http://www.law.duke.edu/journals/djglp/ ... en10p1.htm Cited: 10 Duke J. of Gender L. & Pol'y 1
[*pg 1]
WHO IS ANDREA YATES? A SHORT STORY ABOUT INSANITY
DEBORAH W. DENNO*
INTRODUCTION
I. THE EARLY LIFE AND TRIAL OF ANDREA YATES
A. Meet the Yates Family
B. The Yates Trial
II. THE INSANITY DEFENSE
A. A Brief Overview of the Insanity Defense
B. The Texas Insanity Standard
III. PARK DIETZ'S EXPERTISE AND PSYCHIATRIC PHILOSOPHY
A. Dietz's Background and Reputation
B. Dietz's Limitations in Expertise and Investigation
IV. ANDREA YATES'S HISTORY OF POSTPARTUM DISORDERS
A. The Early Years of Andrea's Marriage
B. The Start of Andrea's Breakdown
C. Andrea's Plunge into Mental Illness
D. Andrea's Killings and The Aftermath
V. PARK DIETZ'S INTERVIEW AND TESTIMONY IN THE ANDREA YATES CASE
A. Dietz's Interview with Andrea
B. Dietz's Empirically Unsupported Conclusions
C. Dietz's Attempts to Give "Logic" to Andrea's Illogical Delusions
D. Dietz's Criticism of Andrea's Inability to Nurture Her Dead Children
VI. OTHER VIEWPOINTS ON THE ANDREA YATES CASE
A. The Overall Defense and Prosecution Perspective
B. The Jurors' Comments
C. Religion and Culture
D. Andrea Yates's Competency
E. Final Comments
CONCLUSION
FOOTNOTES
INTRODUCTION
We all know by now the story of Andrea Yates. Or, at least we think we do. Andrea Yates, high school valedictorian, swim team champion, college graduate, and registered nurse married Russell ("Rusty") Yates in 1993 after a four-year courtship. Both were twenty-eight.1 Over the next seven years, Andrea2 [*pg 2] gave birth to five children and suffered one miscarriage, all the while plunging deeper into mental illness.3 Then on June 20, 2001, in less than an hour, Andrea drowned all of her children in the bathtub, one by one.4 Months later, she was convicted of capital murder in Harris County, Texas,5 where she now serves a life sentence.6
Some may think that a mentally ill mother who committed such an act should be judged insane. Yet, news accounts and court records suggest that Andrea impaired her attorneys' efforts to plead insanity.7 Such defense plans were already encumbered by the unusually strict Texas insanity standard8 and the state's renowned retributive culture.9 After a jury found Andrea competent to stand trial,10 she resented the efforts that her attorneys mounted on her behalf11 even as she faced possible execution.12 Andrea insisted there was nothing wrong with her mind13 and that she deserved to die.14 She seemed to be awaiting punishment for her sins.
To those closest to Andrea, this self-blaming reaction came as no surprise. They could testify that Andrea had been tormented by bouts of mental illness,15 [*pg 3] and, in fact, both the prosecution and defense agreed that she was mentally ill.16 Andrea's life was also distinguished by religious obsession and a steadfast devotion to tales of sin and Scripture,17 a "repent-or-burn zeal"18 that led her to believe she was a bad mother with ruined offspring.19 According to Andrea, she killed her children to save them from Satan and her own evil maternal influences,20 delusions that did little to help Andrea's defense because they fueled her own desire for punishment.
Public opinion on the Yates killings helps explain some of the more contradictory themes in the case. On the one hand, the public had much sympathy for Andrea and the life that she led.21 Yet, her composed behavior on the day she killed her children stirred a strong retributive response.22 Many were unable to comprehend such violence except by declaring it intentional and evil.23 According to this view, it could be said that Andrea was supremely sane -- her acts rational and premeditated24 -- despite her unquestioned history of postpartum psychosis.25 Andrea propelled this account, spurring the public, her "jury," to see her as the Satanic mother she believed herself to be.26
[*pg 4]
These complex and conflicting aspects of the Yates case fed into the prosecution's depiction of Andrea's mental state on the day she killed her children. But, one psychiatrist's testimony seemed to have a greater impact than the others on the case's outcome.27 The prosecution's star expert, Park Dietz,28 appeared particularly adept at persuading the jury to accept the prosecution's assertion that Andrea was sane and acting intentionally when she killed her children.29 Because the Yates case is on appeal, many of the court records are not available.30 In addition, the defense team still lacks funds to pay for the entire trial transcript31 so it too cannot be examined. Park Dietz's testimony, however, is now accessible32 and it warrants a thorough analysis in its own right.
[*pg 5]
What is most striking about Dietz's testimony is how his opinions about Andrea's mental state could carry so much authority with the jury. Criminal trials commonly involve different sides presenting competing legal "stories" about their version of the facts.33 The law's role is to ensure that just verdicts result from these conflicting representations. Courts must be perceived "as fair and disinterested, capable of rising above the self-serving and adversarial narratives by which cases are presented."34 While the law provides evidentiary standards and procedures to oversee what information is released in court and how,35 an immense amount of discretion exists nonetheless in the ways stories can be told. It remains unclear who is to police these narratives -- beyond the structures already in place -- or whether such oversight is even needed.
In the Yates case, the defense claimed that Andrea's mental illness caused her to believe that killing her children was the right course of action. Although Andrea's attorneys called a number of experts to prove their argument, each expert had a different twist on this central viewpoint.36 Therefore, the defense's story about Andrea, while emphasizing her insanity, was still somewhat muddled. In contrast, the prosecution's story about Andrea's sanity was clearer and also apparently consistent with the cultural norms of Harris County, Texas. The prosecution argued that Andrea may have been gripped by her belief in some demonic command, but she was still fully capable of knowing she was doing something wrong.37 And Andrea seemed to concur, damningly perhaps. Her story was congruent with the prosecution's. She had sinned and deserved punishment for acting out the devil's dictates.38 In all likelihood, however, Andrea's own story was indicative of her mental illness,39 not evidence of the disposition she felt she most deserved. Nonetheless, both her narrative and the prosecution's were accentuated by courtroom storyteller, Park Dietz.
This Article analyzes the problematic aspects of Dietz's testimony in an effort to contribute some balance to the Andrea Yates story. While Dietz's comments may have confirmed the Harris County jury's preconceptions, they were virtually unsubstantiated. Dietz also has no significant expertise in postpartum [*pg 6] depression or psychosis even though both sides agreed that Andrea severely suffered from the disorders and that they significantly affected her conduct.
Of course, expert witnesses are routinely used in litigation.40 Dietz is simply one of the more prominent and prolific examples of what the criminal justice system seeks.41 Despite the long history of expert witnesses in criminal trials, the justice system should question the fairness and efficacy of such an unregulated storytelling process. The potential for inequity is all the more pronounced in a case where the prosecution's story lacks factual justification, both sides agree the defendant is mentally ill, and the death penalty is at stake.
Part I of this Article briefly discusses Andrea's life up to her marriage to Rusty as well as the outcome of her trial. Part II provides an overview of the insanity defense and the strict Texas insanity standard. Part III examines Dietz's background, his reputation, and his psychiatric philosophy, in addition to his proclivity to testify for the prosecution. Part IV describes Andrea's history of mental illness, especially her postpartum psychosis that started with the birth of her first child and ended with a severe psychotic episode. Part V focuses on Dietz's testimony in the Yates trial, beginning with his pre-trial interview with Andrea and ending with an analysis of his conclusions. The discussion emphasizes the speculative nature of many of Dietz's statements and their lack of connection to Andrea's history of mental illness. Part VI presents the other perspectives and experts in the Yates case, and considers how the case might have reached a different result with a more consistent defense strategy or a less rigid insanity standard.
The Andrea Yates case is a vast, book-length, narrative. This commentary covers just a part of the trial. It is beyond this Article's scope, for example, to scrutinize the general role of psychiatric experts in the criminal justice system42 [*pg 7] or to review the research on postpartum depression and postpartum psychosis, which is available elsewhere.43 Nonetheless, examining one piece of the Yates story can be enlightening. "Narrative, we are finally coming to realize, is indeed serious business -- whether in law, in literature, or in life."44
I. THE EARLY LIFE AND TRIAL OF ANDREA YATES
A. Meet the Yates Family
Andrea Yates was raised in the Houston area. Her family background appeared to be middle-American and middle-class.45 Her father was a retired auto shop teacher who died of Alzheimer's disease shortly before the killings.46 Her mother, Jutta Karin, was a homemaker.47 Andrea, the youngest of five, was expected to be a high achiever48 and, in high school, she succeeded: she was captain of the swim team, a National Honor Society member, and valedictorian49 of her 1982 graduating class. Upon completing a two-year pre-nursing program at the University of Houston, she went on to the University of Texas School of Nursing in Houston, graduating in 1986. From 1986 to 1994, she was employed as a registered nurse at the University of Texas M.D. Anderson Cancer Center.50 [*pg 8] Andrea's nursing career ceased entirely, however, soon after her marriage to Rusty.51
Andrea and Rusty first met in 1989 at the Houston apartment complex where they both resided. Both were twenty-five at the time.52 Rusty, "a popular jock" in high school and a summa cum laude graduate of Auburn University, was designing computer systems for NASA.53 Andrea approached him first in conversation -- an uncharacteristically bold move for her, Rusty would later reveal.54 Only after Andrea's arrest would Rusty learn that she had never dated until she had turned twenty-three, that she was recuperating from a romantic break-up at the time they met, and that her directness in initiating contact with him was prompted by intense loneliness and, perhaps, depression.55 Andrea and Rusty spent the next few years becoming acquainted, "living together, reading the Bible, and praying."56
Their April 17, 1993 wedding ceremony was small and simple. Surprisingly, it was also nondenominational,57 perhaps because of the influence of Rusty's spiritual mentor, Michael Woroniecki, from whom "[h]e had learned the faults of organized religion."58 The couple confidently announced to wedding guests that they would not use birth control -- they wanted as many children as nature would provide.59 Their desire for children was immediately fulfilled. Within three months, Andrea was pregnant60 with the first of five children. Eight years later she would kill them all.61
B. The Yates Trial
On July 30, 2001, Andrea was indicted on two counts of capital murder for the deaths of Noah (seven), John (five), and Mary (six months),62 but not for the deaths of her other two children, Luke (three) and Paul (two).63 All of the indictments were for capital murder because they involved more than one person and victims less than six years old.64 On the same day, Andrea's attorneys, George Parnham and Wendell Odom, filed a "notice of intent to offer evidence of the insanity defense," based upon the testimony of two psychiatrists claiming [*pg 9] that Andrea was, at the time of the killings, "mentally insane" as defined by the Texas Penal Code.65
The insanity defense for Andrea would ultimately dissolve.66 Within eight months following her indictment, one jury decided that Andrea was sufficiently competent to stand trial for killing her children67 and another refused her insanity plea.68 Although this second jury declined to impose the death penalty,69 Andrea received a mandatory life sentence for the killings.70 Under the Texas capital felony statute, an inmate must serve forty years in prison before becoming eligible for parole.71 The case is currently on appeal.72
Many theories could explain Andrea's conviction. Of course, the primary theory would speculate that the jury was so horrified by Andrea's acts that any psychiatric evidence offered on her behalf paled in comparison. Yet, the continuing controversy and debate over Andrea's conviction73 suggest that there may be other, more complex, explanations.
Additional rationales primarily point to the retributive aspects of Texas law and culture. As one Harris County resident explained, "There's the rule of law, and there's the rule of law in Texas . . . . The rule of law in Texas is kind of cowboy law."74 For example, Texas consistently executes more individuals than any other state;75 annually it accounts for one-third of all executions in the country,76 a pattern that conflicts with both national and international abolitionist trends. Harris County in particular is responsible for over one-third of the state's death row inmates, making it the harshest death penalty jurisdiction in the country77 [*pg 10] and one of the most punitive in the Western world.78 If Harris County were considered a state, it would follow only two other states (Texas and Virginia) in its number of executions since 1977.79
Because the Yates prosecution sought the death penalty, Andrea's jury was "death qualified." In other words, the prosecution could exclude potential jurors for cause if their negative views toward the death penalty were so strong they "would 'prevent or substantially impair the performance of [their] duties as [jurors]'"80 and therefore render them "unable to faithfully and impartially apply the law."81 Research shows that death qualified juries are more anti-civil libertarian in attitude, particularly with respect to such principles as presumption of innocence and burden of proof, and they are significantly more likely to convict than juries that are not death qualified.82 Presumably, then, Andrea's jury was far less able to "comprehend the inconceivable"83 in evaluating an insanity defense relative to a jury that had not been death qualified.
The Texas insanity standard is a comparably strict rule of law; in the eyes of one legal commentator, it is "one of the most stringent" in the United States.84 [*pg 11] The Yates jury judged psychiatric testimony not only by Texas culture but also by that culture's narrow legal view of what constitutes insanity.
II. THE INSANITY DEFENSE
A. A Brief Overview of the Insanity Defense
Part II explores only the very basics of the insanity defense and how it is applied in the state of Texas.85 The insanity defense is considered one of the most controversial criminal law doctrines, not only because of intense debate over how "insanity" should be defined, but also because of increasing conflict over whether the defense should exist in any form.86 Statistics show that insanity pleas are seldom raised or successful in states throughout the country,87 including Texas.88 Nonetheless, the defense rankles social and community tensions over two conflicting goals: the desire to punish the horrendous, highly publicized crimes that the public typically hears about versus the need to understand that some mentally ill people should not be held responsible for what they do.89
1. The Major Legal Standards for Insanity
The legal standard for insanity varies across the fifty states.90 The first and strictest insanity test of modern usage was introduced in 1843 by the English House of Lords in the M'Naghten case.91 Under M'Naghten, a person is insane if, because of a "disease of the mind" at the time she committed the act, she (1) did [*pg 12] not know the "nature and quality of the act" that she was performing; or (2) if she was aware of the act, she did not know that what she "was doing was wrong," that is, she did not know the difference between right and wrong.92 The M'Naghten rule, which soon became the most widely accepted insanity test in the United States,93 considers only cognitive ability and not volitional conduct.94
Concern over the narrowness of the M'Naghten test prompted attempts over the years to replace it.95 The most successful attempt was the American Law Institute (ALI)'s 1962 insanity test which rapidly gained support from legislatures and courts; by the 1980s, the ALI standard was adopted nearly unanimously by the federal circuit courts and over one-half of the states.96 Under the ALI test, an individual is not responsible for her criminal conduct if, because of mental disease or defect, she either lacked "substantial capacity" to appreciate the "criminality" (or, at the opting of the state legislature, the "wrongfulness") of her conduct, or she failed to "conform" her conduct "to the requirements of law."97
The differences between the ALI and M'Naghten tests are striking. For example, the ALI test accepts both cognitive and volitional impairment as an excuse. In other words, the test considers a defendant's cognitive ability to "appreciate" the criminality or wrongfulness of her conduct as well as her ability to "conform" her conduct to the law.98 This added "conform" requirement is often characterized as a "lack-of-control defense," pertaining to those individuals whose mental disease or defect leads them to lose control over their actions at the time they commit an offense.99
The ALI and M'Naghten standards vary in other important ways. The ALI test requires only that defendants "lack substantial capacity," not total capac-[*pg 13] ity.100 In turn, the ALI applies the broader term "appreciate" rather than "know" when specifying the type of cognitive impairment that leads to insanity; hence, the defendant's lack of emotional understanding can be incorporated into the defense.101 The ALI test also allows the state legislature to consider "wrongfulness" rather than "criminality." This choice enables a finding of insanity if the accused does not know the act was illegal and also if she believes the act was "morally justified" according to community standards.102 At the same time, both the ALI and M'Naghten tests skirt any set definition of the term "mental disease or defect."103 According to the ALI, such an open-ended approach allows the term "to accommodate developing medical understanding"104 and therefore avoid the constraints of old science.
The popularity of the ALI test dwindled in 1981 when a jury found John Hinckley not guilty by reason of insanity, based on an ALI standard, for his attempted assassination of Ronald Reagan.105 The effects of the public furor over Hinckley's acquittal were immediate:106 the federal government and several of the ALI test states abolished the volitional component of the test entirely and imposed other limits, in some cases reverting back to a M'Naghten-type standard.107 According to a 1995 survey of insanity laws, about twenty states still use the ALI test while nearly half of the states apply "ome variation of the M'Naghten/cognitive impairment-only test."108 A handful of states have abolished the insanity defense entirely.109
2. Modern Problems with the M'Naghten Insanity Standard
The return to a M'Naghten-type standard spotlights the problems that the test has always had and why there have been continuing efforts to change it. For [*pg 14] example, the word "know" and the phrase "nature and quality of the act" can be defined either very broadly or narrowly.110 Such vagueness gives legal actors little guidance for interpreting the test and heightens the chance that they will apply it inconsistently across different cases. Likewise, it is not clear whether the "wrong" in the right-and-wrong prong pertains to legal or moral wrongdoing because the language in M'Naghten itself could bolster either approach.111 England has since established that the right-and-wrong element represents the defendant's recognition that an act is legally wrong.112 Yet, American law sides in the opposite direction.113 Most American courts have interpreted the word "wrong" to mean "moral wrong," not "legal wrong."114 This issue was important in the Yates case because Texas law does not specify a particular approach115 and a moral wrong approach would have benefited Andrea. According to some defense experts, Andrea knew that her acts were illegal but she believed they were morally right, given the context of her delusional circumstances.116
In American states that apply the moral right-and-wrong test, questions typically concern whether the defendant knowingly transgressed society's standards of morality, not whether the defendant personally perceived her acts to be morally acceptable. In other words, even if a defendant is mentally ill and, as a result, commits an offense that she believes is morally correct, she is considered sane if she is aware that her conduct is condemned by society.117 As one commentator notes, however, this difference can "be blurred to near extinction" depending on how the particular circumstances in a case are pitched.118 For example, a mentally ill individual "is apt to know that society considers it morally wrong to kill, but if she is acting pursuant to a delusionary belief that God wants her to kill, she might now believe that society would agree with her God-endorsed actions."119
Interpretation of the moral-right-and-wrong standard can vary somewhat in the few M'Naghten jurisdictions that have a "deific decree doctrine," in other words, a rule that allows a mentally disordered defendant to be judged legally insane if she believes that she is acting under the direct command of God (for [*pg 15] example, a belief that God commanded the defendant to kill someone).120 Two primary rationales explain the origins of the deific decree doctrine. First, the doctrine "was merely a logical extension of the Judeo-Christian belief that God would not order a person to kill another" because the Sixth Commandment prohibits murder.121 Therefore, a person thinking that God is commanding her to kill is entertaining a false belief and thus should not be held accountable. Likewise, nineteenth-century courts and juries would not grant the insanity defense to individuals contending that they acted under the command of the Devil or some other religiously corrupt figure because people accepted only "the One True God."122 Second, the doctrine may have been a vehicle for inserting a volitional component exception to the cognitive-only limitations of the M'Naghten rule so that M'Naghten could incorporate at least a narrow category of uncontrolled individuals.123
The exceptions and qualifications for the deific decree doctrine apparently still apply today for defendants experiencing such "command hallucinations."124 The doctrine presumes that the defendant's behavior results from a delusion (a "false belief based on incorrect inference about external reality"),125 and not from a religious conviction,126 although determining the difference between the two can be very difficult.127 While some jurisdictions treat the deific decree rule as an exception to the general insanity standard, other jurisdictions view it as a major factor in assessing an individual's capability to tell right from wrong.128 Irrespective of a jurisdiction's particular approach, these right-wrong issues were key in the Andrea Yates case. Andrea's command hallucinations were a focus of the [*pg 16] expert testimony and what was supposed to be considered "wrong" was neither specified, nor constrained, in the jury charge.129
B. The Texas Insanity Standard
In 1973, Texas joined the ranks of other states and adopted the more lenient ALI definition of insanity.130 A decade later, however, the state returned to a M'Naghten type standard, partly in response to developments surrounding the Hinckley verdict.131 Yet, a critical feature of the Texas test132 is that it is even narrower than M'Naghten, although comparably confusing. The typical M'Naghten standard refers to two parts: the defendant's ability to know (1) the "nature and quality of the act committed" or (2) whether the act was "right or wrong."133 The Texas standard, however, eliminates the first part and refers only to the second, that is, whether the defendant knew the act was right or wrong.134 Texas also limits the defense to cases of severe mental illness and puts the burden of proving insanity on defendants.135 As legal commentators rightly contend, the Texas standard "could hardly be narrower"136 or more "impossible to meet."137
Similar to the M'Naghten standard, defining the terms "right" and "wrong" is a problem.138 For example, the Texas insanity statute does not clarify whether "wrong" should be considered from a legal or a moral standpoint.139 This ambiguity was a key issue in the Yates case, both for the law and the psychiatric pro-[*pg 17] fession. As one psychiatric expert commenting on the case said, there is still no "test" available to determine who is genuinely controlled by command hallucinations; rather, psychiatrists must rely on "a certain degree of approximation[]" in their assessments.140 Likewise, the Yates jury charge did not specify what "wrong" should mean and expert testimony did not seem to restrict the definition of "wrongfulness."141 The Yates jury was free to use the term's "common and ordinary meaning"142 and apply "the statutory language to the facts as it saw fit."143
Such a legally muddled circumstance prompted conflicting approaches to interpreting the Texas insanity standard. As the Yates case evolved, for example, it became clear that both the prosecution and the defense would define the legal-or-moral wrong issue because of the statute's silence. Both sides agreed that Andrea was mentally ill and, in general, that she knew her actions were legally wrong.144 The issue of whether Andrea's mental illness rendered her unable to control her actions, although hotly debated, was moot under the narrow confines of the Texas insanity statute.145 Thus, only one significant question was left for the jury to resolve: Did Andrea know that her actions were morally wrong?
III. PARK DIETZ'S EXPERTISE AND PSYCHIATRIC PHILOSOPHY
There was little legal or psychiatric clarity guiding the determinations to be made in the Yates case. For this reason, the opinions of expert witnesses were especially important. According to a synopsis of the ethical guidelines established by the American Academy of Psychiatry and the Law, "the medical expert is expected to provide a clinical evaluation and a review of the applicable data in light of the legal question posed and in the spirit of honesty and striving for objectivity -- the expert's ethical and professional obligation."146 The Academy specifies that such an obligation "includes a thorough, fair, and impartial review and should not exclude any relevant information in order to create a view favoring either the plaintiff or the defendant."147
According to some legal commentators, Park Dietz's expert testimony was considered "crucial"148 for the conviction of Andrea Yates -- the "defining moment" of the trial.149 Part III examines Dietz's background, experience, and psy-[*pg 18] chiatric philosophy in an effort to explain why Dietz's story about Andrea seemed so much more compelling than the other stories experts had to offer. Notably, much of the information about Dietz derives from interviews with Dietz himself, or from his supporters, in magazines and newspapers. Dietz is commendably forthright about his views in general and was immediately open to commenting on the Yates case as soon as Andrea was sentenced.150 What becomes apparent is how his own self-described, pro-prosecution leanings could mesh so well with a death qualified, Harris County jury.
A. Dietz's Background and Reputation
Park Dietz is considered one of the most "prominent and provocative" psychiatric expert witnesses in the country.151 In one professional capacity or another, he has been involved with a long list of famous homicide defendants: John Hinckley, Jr., Jeffrey Dahmer, Susan Smith, Melissa Drexler, the Menendez brothers, O.J. Simpson (in the civil case), and Ted Kaczynski, to name a few.152 He can now add Andrea Yates to that list. As the prosecution's star witness in the Yates case,153 he both interviewed and videotaped Andrea,154 and he subsequently testified in court about his evaluation.155
Dietz also has extensive professional credentials. He acquired a B.A. from Cornell in biology and psychology, an M.D. from Johns Hopkins School of Medicine, and a Masters in Public Health and Ph.D. in sociology, both from Johns Hopkins. He has held academic posts at Johns Hopkins, the University of Pennsylvania, Harvard, and the University of Virginia.156 His professional experience is substantial, including consulting positions with the Department of Justice and the Federal Bureau of Investigation.157 In addition, Dietz has over one hundred publications, "nearly all" of which concern violent or injurious behavior,158 and he has examined "thousands" of criminal defendants for forensic psychiatric purposes, including sanity determinations.159
Currently (and at the time he testified in the Yates trial), Dietz runs two businesses in Newport Beach, California. He is the president and founder of Park Dietz & Associates, Inc., forensic consultants in medicine and the behav-[*pg 19] ioral sciences, as well as president and founder of Threat Assessment Group, Inc. (TAG), which specializes in the prevention of workplace violence.160 Before arriving in Houston to testify in the Yates case, Dietz mailed his business brochure (describing his companies and the types of cases on which they work) to a wide range of members of Houston's legal community -- prosecutors, defense attorneys, attorneys specializing in premises liability for violent crime, and lawyers representing elder abuse victims.161 Although the Yates defense brought forth evidence of Dietz's brochure distribution during cross-examination in an effort to portray Dietz as a "professional testifier,"162 Dietz did not seem apologetic.163 Nor did such a revelation appear to dent the perceived validity of his testimony.
1. A Desire to Emphasize "Facts"
Media articles about Dietz claim he is known for emphasizing "facts" rather than "theoretical conjecture" when evaluating a case.164 Indeed, both Dr. Jonas Rappeport, a renowned professor of Dietz's at Johns Hopkins Medical School,165 as well as Roger Adelman, one of the prosecutors in the Hinckley case,166 credit Dietz's precision and "focus on the facts" as major contributions Dietz has brought to modernizing the field of forensic psychiatry.
In line with this facts-driven orientation, Dietz seems to be more concerned with the physical evidence linked to a crime than with the defendant's history that can be acquired in an interview.167 According to Dietz, for example, interviews with defendants have typically "been the linchpin of forensic assessments"; yet, there are "serious risks" associated with them because the "[n]atural human techniques for gaining information from an interview unthinkingly cut corners by suggesting answers or guessing at the answer or offering multiple choices."168 Such leading or suggestive procedures are comparable to crime scene evidence that has been contaminated or corrupted.169 Dietz favors instead the second source of mental evidence, which includes examining the crime scene, analyzing autopsies and weapons, and interviewing witnesses to the crime.170 Although "the ideal" would be to have both types of evidence when making an evaluation, Dietz has stated that, "
f I had to choose between [*pg 20] the interview [with the defendant] only or everything except the interview as a means of getting to the truth, I'd prefer everything except the interview because it would get me to the truth more often."171
Dietz's apparent stress on facts,172 combined with what even Rappeport views as a "rigid" approach towards defendants,173 has prompted criticism. According to an article about Dietz in Johns Hopkins Magazine, "ome forensic psychiatrists" have accused him of presenting "mere informed opinion as solid fact, and [complain] that his standard of criminal responsibility is harsh and unforgiving of mentally ill defendants."174 For example, during his testimony in the Yates case, Dietz indicated that because Andrea claimed that Satan, rather than God, told her to kill her children, she knew her actions were wrong.175 Andrea also failed to act in a way a loving mother would if she really thought she was saving her children from hell by killing them. As Dietz stated, "I would expect her to comfort the children, telling them they are going to be with Jesus or be with God, but she does not offer words of comfort to the children."176 However, there appears to be no empirical support for this kind of interpretation of the deific decree doctrine, if in fact that is what Dietz was referencing.177 Rather, if Dietz's explanation has any source at all, it seems to derive from the centuries-old, Judeo-Christian origins of the doctrine itself.178 As one legal critic asked in response to Dietz's comments, "Is one to infer that it is somehow more loving to invoke the name of Jesus while you drown your children than to drown them without any religious commentary?"179 In other words, Dietz appears to be stressing religion, not facts, a focus more aligned with Southern Bible belt culture180 rather than with a medical assessment of Andrea's mental state.
Even Dietz's supporters have admitted that his inflexible approach may prevent him from being able (or willing) to comprehend "some of the psychological nuances of human behavior."181 According to Rappeport, a strong advocate,182 Dietz has the capability to understand and apply knowledge of human behavior, he simply chooses not to.183 As Rappeport explained, "I have a suspicion he may not like to do that. So he may find himself more frequently on the side of the prosecutor, who doesn't like to do those things either."184 Such an omission is a troubling handicap in a field where "[f]ifty percent or more of [*pg 21] medicine is emotional."185 It is particularly problematic given that the cases that typically involve Dietz's testimony often turn on the very "nuances" that Dietz discounts.
Indeed, in media interviews186 and his testimony in the Yates case,187 Dietz has made clear that he does not treat patients in a psychiatry practice. This lack of engagement with patients is "rare" among medical expert witnesses.188 Rather, Dietz opts to concentrate on research and one-time interviews with criminal defendants.189 Yet, such a view of the psychiatric world is distorted. For example, it is difficult to comprehend how Dietz can evaluate an individual's normality or abnormality if he only engages in short-term interviews with highly abnormal people. By encountering briefly only the most extreme criminal cases, all Dietz sees is pathology. He has no "control group" as a comparison, no in-depth evaluations of individuals from whom he can learn nuances. Such an approach may explain additional criticisms concerning where Dietz draws the line for distinguishing sanity from insanity. According to Fred S. Berlin, associate professor of psychiatry at Johns Hopkins and one of the defense's psychiatric experts in the Jeffrey Dahmer case, Dietz's line is too stringent. "He has a high threshold for evidence that tends to suggest impairment. A narrow range for what he defines as psychiatric disorder."190
Consistent with this view, in the Yates case Dietz minimized the defense expert witnesses' testimony that Andrea had suffered years of delusions, auditory hallucinations, and visions of violence.191 Instead, Dietz claimed that Andrea had, at most, experienced "obsessional intrusive thoughts."192 Yet, contrary to other high profile defendants pleading insanity, Andrea had a substantial and documented history of mental illness before she killed her children.193 Not only had she twice attempted suicide, she had also been hospitalized and prescribed anti-psychotic drugs after the birth of her fourth and fifth children.194 The de-[*pg 22] fense could call experts who had actually treated Andrea, some repeatedly,195 in sharp contrast to Dietz's relatively brief interview. As one scholar on expert testimony emphasizes, "[t]he legal system assumes that the treating doctor is more credible than a nontreating doctor"; therefore, the treating physician "is frequently sought to provide expert testimony."196
Nonetheless, Dietz's effectiveness as a witness appears to be due to his alleged emphasis on fact. Because jurors received conflicting expert testimony during the Yates trial, minimal statutory guidance, and unclear stories from both the prosecution and defense, they were left with little to rely on other than the supposed "facts."197 Compounding this dilemma, the multiple defense psychiatrists gave somewhat contradictory analyses of Andrea's mental state,198 presumably in part because she had been treated or assessed by a number of them during different stages of her illness. Such a multiple-theory defense narrative contrasted with the more uniform "factual" narrative presented by Dietz. Given a choice, Dietz's story may have been the preferred alternative; the jury could base a decision on something tangible -- "facts" -- rather than confusion.
2. A Prosecutorial Bent
Almost immediately, Dietz's testimony and post-trial commentary about the Yates case sparked notoriety for the views he expressed both inside and outside the courtroom. In an interview with the New York Times six weeks after his trial testimony, Dietz stressed that his involvement in the Yates case was "troubling," both "professionally and personally."199 As he explained, "t was obvious where public opinion lay, it was obvious she was mentally ill, it was obvious where professional organizations would like the case to go."200 Therefore, while "t would have been the easier course of action to distort the law a little, ignore the evidence a little, and pretend she didn't know what she did was wrong," it also would have been "wrong . . . to stretch the truth and try to engineer the outcome" in that way.201
Dietz also tried to justify his career-long tendency to appear primarily for the prosecution. According to Dietz, prosecutors, like good forensic psychiatrists, strive "to seek truth and justice" and therefore to make available all the information important in a case.202 In contrast, defense attorneys attempt to help their clients -- a goal that conflicts with a thorough search for data. "- ften there are pieces of evidence that are not in their client's interest to have disclosed or [*pg 23] produced."203 Of course, Dietz's statements imply that defense attorneys and their witnesses want to distort information in some way and shield the truth.204
The irony of Dietz's points, however, were spotlighted a week later by Andrea's attorneys. They discovered a factual error that Dietz had made during cross-examination. As the next section discusses, their research showed that Dietz had testified incorrectly about the existence of a television episode about postpartum depression that never aired.205
3. A Mistake in Testimony
Dietz is a technical advisor to two television shows: Law & Order and Law & Order Criminal Intent. In his advisory capacity, he has viewed nearly three hundred episodes of both shows.206 During the Yates trial, Dietz mistakenly testified that, shortly before Andrea killed her children, Law & Order aired an episode involving a postpartum depressed mother who successfully won an insanity appeal after drowning her children in a bathtub.207 The episode never existed.208 When Dietz learned of his error, he wrote prosecutors Joe Owmby and Kaylynn Williford and informed them that he had confused the insanity episode he testified about with other Law & Order episodes and infanticide cases. Dietz's mistake about such a fact, however, may be part of the grounds for Yates's appeal.209 It is not a stretch to think the jury may have been affected by Dietz's implication that Andrea was somehow influenced by the show.
Dietz's statements about the "truth seeking" differences between the prosecution and the defense were also problematic in other ways totally beyond his control and, presumably, his awareness. For example, trial testimony revealed that the defense was not able to acquire copies of particular documents, including Andrea's police offense report. George Parnham, Andrea's attorney, was allowed only to read her police report but not to photocopy it.210 Therefore, Parnham resorted to taking notes on the report, based only on what he could remember of it. As one defense expert later revealed, having only Parnham's notes on Andrea's report put the expert "at a real disadvantage."211
Dietz also claimed that the defense experts asked "shocking examples of leading questions" of Andrea and provided only partial, and biased, videotapes of their interviews with her.212 Predictably, his accusation prompted a response. According to Lucy Puryear, a Houston psychiatrist who testified for Andrea's [*pg 24] defense, Dietz did the same.213 Puryear added that Dietz edited his eight hours of videotaped interviews with Andrea and only "showed the jury portions that supported his testimony."214
Such media debates simply seem to accentuate the general problems associated with incorporating psychiatric testimony in an adversarial process, as well as the weaknesses of the profession itself. Legal commentators emphasized the extent to which both sides in the Yates case differed in their conclusions about Andrea's mental state given that they were purportedly examining the same evidence.215 As the following sections suggest, however, the backgrounds of the experts appeared to have an impact on what kind of evidence they believed was most significant and why.
B. Dietz's Limitations in Expertise and Investigation
This section examines the extent of Park Dietz's background and experience for testifying in a case involving a defendant with an undisputed history of postpartum depression and postpartum psychosis. As one scholar on expert witnesses has emphasized, "[m]edical professionals who undertake the role of expert witnesses are generally expected . . . to be knowledgeable and experienced in the area in which they are functioning as a medical expert."216
1. Postpartum Depression and Postpartum Psychosis
The Yates trial revealed the degree to which Dietz was unfamiliar with patients diagnosed with postpartum depression or postpartum psychosis and his admitted void in treating patients.217 This observation is not meant to elevate the psychiatric classification of postpartum disorders to a level of scientific precision and sophistication that it does not deserve.218 Rather, this section makes clear that there is still much to be learned about postpartum disorders and how much they can justifiably mitigate criminal culpability, if at all. At the same time, what is known medically about the disorders -- especially their neurobiological aspects -- should not be ignored. Two postpartum experts highlighted the problem of such informational inadequacy specifically with respect to the prosecution's approach in the Yates case: "The real challenge for psychiatry is to educate the legal profession and juries about the physiological underpinnings of postpartum disorders and other psychoses . . . and, ultimately, to encourage verdicts based on facts."219
Of course, Park Dietz was not responsible for such a lack of education. It is not the role of the expert witness to provide answers to questions that are never asked or to draw conclusions without a foundation. Andrea's defense attorneys could have more aggressively revealed Dietz's gaps and confronted him with [*pg 25] the history of Andrea's illnesses that Dietz bypassed in his evaluations. Nonetheless, without a fuller expertise on postpartum issues, Dietz's story about Andrea offered a much simpler mental landscape -- and a greater level of speculation -- than may have been warranted given her background.
Direct and cross examinations in the Yates trial made clear that Dietz has been asked to consult on an "unusually high proportion" of cases concerning mothers who kill their children.220 Yet, according to his testimony, the last time he ever treated a female patient with postpartum depression was twenty-five years ago (in 1977).221 Nor was Dietz "sure" that he ever treated a patient for postpartum depression with "psychotic features."222 Dietz conceded that he stopped treating patients totally "many many years ago," in "1981 or 1982"223 and that he has no expertise in women's mental health.224 Dietz's error concerning the showing of a Law & Order episode on postpartum depression225 came about when Parnham was cross-examining him to assess two issues: the sources of Dietz's income, but also whether Dietz had any more expertise in postpartum disorders, even at the level of consulting for television shows, than what he indicated in his testimony on direct examination.226 It appears Dietz did not have more background because he did not offer any information other than his consultancy on a nonexistent show. Such inexperience does not comport with accepted diagnostic principles of psychiatry.227
Dietz's lack of expertise in postpartum depression and postpartum psychosis is striking given the psychiatric community's recognition of postpartum disorders228 and the acceptance by both sides that Andrea was afflicted with one.229 The disorders are included in the Diagnostic and Statistical Manual of Mental Disorders (DSM), published by the American Psychiatric Association, and now in its fourth (text revised) edition (DSM-IV-TR).230 As courts and professionals have noted, "[t]he DSM is often referred to as 'the psychiatric profession's diagnostic Bible.'"231 DSM-IV-TR also clearly recognizes the link between postpartum-related mental disorder and infanticide in the context of delusions.232 Notably, [*pg 26] however, postpartum psychosis is not presently treated as an individual diagnostic classification in the DSM-IV-TR. Rather, the symptoms are categorized according to the established criteria used to diagnose psychosis (for example, major depressive, manic, or mixed episode). The "postpartum onset specifier" applies if symptoms occur within four weeks after childbirth.233
2. Andrea's Postpartum Risk Factors and Life Stressors
It appears that Dietz never really adequately investigated or acknowledged Andrea's postpartum risk factors -- most particularly in the context of the postpartum period's "unique . . . degree of neuroendocrine alterations and psychosocial adjustments," which the DSM emphasizes.234 In other words, the medical literature stresses that the risk factors for postpartum disorders cover a broad scope of biological, psychological, and social influences. These factors include an individual's personal and family history of depression, biochemical imbalances, recent stressful events, marital conflict, and perceived lack of support from the partner, family, or friends.235
[*pg 27]
Andrea experienced all of the postpartum risk factors that the DSM mentions.236 She was also subject to a host of family and environmental life stressors shown to be linked to postpartum depression and postpartum psychosis.237 Dietz only occasionally alluded to these stressors if he mentioned them at all in his testimony. Even if it could be argued that the direct and cross examinations of Dietz did not prompt further references to Andrea's disorders, it would be expected that they would be part of Dietz's evaluation of Andrea independent of his courtroom testimony.
Andrea's stressors were numerous. First, over the course of her marriage to Rusty (during which she was nearly always either pregnant or breastfeeding), Andrea consistently demonstrated DSM-listed criteria for postpartum mood disorder: "fluctuations in mood, mood lability, and preoccupation with infant well-being."238 Like the DSM specification, these feelings "ranged from overconcern to frank delusions"239 and they also took the form of suicide attempts related to the other circumstances in Andrea's life -- uprooted living conditions and transiency, home schooling her five children, her father's death, depressive illnesses throughout her family, Rusty's own bizarre behavior and pressure for more children, as well as Andrea's increasing obsession with religious doctrine, particularly as it was pitched by Michael Woroniecki and his wife, Rachel.240 As the DSM notes, "[t]he presence of severe ruminations or delusional thoughts about the infant is associated with a significantly increased risk of harm to the infant."241 Part IV considers in further detail how Andrea wove such delusional thoughts into a highly stressed life that seemed to spur the thoughts all the more.
IV. ANDREA YATES'S HISTORY OF POSTPARTUM DISORDERS
A. The Early Years of Andrea's Marriage
Andrea's postpartum difficulties appeared with her first pregnancy. Soon after Noah's birth in 1994,242 for example, Andrea experienced hallucinations -- a striking vision of a knife and her stabbing someone. She dismissed the image and never revealed it to anyone243until after her arrest, when she told Rusty.244 As research shows, postpartum depressed or psychotic women often feel ashamed or embarrassed to admit to others their thoughts about harming their infants.245
When Andrea became pregnant a second time in 1995 (with John), she gave up swimming and jogging and also saw less of her friends.246 Her lifestyle [*pg 28] switched yet again in 1996, when Rusty was offered work on a six-month NASA-related project in Florida -- an event that prompted the leasing of their four-bedroom suburban house and a drive to Florida in a thirty-eight foot trailer.247 That trailer would become their "home" in a recreational-vehicle community where Andrea would care for Noah and John while Rusty worked.248 In Florida, Andrea miscarried but then became pregnant a third time just when Rusty had completed his job and was ready to move back to Houston.249
The return to Houston did not mean re-inhabiting their house even though in 1997 Andrea gave birth to a third child, Paul.250 Rusty had other ideas. In an effort to live "light" and "easy," the Yateses rented a lot for their trailer.251 By 1998, after several months of trailer living, Rusty's "easy living" philosophy took a new twist. He learned that a traveling evangelist, Michael Woroniecki, whose advice had inspired Rusty in college, was selling a motor home that Woroniecki had converted from a 1978 Greyhound bus.252 Woroniecki, his wife Rachel, and their children had used the 350-square feet of bus for home and travel for their mobile lifestyle.253 Because Andrea and Noah preferred the bus to the trailer, Rusty bought it. Noah and John slept in the luggage compartment, while Andrea, Rusty, Paul, and now, Luke, who was born in 1999, slept in the cabin.254
While her brood expanded, Andrea also became devoted to helping her father, who now had Alzheimer's disease. This task was overwhelming for Andrea.255 At the same time, Andrea became further isolated from everyone. When she did choose to see people, she always visited them, never reciprocating by inviting them to the trailer.256
Rusty's role in Andrea's increasing aloneness, oddity of lifestyle, religious obsession, and continual state of pregnancy should not be downplayed with respect to any facet of Andrea's behavior.257 And it may never be known to what extent Andrea's pregnancies were based on a mutual decision with Rusty or primarily a product of Rusty's desire for a large family. A number of people, including Andrea's mother and her friend Debbie Holmes, suggested Rusty was a dominating force in the Yates family, including the decision to have babies.258
[*pg 29]
B. The Start of Andrea's Breakdown
On June 16, 1999, Andrea called Rusty at work, sobbing and hysterical. He returned to find her shaking uncontrollably and biting her fingers.259 His efforts to calm her to no avail, Rusty took Andrea to her parents' home that evening.260 The next day, while Andrea's mother was napping and Rusty was out doing errands, the full force of Andrea's troubles became unmistakably clear. She attempted suicide by taking forty pills of her mother's antidepressant medication.261 An unconscious Andrea was rushed by ambulance to Methodist Hospital, with Rusty following behind.262
Andrea told the staff at Methodist Hospital that she had consumed the pills to "sleep forever," but afterwards she felt guilty because she had her "family to live for."263 At the same time, her recovery was slow. According to notes taken by a hospital psychiatrist and a social worker, Andrea was evasive about the reasons for her suicide attempt and deflected questions.264 Although Andrea was still depressed, the hospital discharged her for "insurance reasons," the explanation written on her medical chart. The psychiatrist prescribed Zoloft, an antidepressant, and Rusty took Andrea back to her parents' home to rest.265
Andrea did not like taking the medication, however, and her condition only worsened.266 She would stay in bed all day and self-mutilate. At one point, she scratched four bald patches on her scalp, picked sores in her nose, and obsessively scraped "score marks" on her legs and arms.267 Later, she would tell psychiatrists that during this time, she saw visions and heard voices, telling her to get a knife. She also watched a person being stabbed, although she would not identify the victim.268 At the same time, Andrea refused to feed her children or nurse her baby Luke, claiming that they were "all eating too much."269 Such delusions and thoughts about her children are consistent with the criteria listed for postpartum disorders in the DSM.270
It was only after Andrea's attempted suicide that her relatives discovered the extent of her family history of mental illness: Andrea's brother and sister had ongoing treatment for depression, another brother was bipolar, and in hindsight, her father also suffered from depression.271 According to the DSM, this [*pg 30] family history of mental disorder (particularly bipolar disorder),272 along with Andrea's pre- and post-pregnancy experiences with depression,273 are all factors that would heighten the likelihood of postpartum psychotic features. As the DSM explains, "- nce a woman has had a postpartum episode with psychotic features, the risk of recurrence with each subsequent delivery is between 30% and 50%."274
At different times, Andrea also experienced bizarre delusions and hallucinations. She believed that there were video cameras in the ceilings watching her in various rooms in the house and that television characters were communicating with her. She told Rusty of these hallucinations; however, neither of them informed Andrea's doctors, even though Andrea was continually asked whether she had hallucinations.275
Of all of her family members, Andrea seemed to suffer the most and her condition continued to deteriorate. The day before she had an appointment with one of her psychiatrists, Eileen Starbranch, Rusty found Andrea in the bathroom looking at the mirror with a knife at her throat. Rusty had to grab the knife away.276 When Rusty told Starbranch of the incident, she insisted that Andrea be hospitalized again, this time at Memorial Spring Shadows Glen, a private facility in Houston.277
The initial results of this hospitalization were disastrous. Andrea was virtually catatonic for ten days.278 According to clinicians, catatonia is an objective sign of mental disorder whether or not an individual reveals what he or she is thinking.279 It was also only during Andrea's stay at Memorial Spring Shadows Glen that there would ever be any record suggesting that she experienced hallu-[*pg 31] cinations.280 This record was based on a doctor's report and observations by the doctor's assistant.281
Starbranch gave Andrea a multi-drug injection that immediately improved Andrea's behavior, according to Rusty.282 After a sound sleep, Andrea seemed much more like the person he had first met and they had in the evening what he thought was one of their best conversations.283 Only later did Andrea assert that she considered the injection a "truth serum" that led her to lose self-control in a way she abhorred.284 Andrea's view of the injection as a "truth serum" could be considered yet one more bizarre delusion on her part.285
When Andrea returned to her family after treatment, "home" was neither her parents' house (which was too small) nor the bus, which her parents considered unhealthy for her and the children. With her parents' urging, Rusty, a well-salaried ($80,000 a year) project manager at NASA, bought a three-bedroom, two-bath house in a tree-lined, residential neighborhood.286 The house even had a place to park the bus, which was still very important to Rusty. In the more serene surroundings, Andrea apparently prospered -- swimming laps at dawn, baking and sewing, playing with her children, and fostering an environment for home schooling,287 which Rusty encouraged despite the past stress on Andrea.288 At this point, Andrea admitted to Rusty that she had "failed" at their life in the bus; this new phase in their life was a chance to succeed.289
During this period, the family was engaging in three nights per week of Bible study in the living room because Rusty did not like any of the churches in their area. Again, the views of the bus-selling traveling minister Michael Woroniecki would come to have a profound effect on the lives of Andrea and Rusty.290 Through Woroniecki, Rusty came to doubt organized religion, even though Rusty was not in complete agreement with Woroniecki's views.291 Andrea was another story, however. Woroniecki's "repent-or-burn zeal"292 captivated her and she corresponded with Woroniecki and his wife for years after she and Rusty bought their bus.293 Indeed, at times, the Yates family seemed to imitate the Woronieckis -- a bus-living, home-schooling, Bible-reading brood relishing the isolation of itinerancy.294 According to Woroniecki, "the role of woman is derived . . . from the sin of Eve."295 Likewise, he thought that "bad mothers" create [*pg 32] "bad children."296 There came a time when Woroniecki's "hell burning" influence on Andrea was so great, it distressed both her parents and even Rusty.297
By the spring of 2000, Andrea became pregnant again, a decision seemingly made with Rusty when Andrea started to improve so markedly.298 Yet, the news greatly alarmed Starbranch, who had warned that Andrea's problems could be far more serious if they returned,299 as well as Andrea's mother, who had believed all along that Rusty's demands prompted Andrea's breakdown.300 Debbie Holmes, a former nursing colleague of Andrea's, echoed this view of Rusty, claiming that Andrea continually depicted Rusty as manipulative and controlling and that Rusty pushed her to have the fifth baby.301
C. Andrea's Plunge into Mental Illness
Starbranch's predictions rang true. Andrea's pregnancy was met by another downward dive into mental illness, this time precipitated by the death of Andrea's father.302 Andrea also became more absorbed with the teachings of the Bible.303 The effects of the traumatic circumstances surrounding her father's death were obvious: Andrea stopped talking; she would continually hold Mary but not feed her; she would not drink liquids; she scratched and picked at her scalp until she started to become bald again.304
On March 31, 2001, four months after Mary's birth,305 Rusty sought to rehospitalize Andrea, with Starbranch's urging. This time, Rusty took Andrea to the Devereux Texas Treatment Center Network,306 a trip that Andrea adamantly resisted.307 Only with much prodding from Rusty and her brother did Andrea finally agree to go to the hospital. Once there, she refused to sign forms admitting herself. Because he thought Andrea's condition was dangerous, her attending psychiatrist, Mohammed Saeed,308 initiated the process of requesting that a state judge confine Andrea to Austin State Hospital.309 Only after Rusty's continual pleading did Andrea finally agree to sign the forms admitting herself to Devereux.310
Saeed's account of Andrea's condition appeared to be based entirely on Rusty's description rather than from Andrea's treating psychiatrists or from Andrea herself who, Saeed said, rarely spoke.311 When Rusty insisted that Saeed [*pg 33] put Andrea on Haldol,312 a drug that had been helpful to her in the past, Saeed complied.313 Saeed discontinued the treatment shortly thereafter because, he said, her "flat face" seemed to be a side effect.314 Later, Saeed would testify that, based on the little Andrea said, she did not seem psychotic, never described the torment she was going through, and denied experiencing hallucinations and delusions.315
After ten days at Devereux, Andrea finally started feeding herself again -- a behavioral improvement which, in Saeed's opinion, justified discharging her even though her medication regime was still not stable.316 Also, Andrea wanted to go home and Saeed thought that Rusty could take care of her.317
When Andrea returned home, Rusty's mother, Dora, visited from Tennessee to help out during the day while she stayed at a motel in the evenings.318 Yet, there were clear signals of Andrea