Pleading the Fifth

In 1637, John Lilburne was accused with the crime of shipping seditious books into England from Holland.  Forced to appear in the now-infamous Court of Star Chamber — so-named because of the stars painted on the roof of the room in which its proceedings were held — Lilburne refused to swear the “oath ex officio.”  That refusal was intimately related to the ultimate abolition of the Court of Star Chamber by the English Parliament in 1641 and to the development of one of the most important modern legal rights, the right against self-incrimination.

John Lilburne

Derived from ecclesiastical Courts of Inquisition, the oath ex officio compelled an individual to answer all questions put to him truthfully, and was typically administered to those about to be accused of a crime but before being advised of the nature and scope of the charges.  Because it was complete in scope, requiring a truthful answer to any question, it became a convenient mechanism for forcing self-incriminating testimony, and was increasingly abused by the Court of Star Chamber as its proceedings became ever more secret and politically oppressive.

Like so many of the men and women who prompted the development of rights that we (perhaps too complacently) now take for granted, Lilburne was, by all accounts, an argumentative and combative man.  One of his contemporaries commented that “if John Lilburn were the last man in the world, John would fight with Lilburne, and Lilburne would fight with John.”  But this characteristic of steadfast defiance was instrumental in effecting the ultimate abolition of the Court.

After several appearances in which he refused to swear the oath, Lilburn was sentenced to a fine of £500, punishment in the pillory, and imprisonment until he acquiesced.  He was whipped more than 200 times in the street on his way to the pillory, where he harangued the gathering crowds that no one should be forced to accuse himself of a crime.  During the time of his imprisonment, which included at least one four-month period of solitary confinement, he wrote nine pamphlets.  As part of these efforts to compel him to swear the oath — without even yet addressing his alleged crime of importing seditious books — Lilburne spent nearly three years imprisoned.  It was only finally when the Long Parliament met near the end of 1640 and was inspired by a speech delivered by Oliver Cromwell that action was taken to release victims of the Court of Star Chamber’s oppression, including Lilburne.

Lilburne’s example was one of several that served to develop the modern concepts of a presumption of innocence, the right not to incriminate oneself, and the recognition that the refusal to answer an incriminating question carries no implication of guilt.  For hundreds of years now, these have formed part of the bedrock of criminal law and we are all too aware of the potential for oppression by the state when they are relaxed.  In the United States, the right not to incriminate oneself is, of course, enshrouded in the Fifth Amendment of the Bill of Rights:  “No person … shall be compelled in any criminal case to be a witness against himself.”

That right was implicated in a case in Colorado this week that is interesting because of its involvement of modern electronics technology.  The Fifth Amendment right is being asserted by Ramona Fricosu, who was indicted in 2010 on charges arising from allegedly fraudulent real-estate transactions.  As part of its investigation, the government seized, under warrant, six computers from her home, one of which is a laptop whose contents are encrypted.  Suspecting that the encrypted contents contain information relevant to its investigation and likely to be incriminating, the government has sought to compel Fricosu to type her password into the laptop so that the contents can be read.  She has refused.

This week, the U.S. District Court for the District of Colorado ordered that Fricosu supply an unencrypted copy of the contents of her laptop to the government by February 21, 2012, presumably by typing her password into the device.  A copy of the order can be read here.  If Fricosu continues to refuse, she may be punished for contempt of the court’s order.

There is no doubt that modern digital encryption techniques have an important and legitimate function.  The very existence of electronic commerce depends fundamentally on the use of such techniques, and there have been increasing recommendations issued by electronic-security professionals to routinely encrypt the content of electronic devices as a precaution against loss or theft.  Sensitive personal and business information is now commonly stored on such devices, and encryption is the only effective mechanism for preventing unauthorized access to the information.

Fricosu’s case is, in some ways, a difficult one.  This is not an instance in which the government has an inchoate suspicion that there is something incriminating on the laptop, but rather one in which there is sufficient evidence to support the probable cause for issuing a warrant.  It is therefore easy to understand the court’s order and to have sympathy with the desire by police simply to read what they have legitimately seized as part of their investigation into a crime that defrauded others.

But at the same time, I have considerable discomfort with the order precisely because of the Fifth Amendment implications.  It is a mistake to view this case purely through the procedural lens of the Fourth Amendment and to be satisfied that the police have complied with all warrant requirements.  History has taught us of the dangers of vesting the state with the power to compel self-incrimination by individuals.  The potential for oppression is sufficiently great that free societies have determined that it is better not to vest the state with that power.  There may be some cases in which the burden on the government to prove guilt is greater, and some even in which guilty persons are not convicted because the burden is too great to satisfy.

But this is the cost we have agreed to pay for maintaining a barrier against abuses that we have recognized as being too seductive to governments in the past.

A Mean Act of Revenge Upon Lifeless Clay

Jack Kevorkian died today, and many are commenting about his role in the “right to die” movement.  While I am a supporter of the movement generally, I did not find Kevorkian to be a courageous man.  His actions had a significant detrimental impact on the efforts of others to provide ways for physicians to aid the terminally ill to end their lives on their own terms and with dignity. 

Consider for a moment the case of Diane, and imagine the circumstances she found herself in.  She had been raised by an alcoholic family when she was a child and had suffered a great number of torments in her life, including vaginal cancer as a young woman, clinical depression, and her own alcoholism.  When her physician diagnosed her with myelomonocytic leukemia, she was presented with the options:  She could proceed without treatment and survive for a few weeks or perhaps even a few months if she was lucky, but the last days of her life would surely be spent in pain and without dignity; it was not how she wanted her friends and family to remember her.  If she accepted the treatment her doctor had discussed, there was a 25% change of long-term survival, but the treatment itself — chemotherapy, bone marrow transplantation, irradiation — would also rob her of much of what she valued about life, and would likely result in as much pain as doing nothing.  For her, the 25% chance that such treatment would succeed was not worth it.  Others might have differed in their assessment, but this was hers. 

Neither option presented to her — let the disease run its course or accept a treatment she had rejected — was acceptable, and so she considered the unspoken alternative.  Diane’s physician told her of the Hemlock Society, even knowing that he could be subject to criminal prosecution and professional review, potentially losing his license to practice medicine.  But by having a physician who knew her involved in her decision, her mental state could be assessed to ensure that it was well-considered and not a result of overwhelming despair.  Her physician could explain how to use the drugs he prescribed — ostensibly to help her sleep — so that until the time came, she could live her life with confidence that she had control over when to end it.  She could enjoy the short time she had remaining without being haunted by fears that it would be ineffective or result in any number of consequences she did not want.  In the end, Diane died alone, without her husband or her son at her side, and without her physician there.  She did it alone so that she could protect all of them, but died in the way that she herself chose. 

The story of Diane is one that her physician, Dr. Timothy Quill, published in the New England Journal of Medicine in 1991.  A copy of it can be found here.  It was one of the first public accounts of a physician acknowledging that he had aided a patient in taking her own life.  It was to prompt a debate about the role of physicians at the end of life, and a subsequent study published by the same journal in 1996 found that about 20% of physicians in the United States had knowingly and intentionally prescribed medication to hasten their patients’ deaths. 

But the quiet, thoughtful, and sober approach adopted by Quill and many other physicians to the issue of physician-assisted suicide was very much derailed by the grandstanding antics of Kevorkian.  His theatrical flouting of the law, prompting law-enforcement agencies to act in making an example of him rather than seriously considering the merits of his views, were counterproductive to the medical debate. 

Kevorkian’s fascination with death was long part of his life.  He was not, as many believe, christened with the nickname “Dr. Death” because of his efforts promoting physician-assisted suicide.  That happened long before, during the 1950’s shortly after receiving his medical degree.  While a resident at the University of Michigan hospital, he photographed the eyes of terminally ill patients, ostensibly to identify the actual moment of death as a diagnostic method, but more truly “because it was interesting [and] a taboo subject.”  Later, he presented a paper to the American Association for the Advancement of Science advocating “terminal human experimentation” on condemned convicts before they were executed.  Another of his proposals was to euthanize death-row inmates so that their organs could be harvested for transplantation. 

His views have politely been described as “controversial,” but are perhaps more accurately considered gruesome and bizarre, such as his experiments aimed at transfusing blood from corpses into injured soldiers when other sources of blood were unavailable.  The result of his various investigations was considerable professional damage, causing him to resign or be dismissed from a number of medical centers and hospitals.  His own clinic failed as a business.  For all his current notoriety, Kevorkian was throughout his career considered very much an outsider to the mainstream medical-science community. 

In considering the legacy of Kevorkian, it is important to recognize the long history of the debate over physician-assisted suicide, which dates at least from the days of ancient Greece and Rome.  The modern debate in the United States has its origins in the development of modern anaesthesia.  The first surgeon to use ether as an anaesthetic, J.C. Warren, suggested it could be used “in mitigating the agonies of death.”  In 1870, the nonphysician Samuel D. Williams suggested the use of chloroform and other medications not just to relieve the pain of dying, but to spare a patient that pain completely by ending his life.  Although the proposal was made by a relatively obscure person, it attracted attention, being quoted and discussed in prominent journals and prompting significant discussion within the medical profession.  The various discussions culminated in a formal attempt to legalize physician-assisted suicide in Ohio in 1906, although the act was rejected by the legislature in a vote of 79 to 23. 

Today, there are three states that have legalized the practice of physician-assisted suicide — Oregon, Washington, and Montana.  The history of how that legislation came to pass, and the various court challenges that have been raised, is fascinating in its own right.  For now, suffice it to say that my own view is that those states legalized the practice because of the courageous efforts of physicians who are largely unknown, not because of the actions of Kevorkian.  Indeed their courage is all the greater that they achieved as much as they did despite his activities.

A Mere Heart of Stone

The fictional legal trials I watched on television as a boy were gripping drama.  It was almost a given that there would be some irregularity in procedure — a last-minute witness would be called out of order, perhaps, and give testimony that would give a completely new perspective on everything that had happened before.  Or a juror might be called to the witness stand and compelled to testify over the objections of the defendant’s attorneys.  Or maybe one of the attorneys himself might be called to the stand.  These events would always be accompanied by a sober judge intoning that it was highly irregular but that in the interests of justice — always in the interests of justice — he “would allow it.” 

Those sorts of things never happen in real-life trials.  At least not very often. 

One of the most dramatic events in a trial was surely when Clarence Darrow called prosecuting counsel William Jennings Bryan to the witness stand in the famous trial of State of Tennessee v. Scopes.  It was to be a battle of wits between two of the most famous of American attorneys as Darrow quizzed Bryan about his interpretation of the Bible and whether everything in it should be interpreted literally — they covered Jonah being swallowed by the whale (while disagreeing over whether it was a whale or a big fish), Joshua causing the sun to stand still at Jericho, and the date of the great flood.  Darrow pressed Bryan on the inconsistency between what is taught in the bible and the scientific discoveries of humanity.  It was truly gripping stuff and reports exist of people from miles around Dayton, Tennessee pressing into the courtroom to witness the drama. 

In the end, Judge John T. Raulston ordered the whole session — which had taken place outside the view of the jury — expunged from the record.  The result of the Scopes Monkey Trial is well known.  Scopes was found guilty of teaching evolution to schoolchildren in violation of the Butler Act and fined $100.  His conviction would ultimately be reversed and dismissed by the Tennessee Supreme Court on technical grounds — rather than constitutional grounds — with the weary comment that “Nothing is to be gained by prolonging the life of this bizarre case.”  The Butler Act, forbidding the teaching that Man evolved from “lower orders of animals,” would remain valid law in Tennessee until it was repealed in 1967. 

It is interesting that Tennessee has again become a sort of battleground over the teaching of evolution to schoolchildren. It is being watched by those who take an interest in the teaching of evolution and creationism because of bills that are currently pending in that state’s legislature.  Those bills are similar to bills that have been defeated in other states like Oklahoma and New Mexico, and represent a new strategy by those who wish to have creationism taught as science to American schoolchildren.  Other tactics, such as directly challenging the exclusion of creationism from scientific curricula or couching creationism as “creation science” or “intelligent design” have failed. 

The new approach is to use scientists’ love of critical thinking as a wedge to insist that teachers not be prohibited from “helping students understand, analyze, critique, and review in an objective manner the scientific strengths and scientific weaknesses of existing scientific theories covered in the course being taught.”  This is to be done to further the “purpose of science education … to inform students about scientific evidence and to help students develop critical thinking skills necessary to become intelligent, productive, and scientifically informed citizens.”  All good stuff.  A copy of the house bill can be read here; the senate version is essentially identical.

Critics of the bill point to the fact that the examples of controversies given in the bill — namely “biological evolution, the chemical origins of life, global warming, and human cloning” — are not, in fact, subject to significant scientific controversy.  This is certainly true at the level at which such subjects might be addressed to childen.  The bill is seen as dangerous to the integrity of scientific education because it may provide a license to creationist teachers to proselytize contrived and misleading criticisms of dominant scientific theories under the guise of science.  This is a legitimate concern. 

But is such a bill really so troublesome?  It includes provisions intended to appease scientists, such as by insisting that it “shall not be construed to promote any religious or non-religious doctrine, promote discrimination for or against a particular set of religious beliefs or non-beliefs, or promote discrimination for or against religion or non-religion.” 

It is possible that I am naïve, but I don’t particularly fear this type of legislation, even as I recognize the strategic and tactical motivations pressing for it.  It is clearly another attempt by creationists to infiltrate scientific teaching to children in accordance with the Discovery Institute’s infamous “wedge strategy.”  A copy of the leaked 1999 document describing that strategy is available here.  But it is not a deceptive “equal time” provision as has been attempted in the past to compel the teaching of nonscientific ideas in a science class.  And teachers already have a great deal of discretion and influence in classrooms.  Even with those parts of a curriculum they are mandated to teach, they undoubtedly affect students’ perceptions by their own demeanor and subtle expressions of opinion and skepticism.  It is unrealistic even to attempt to regulate every nuance of a teacher’s presentation. 

Doesn’t there come a point when scientists are ready to say “Bring it on”?  I have confidence in the scientific method and confidence that views developed using that method will prevail.  And part of the scientific method is to allow criticism so that it can be exposed and defeated when it is wrong.  There is no doubt that past strategies by creationists were antithetical to science; I just have trouble seeing exactly how this one is.

Rule, Supremacy, and Sway

When the Oklahoma State Trooper came upon the scene on November 13, 1974, it seemed apparent enough that the 28-year-old woman had died in a sleeping-driver accident.  There were no other automobiles involved and the pattern of evidence was classic. 

But the facts surrounding the death of Karen Silkwood were suspicious.  Her blood contained 0.35 mg / mL of methaqualone (Quaalude), roughly double the level needed to induce drowsiness.  While it might have been possible to dismiss the accident as resulting from a young woman in the 1970’s ingesting too much of a common-enough recreational drug, there were too many other contextual factors over the preceding week for such a conclusion to be readily accepted. 

When she died, Silkwood had been returning from a meeting of the Oil, Chemical, and Atomic Workers’ Union, and was on her way to meet a journalist.  It is believed that she had been gathering evidence on behalf of her union to support a claim that her employer — the Kerr-McGee plutonium fuels production plant — was engaging in negligent safety practices.  She was by then known as somewhat of a troublemaker.  Already earlier that summer, she had testified about improper safety practices at the company to the Atomic Energy Commission. 

On three distinct occasions during the week before her accident, Silkwood found herself contaminated by plutonium.  The incidents required not only that she undergo decontamination procedures but that her apartment and roommate also be subjected to them.  Many of her belongings needed to be destroyed during the course of the decontamination procedures.  The source of the plutonium contamination remains unclear.  Many believe that Silkwood was a victim of retaliation for her whistle-blowing efforts, while others claimed that she had deliberately contaminated herself as part of an orchestrated effort to generate negative publicity for the company. 

After her death, Silkwood’s father brought an action against the plutonium plant, during which a jury that considered the detailed facts rejected the theory that she had deliberately contaminated herself.  Indeed, the jury awarded $505,000 in compensatory damages and $10 million in punitive damages under Oklahoma state law.  But that decision was not to be the end of the case.  There was an important conflict between federal and state law that ultimately needed to be resolved by the Supreme Court of the United States. 

Under the federal system of the United States, the Supremacy Clause of the U.S. Constitution requires that “the laws of the United States … shall be the supreme law of the land … anything in the constitutions or laws of any State to the contrary notwithstanding.”  Such preemption means that when there is a conflict between federal and state law, the federal law prevails.  And in the Silkwood case, there was a relevant provision in the Atomic Energy Act, which had been passed by Congress in 1954.  In that act, Congress granted the Nuclear Regulatory Commission (then the Atomic Energy Commission) exclusive authority to set safety standards in the nuclear industry.  Barring some minor lapses, Kerr-McGee had generally complied with those federal regulations. 

When the Court of Appeals for the Tenth Circuit heard the case, it reduced the total award to a mere $5000 (the amount of the property damage when Silkwood’s belongings were destroyed), finding that Oklahoma state tort law had been preempted by the federal Atomic Energy Act. 

The case illustrates the dramatic consequences that can result from the act of federal preemption in specific areas — a difference in result between $5000 and one well in excess of $10 million.  Last week, the Supreme Court of the United States considered another case that turned very much on the issue of federal preemption, but this time in the context of the so-called “Vaccine Court.”  A few months ago, I commented on the case here, just before oral arguments were to be heard. 

Created in 1986 by passage of the National Childhood Vaccine Injury Act, the Vaccine Court has a very limited and specific role:  it applies a form of “no fault” system to the adjudication of injuries that result from the administration of vaccines.  The court was created in response to vaccine-related tort actions, most particularly in response to use of the diphtheria, tetanus, and pertussis (“DTP”) vaccine, which was being blamed for an increased incidence of certain developmental disorders in children.  There was a real concern that the potential tort liability for vaccines was driving vaccine manufacturers from the market, and that this was at odds with the government’s public-health objectives. 

The creation of the Vaccine Court and a no-fault system of adjudication gave vaccine manufacturers a concession by limiting their liability in a highly predictable way.  While the structure of the system was successful in stabilizing the vaccination market, the clear downside is that the no-fault nature of the program may provide insufficient safety incentives to vaccine manufacturers and that some of those who are injured by vaccines are significantly limited in their legal recovery, often limited to recovering awards that are significantly less than their actual damages.  Those who suffer the greatest harms would very much like to have access to state tort laws as an avenue for recovering their full damages and it is difficult not to be sympathetic to those innocent children who had reactions to vaccines that significantly impair the quality of their lives. 

There is an important similarity between regulations of safety practices in the nuclear industry and the administration of the Vaccine Court — in both cases, Congress’s decision to preempt a portion of state tort law potentially has the effect of barring people who suffer from very real injuries from recovering their full damages.  In passing this kind of legislation, Congress presumably believes that a broader purpose is served by excluding legal options that would otherwise exist.  The task of the Supreme Court when confronted with such cases is, though, to decide the precise scope of what Congress has preempted and whether it is constitutionally permitted to do so. 

The Supreme Court did ultimately consider the Silkwood case, and reversed the appellate court to reinstate the large damages award.  A copy of the court’s decision in Silkwood can be read here.  But in the case decided last week, the Court upheld the framework of the Vaccine Court, holding that all design-defect claims against vaccine manufacturers are preempted.  A copy of that decision can be read here.  Why the difference? 

The answer is simple— the decisions hinged critically on the statutory language chosen by Congress, with the Court determining that it was Congress’s intention to exclude all avenues of recovery for faulty designs of vaccines except through the Vaccine Court.  The question remains, though:  Given what we know about the safety issues that exist with vaccines, the effect on public health of having them administered, and the impact on manufacturer behavior resulting from its insulation from state tort liability, is Congress’s choice a wise one or a foolish one?

The Other Side of Despair

Queen Victoria was not at all amused:  “We do not believe that anyone could be insane who wanted to murder a Conservative Prime Minister.”  The target of a number of assassination attempts herself, she was upset enough upon hearing the verdict in the trial of Daniel M’Naghten that she complained to the Prime Minister, prompting the House of Lords to revive an ancient right to intervene by directing the Supreme Court of Judicature to respond to five questions about the case. 

M’Naghten was a Scottish woodcutter who acted on delusions that he was being persecuted by the Conservatives by attempting to murder tory Prime Minister Robert Peel:  “The Tories have … compelled me to do this.  They follow, persecute me wherever I go, and have entirely destroyed my peace of mind ….” 

Setting aside the queen’s implication about the political affiliation of the victim, the facts are well known.  M’Naghten had exhibited symptoms of mental illness for at least a couple of years, his landlady reporting that he believed there were devils in human form seeking to kill him.  Those devils evidently enjoyed the form of conservative politicians because some time later, he applied to the London Police for protection from the Tory persecutors who sought his life.  On January 20, 1843, a few days after being seen loitering suspiciously near Whitehall, he mistook Edward Drummond for Peel, walked up to him in broad daylight, drew his gun, and shot him in the back.  When brought for trial and asked for his plea, he asserted that he was “driven to desperation by persecution” and that he was “guilty of firing.”  After testimony by a number of witnesses, including physicians who testified that M’Naghten’s delusions deprived him of control over his actions, he was found not guilty on the ground of insanity. 

For many years, the M’Naghten test defined the circumstances under which a defendant could succeed in using insanity as a defense to a crime.  The test developed in response to the five questions put to the Court of Judicature by the House of Lords:  a person is not guilty if, at the time of the person’s actions, he either did not know the nature or quality of his actions or did not know what he was doing was wrong.  In modern parlance, there are two tests, either of which may be sufficient for an insanity defense to succeed:  the “cognitive capacity” test and the “moral capacity” test.  But it is the response to the fifth question in particular that interests me today, in which the Court noted the impracticality of deciding the legal question of insanity with science alone; instead, “each of those questions involves the determination of the truth of the facts deposed to, which it is for the jury to decide, and the questions are not mere questions upon a matter of science.”  The record of the case can be found here.  

As Jared Loughner entered a plea of not guilty on Monday after opening fire in Tucson earlier this month, killing a federal judge and seriously injuring a member of Congress, many have questioned to what extent an insanity defense might apply.  The parallels with the M’Naghten case are difficult to avoid:  Loughner appears to have been suffering from a number of delusions, and irrationally focuses responsibility for his ills on the U.S. government.  What is perhaps most surprising, though, is how little the view about the role of psychiatric science in bearing on the question of insanity has changed in 150 years.  After all, the last few years have seen tremendous advances in understanding the structure of the human brain and the physiological relationship with many mental illnesses:  the technique of functional MRI allows specific brain structures to be associated with different activities, different thoughts, different senses, etc. 

Coincidentally, it was the State of Arizona’s insanity test that was the subject of the U.S. Supreme Court’s most recent ruling on the defense, the same state where Loughner went on his rampage.  While Arizona had historically codified the full  M’Naghten test, with both the cognitive-capacity and moral-capacity prongs, it dropped the cognitive-capacity test in 1993 so that a defendant must lack moral capacity for an insanity argument to succeed.  There is a great deal of discussion of different issues related to insanity in the Court’s opinion in Clark v. Arizona (which can be read here), but of particular interest is its general distrust of evidence related to mental disease — and acceptance of Arizona’s restrictions on considering such evidence — even as recently as 2006: 

[T]he diagnosis may mask vigorous debate within the profession….  [T]his professional ferment is a general caution in treating psychological classifications as predicates for excusing otherwise criminal behavior….      [T]here is the potential of mental-disease evidence to mislead jurors … through the power of this kind of evidence to suggest that a defendant suffering from a recognized mental disease lacks cognitive, moral, volitional, or other capacity, when that may not be a sound conclusion at all….  These dangers arise because of the imperfect fit between the questions of ultimate concern to law and the information contained in a clinical diagnosis.

 The Court’s conclusion that a psychiatrist has no more particularly relevant knowledge on the legal and moral issues related to an insanity defense than does a layman has, unsurprisingly, been subject to significant criticism in psychiatric circles.  The American Psychiatric Association and the American Psychological Association have both taken the position, seemingly reasonable, that those who study the human brain are in a position to offer expert evidence about how the functioning of the brain can lead individuals to take actions that we all consider horrific.  A copy of their joint amicus brief in Clark can be read here.  Indeed, it seems more that there is a widespread misunderstanding of mental disease that could be demystified through expert testimony rather than the Court’s view that such testimony may have the effect of clouding the issues and misleading jurors.  Disagreement among experts is nothing new:  juries are called upon all the time to evaluate disagreements, even among experts, so they can reach decisions in cases that are in some sense “fair.” 

Contrary to widespread conception, insanity defenses are only rarely raised and even more rarely succeed.  By all accounts, it appears that Loughner did indeed possess moral capacity when he opened fire, even if he was suffering from certain delusions:  his statements on his Myspace page and his quick statement to “plead the Fifth” when he was arrested demonstrate a knowledge that he well understood what he did was wrong.  I suspect that an insanity defense, if raised, will fail.  But at the same time, as someone who generally believes that we make better decisions when we have more information rather than less, I’m uncomfortable with limiting the evidence of mental illness that a jury can consider when evaluating that defense.