanted children and was devastated that she would not live to have them. At the same time Caroline found she was pregnant. Feeling it would be cruel to tell Alison, she kept her pregnancy secret from everyone.
After Alison’s death, Paul found Caroline changed and their relationship was under strain. She had “visions” of Alison. After Alison’s death she identified her own baby with Alison and believed the baby was dead too. Her great stress was noticed at work and she was offered compassionate leave. Having still told no one about her nearly nine-month pregnancy, she went to New York with Paul and his brothers. Later she said, “I couldn’t tell anyone. It wasn’t that Paul and I weren’t speaking but we were going through a bit of a rocky patch … I didn’t really want to go to New York … but I just went along with it. I didn’t know what to do. And so I found myself sitting on the underground going to Heathrow and not being able to tell anybody what was happening.”6
Caroline felt in a daze or a dream the whole time. On the last night in New York, she went to bed while the others went out. In the bath she gave birth. Since Alison’s death, she had thought her baby was dead in the womb. In the bath, “I think I might have passed out. The baby’s head was coming out. I knew she wasn’t alive. Then I was awake again. I was so scared because I thought I was going to die as well. I just didn’t know what to do. There were scissors on the sink so I just cut the cord with scissors and left her in the bath while I went to the loo … I was like on automatic pilot. I took her out of the bath and I didn’t know what to do. I put her in a bag because I didn’t know what else to do. I had to get her back to England, somehow.”7 The next day she carried the baby in the bag to the airport. A security guard found the baby and she was arrested.
The autopsy said the baby had taken a breath. Caroline was charged with murder. She was not granted bail, but was remanded to Rikers Island prison. In a diary she wrote, “The loneliness, the fear, the loss of self dignity, the hunger (and that really does exist). The worst thing of all is going to bed hungry and drinking water to fill yourself up. I think if a judge or DA was to spend a month here they would certainly be more hesitant about incarcerating people in this place. There must be some other way of doing things.”8 She had no access to treatment.
In England, Caroline’s father had been put in touch with Dr. Channi Kumar at the Maudsley Hospital. Help was also offered by Professor Ian Brockington of Birmingham University, who went to New York to interview Caroline. Channi Kumar contacted a New York colleague, Dr. Meg Spinelli (the Margaret Spinelli whose paper “Maternal Infanticide Associated with Mental Illness” is cited in this chapter’s notes). She was instrumental in helping Caroline obtain her defense lawyer, Michael Dowd, who negotiated terms with the district attorney, Marjorie Fisher, for release on bail. While on bail Caroline said she would kill herself rather than go back to Rikers Island.
Ian Brockington had found that leading experts were very skeptical about the autopsy claim that the hemorrhages proved suffocation rather than some injury during delivery. He, Channi Kumar, and Michael Dowd also assembled a powerful team of psychiatric experts for the defense.
On Caroline’s condition, the psychiatrists were divided. Dr. Naomi Goldstein, interviewing Caroline for the prosecution, accepted that she had clinical depression and impaired judgment. But she did not think that at the time of the act Caroline had the psychosis needed for an insanity defense.
The defense psychiatrists, while differing on diagnosis, agreed she was not responsible. After spending time with her, Meg Spinelli had concluded that Caroline did not have postnatal depression but that she had had visual and auditory hallucinations and had been in denial about her pregnancy. Carrying the dead baby through the airport was a sign of acute psychosis. “In my judgement she lacked capacity to form intent by virtue of a defect in her ability to reason and should therefore be found innocent of charges under the insanity defense.”9 Channi Kumar interviewed Caroline twice, and diagnosed a major depressive disorder, linked to a pathological reaction to Alison’s death. He believed that Caroline’s disorder caused what she did. Criminal prosecution was inappropriate. He offered to treat her serious psychiatric illness at the Maudsley Hospital.
Michael Dowd was worried that in Queens a mainly Catholic jury might be prejudiced against anyone accused of harming a baby. He worried Caroline would not stand up well to a long trial. These concerns, and the risk of a very long sentence, made him support plea bargaining. At first, in exchange for a guilty plea, Marjorie Fisher would accept a sentence of from two to four and a half years. The final offer was that if Caroline pleaded guilty to manslaughter, she could receive a sentence that would not exceed the time she had already served in prison and so could return to England.
Because Caroline had believed her baby was dead, pleading guilty to manslaughter was a terrible and humiliating thing to have to do. She agreed to do it because the alternative risked so much. She was released and flew back to England. She had treatment and support from Channi Kumar and others at the Maudsley and after a storm of media attention started to get back to her life.
This tragedy has many layers. Caroline Beale was drawn into denial and concealment of the pregnancy by not wanting to hurt Alison. She found herself trapped in the deception, and invaded by a strange psychiatric condition triggered by the death of her friend. Then came the trauma of what she perhaps rightly thought was a stillbirth, with no one to turn to. It was followed by her “automatic pilot” state and the bizarre project of taking the dead baby to England. Then came the further trauma of arrest.
The next layers of the tragedy involve questionable decisions of others. Someone better trained might have produced a less overconfident autopsy report. A legal system that saw infanticide as the rest of the Western world does would not have brought a murder charge. In England people were shocked that the district attorney, Marjorie Fisher, made choices leading to Caroline’s being confined in a hellhole of a prison and denied psychiatric treatment. The attorney became something of a hated figure. Though some of the criticism was justified, this response is too shallow.
Marjorie Fisher knew about the criticism in England. Asked about her decisions, she said, “In the United States you don’t get extra points for killing babies.” She also said, “A system that takes into account the life of a child is a very compassionate one. Sometimes I got the impression that people were more concerned about the defendant than about the victim … She hid it, she concealed it. To me that merited a murder charge.” And she said, “I was satisfied with the sentence and I was most satisfied that she stood up and said what she did. She said that she knew that what she did at the time could end the child’s life. She knew that and the people in Queens and the people in England knew that.”10
Of these remarks, the most notorious was the suggestion that jurisdictions outside the United States give “extra points for killing babies.” Perhaps a district attorney need not know about the development of law in other jurisdictions. Her ignorance that the special laws about infanticide come from concern about diminished responsibility, and not from a failure to “take into account the life of a child,” is perhaps excusable. (Though the ignorance was aggressive: “extra points” suggests positive enthusiasm for a baby’s death.) The other comments are weak. To the psychiatrically informed or the humanly imaginative, “She hid it, she concealed it” does not make murder more likely than that the concealment came from delusion, denial, and panic. And “she stood up and said what she did” is entirely undermined as evidence for her having committed murder by the system of plea bargaining.
The main issue goes deeper than the district attorney’s decisions. It is about laws that, following the M’Naghten rules, are concerned only with whether a woman knew what she was doing, to the exclusion of issues about psychiatric conditions creating inner compulsions. This legal defect is the outer layer of the tragedy, making it very hard to avoid the kind of injustice and cruelty seen in this case.
This was demonstrated seven years later in Houston, Texas. In 2001 Andrea Yates drowned her five children in the bath. She was a devoted mother, whose first psychotic episode was after the birth of her first child. “She hid the episode in the hope that this would protect them from being harmed by Satan.” After her fourth child was born, she attempted suicide twice while trying to resist “satanic commands” to kill her baby. When her fifth child was six months, she was “catatonic” and “like a caged animal.” She was twice confined in a psychiatric hospital. When her medication was stopped, she obeyed Satan’s command to save her children from hell by killing them. She said she was Satan.
Again, the story is a multiple tragedy. The horrifying tragedy of the five children killed, and by their own mother. And the tragedy of Andrea Yates herself: of her tormenting insanity and the indelibly horrible thing it made her do.
The tragedy has another dimension. Andrea Yates was charged with capital murder. Her plea was innocent by reason of insanity. The State of Texas still used the M’Naghten tests. Did she know the nature and quality of her act? Did she know it was wrong? The prosecution argued that she knew right from wrong because she called 911 and her husband after the killings. The jury found her guilty of capital murder but substituted life imprisonment for the death sentence the prosecution demanded.11
The cases of Caroline Beale and Andrea Yates show that the modern version of the central framework of responsibility has not percolated everywhere. In some places, more than a century and a half after the hanging of Rebecca Smith, legal failure to recognize internal compulsions still multiplies their cruel consequences.
Including internal compulsions was a great advance. Backward jurisdictions still need to be persuaded of this, and there is still intellectual work to be done. There is a need for a finer-grained analysis of what counts as excusing internal constraint. This needs to be seen in the context of particular psychiatric conditions.
The continuing dominance of the central framework of Aristotle’s two excuses, ignorance and compulsion (in modified form), reflects the way actions are guided by beliefs and motivated by desires.
It also reflects links between judgments of responsibility and how we view motives, values, and character. If someone knowingly steals money, this reflects on his honesty. This is not so if he just picks up a pile of documents, without seeing the paper money among them. If a bank clerk hands over the key to the safe because the robber demanding it is armed and dangerous, this reflects fear, not complicity in the crime. What matters is motivation, which tells us something about the person’s values and character. Pleas of ignorance (“He did not see the money”) or of compulsion (“She was forced to hand over the key”) defend a person’s character. They defeat accusations of being a thief or an abbettor.
The psychiatric versions of the two pleas—ignorance and compulsion—raise problems of their own. A delusion may stop someone from knowing what he is doing. But it is often hard to tell how far someone really is deluded. And the strangeness of delusions can make them hard to understand. Psychiatric states may involve compulsion, but the “inner” compulsion of addiction is harder to assess than the external kind used by the armed robber.
Morally blaming people is bound up with the belief that what they did reflects badly on them. The excuses of the central framework, ignorance and compulsion, sever the link between action and character, and so prevent the person’s character being stained by the action.
These excuses, with complications, apply in many psychiatric cases. (Though where the person’s character itself has defects caused by the psychiatric disorder, the problems are much more acute.)
Blaming someone, judging their character, can have the unemotional detachment of the recording angel. But there is another view of blame that makes it less detached by locating it among people’s relationships and attitudes.
In his classic paper “Freedom and Resentment,” P. F. Strawson argued against those who see determinism as undermining blame. He suggested that they overlook the way blame is embedded in a network of “interpersonal reactive attitudes” such as gratitude and resentment. He argued that blame cannot be separated from these other attitudes—that they stand or fall together. And, he argued, determinism could not justify giving up this whole network of attitudes. It would be psychologically impossible for us to give them up. Even if it were possible, it would not be rational to do so. These attitudes are at the core of human relationships. Giving them up would impoverish us. It would replace our present emotionally involved reactions to each other with the clinical objectivity of a psychiatrist or a social worker.12
This line of thought might not succeed in defusing the issue of determinism. (Is it too hasty to see the reactive attitudes as an indivisible web? Will undermining blame really make all relationships so cold and detached?) But one merit of this approach is that it makes questions about blame less abstract, shifting the focus to our reactions and attitudes to the person.
Relationships are shaped by people’s attitudes toward each other. Some attitudes pervade a whole relationship: liking, disliking, love, hatred, friendship, and enmity. These attitudes are often responses to what the other person does. But not always. I may not deserve the dislike of one person or the love of another, yet have both.
Some attitudes are less free-floating because they are based on evaluation. Respect, admiration, and contempt claim to reflect someone’s good or bad qualities. Some attitudes reflect people’s good or bad luck. If I break a leg trying to windsurf, I may trigger sympathy, gloating, or amusement. Other attitudes are reflexive. Jealousy, envy, and condescension all compare the other person with oneself. All of these are “interpersonal reactive attitudes.” Each may be, and some must be, a reaction to what other people are like. But without further layers, they are limited.
Relationships include interplay between attitudes to each other. We have attitude spirals. I have attitudes to your attitude to me. You have attitudes to my attitudes to your attitudes. Lovers or friends respond to each other’s attractiveness or wit. But the relationship is thin without reactions to each other’s love or trust. Enemies may dislike each other’s faces or voices, but the enmity is shallow unless they mind each other’s contempt or gloating.
Relationships go deeper as we develop attitude spirals. Some attitudes are centrally responsive to the attitudes of others. At this level are gratitude, indignation, vengefulness, blame, and forgiveness. (These need not be in response to attitudes to ourselves. We might be indignant on someone else’s behalf, or blame a person for generalized malevolence.) At least mainly, we see in actions the attitudes that shine through them. These attitudes to attitudes are at the core of human relationships. For those who know people whose psychiatric conditions contribute to objectionable behavior, there is a possibly insoluble dilemma. Blaming them may be unfair. But just as the reactive responses are at the core of relationships, exclusion from these responses is exclusion from part of that core.