People of Michigan v. Jasmine Gordon

STATE OF MICHIGAN

COURT OF APPEALS

PEOPLE OF THE STATE OF MICHIGAN, UNPUBLISHED April 27, 2017 Plaintiff-Appellee,

v No. 329448 Wayne Circuit Court CLIFFORD DANIEL THOMAS, LC No. 15-001020-01-FC

Defendant-Appellant.

PEOPLE OF THE STATE OF MICHIGAN,

Plaintiff-Appellee,

v No. 329449 Wayne Circuit Court JASMINE GORDON, LC No. 15-001020-02-FC

Defendant-Appellant. Before: O’CONNELL, P.J., and GLEICHER and BOONSTRA, JJ.

PER CURIAM.

Defendants Jasmine Gordon and Clifford Daniel Thomas jointly stood trial for the death of three-year-old Jamila Smith. The prosecution attributed Jamila’s death to child abuse. Separate juries convicted Gordon of first-degree child abuse and involuntary manslaughter, and Thomas of involuntary manslaughter and assaulting, resisting or obstructing a police officer. In these consolidated appeals defendants raise several challenges to their convictions, and Thomas additionally challenges his involuntary manslaughter sentence. Thomas’s claims lack merit, and in Docket No. 329448, we affirm Thomas’s convictions and sentences.

Gordon, on the other hand, raises a valid constitutional challenge to the performance of her trial counsel. Counsel abandoned a viable defense strategy supported by expert testimony. Thomas’s jury heard this evidence and acquitted him of the child abuse charge. Gordon’s attorney fabricated a causation theory and presented no evidence to support his personal hypothesis. As a result, Gordon had no defense at all. In Docket No. 329449, we vacate Gordon’s convictions and sentences and remand for a new trial.

-1-

I. BACKGROUND

A. JAMILA’S MEDICAL HISTORY

Jamila’s cause of death formed the central contested issue in the joint trials. The prosecution asserted that the immediate cause was bilateral bronchial pneumonia, and that the pneumonia resulted from an infection (sepsis) triggered by a laceration of her pancreas. Thomas inflicted the pancreatic laceration, the prosecution argued, and Gordon aided and abetted Thomas by turning a blind eye to indications that he was physically abusing the child. The defense contended that Jamila died a natural death due to pneumonia and that the incidental pancreatic laceration did not contribute to her demise.

Jamila’s unusual medical history supplied the backdrop for these competing claims. Jamila weighed just over eight pounds at birth and appeared to be a normal newborn. But at 22 months of age she could not talk, walk, or stand without support. Her height, weight and head circumference were well below normal limits. Physicians at Children’s Hospital of Michigan (CHM) labeled her as “failure to thrive.” Further evaluation revealed that she had a brain malformation of the frontal lobes resulting in a 30% loss of brain matter. Children with this disorder have delayed motor and language development and are intellectually disabled. Gordon and Smith were provided with nutritional counseling and encouraged to enroll Jamila in the Early On program, which offers special education programming for very young children.

Jamila’s medical history included two additional diagnoses that became significant during the trial. At five months of age, Jamila had been treated for nummular eczema, which manifests with coin-shaped lesions on the skin that can itch and crust over. Two months before her death, Jamila contracted hand, foot and mouth disease, a viral illness featuring a rash. Whether the lesions on Jamila’s body were bruises or a rash was a subject of disagreement after her death.

Gordon and Smith separated in March 2014. Jamila and her older brother remained with Gordon. Several months later, Gordon and Thomas began living together. Gordon was the breadwinner of this new family unit. While Gordon worked, her sister and father sometimes cared for her children. At other times, Thomas was in charge. Over time, Thomas became Jamila’s regular caregiver.

In August 2014, Gordon took Jamila to CHM after Smith expressed concern about a bruise on Jamila’s left inner thigh. A doctor measured the bruise at four by five centimeters, and noted the presence of a rash in the same area. A report of suspected child abuse was made to Child Protective Services (CPS). When CPS came to Gordon’s home to investigate, Gordon neglected to mention that Thomas provided care for her children. CPS did not substantiate the child abuse report. One month later, Jamila was dead. And after calling Gordon about the bruise, Smith never saw his daughter again.

B. THE DAY OF JAMILA’S DEATH

During the afternoon of September 18, 2014, Thomas drove Gordon to work. Gordon testified that she called home several times that afternoon and evening to check on the children, and Thomas reassured her that all was well. At approximately 9:00 p.m., Thomas called Gordon -2-

and told her to “come outside” as he had arrived with her car and the children, and something was wrong with Jamila. Gordon quickly ascertained that Jamila was not breathing. She “hopped” into the back seat and started CPR.

Thomas claimed that Jamila had been well throughout that day. He recounted that after taking Gordon to work, he and the children went to a park. Jamila fell while they were there, but he believed nothing was wrong with her. Later, Thomas fed Jamila and she vomited. Thomas described that when he and the children set out to retrieve Gordon from work, Jamila walked to the car under her own power. As they were driving on the freeway, Jamila’s older brother told Thomas that Jamila had stopped breathing or was “playing sleeping.” Thomas claimed that he began driving with one hand and administering CPR with the other, reaching into the backseat to do so.

When they arrived at Henry Ford Hospital, Thomas pulled into a back parking lot instead of the emergency room entrance. Jessica Vollstaedt, an emergency medical technician, was making her way to the back of the hospital when she heard yelling and screaming that a child was not breathing. She responded, determined that Jamila had no pulse, and started CPR. Neither Vollsteadt nor the code team could revive Jamila. Vollsteadt recalled that Jamila had “an abrasion to her forehead and several bruises throughout her body.” She wore only a diaper.

The police were summoned after hospital personnel observed bruising on Jamila’s body and a wound on her forehead.1 When asked where they lived, Thomas offered an incorrect address on Webb in Detroit. The police found only an abandoned house. Thomas eventually admitted that he lived on Rockdale; when the police went to the home they found a bucket containing water and vomit near the bedrooms.

C. THE WEEKS BEFORE JAMILA’S DEATH

Thomas and Gordon were separately interrogated within hours of Jamila’s death. Gordon submitted to a second interview several months later while in custody. All three lengthy interrogations were played for the jury.

Thomas reported that during the last two weeks of Jamila’s life, he cared for her “every other day.” He claimed that about a week and a half earlier, Jamila had hit her head on a doorknob, door edge, or some other part of a door; the exact mechanism changed several times during the interrogation. He put ice and witch hazel on the head bump and administered Benadryl and Pedialyte for her subsequent vomiting. The head injury swelled to the size of a golf ball. “Ever since she hit her head she’s been going down,” Thomas admitted. Jamila did not want to eat and vomited. Thomas further conceded that he made a “bad judgment call” in not taking Jamila to the hospital. He agreed that it was his “responsibility to see that they were cared for during the times” that he had the children.

1 A large, open head wound is readily visible in the autopsy photographs. Two smaller head wounds are also visible.

-3-

Gordon claimed that that she did not take Jamila to the hospital after first seeing the head wound because she was “impaired” when she came home from work that evening due to her ingestion of wine and marijuana. Thomas told her that Jamila ran into the wall or the door while being chased by her brother. Gordon had no explanation for why she did not take Jamila to a doctor the next day, when she was no longer impaired. Instead, Gordon treated Jamila’s head bump with cocoa butter and witch hazel and believed that the witch hazel caused the bump to scab and bleed.

Two days after hitting her head, Jamila’s stomach bloated. Two or three days before her death, the child started vomiting. Gordon volunteered that she planned to take Jamila to the doctor “on my day off,” as she thought the child had only a stomach virus. She described that Jamila “would lay down. She wouldn’t sleep and she had a fever and her bowel movements was a little irregular as well.” Gordon insisted during her interviews that Jamila had rashes, not bruises. She stressed that Jamila suffered from eczema and had recently been treated for hand, foot and mouth disease. At the trial, however, she expressed that she had changed her mind, and believed the marks to be bruises. She blamed Thomas for what happened, asserting that she went to work every day, regularly called to check on her children, and did not know about the pancreatic injury until after the autopsy.

D. THE MEDICAL TESTIMONY

Dr. Kilak Kesha, a forensic pathologist employed by the Wayne County Medical Examiner’s Office, performed an autopsy on Jamila. He found 18 to 23 abrasions (“scrapes”) on her face, back, left thigh and right arm. Dr. Kesha attributed all of the abrasions to blunt force trauma. He also observed 10 to 15 contusions (bruises) on the child’s chest and abdomen, and 12 contusions on her thighs and forearms. The abrasions, too, were caused by “a blunt object.” The bruises were of “varying ages,” Dr. Kesha explained, “from weeks before death until death.”

Most notably, Dr. Kesha discovered a one-half inch tear or laceration of Jamila’s pancreas and fresh blood in her abdominal cavity. The pancreatic laceration had occurred about one week prior to death, he opined. Dr. Kesha described the mechanics of this injury as follows:

A laceration of the pancreas is extremely rare. It usually occurs when a blunt object or force is applied to the front of the abdomen and the back is supported by either [sic] like a throw to the wall. The reason for this is the pancreas is a long organ that crosses the midline of the back. When there’s pressure from the front the pancreas rubs up against the spine. The spine is a bone, it’s hard and this causes the pancreas to split into two.

A child cannot tear her pancreas by playing or falling; this injury requires “significant” force, Dr. Kesha insisted. “It’s rarely seen,” he indicated, other than in car accidents where “a child is unrestrained and the vehicle is travelling at a high rate of speed.” A child with a pancreatic injury would appear ill, have abnormal bowel movements or urination, pain, drowsiness, lethargy, vomiting or a desire to be held all the time. The child’s illness would have been obvious to a caretaker, he concluded.

-4-

Evidence of bleeding was also discovered in Jamila’s mesentery, which is the “fat in the abdomen.” According to Dr. Kesha, “blunt trauma to the abdomen” caused this bleeding, as the mesentery is “very close” to the pancreas. One blow could have caused both areas to bleed, but there was also a second, fresh area of hemorrhage in the mesentery, which means that there were “[m]ultiple blows.” Dr. Kesha conceded that the newer bleeding alternatively could have occurred during “play,” as the pancreas was already injured; he admitted that he did not know how much pressure would be required to reinjure a lacerated pancreas. The first injury to Jamila’s pancreas initiated the infection that led to her death, in his view.

Dr. Kesha found additional bleeding between Jamila’s scalp and skull bones caused by blunt trauma to the head. The magnitude of the hemorrhage was inconsistent with a fall. A knee, foot, fist, piece of wood or bat could have caused her head injuries, which appeared to be about a week old. He conceded that the head injuries did not cause Jamila’s death, nor did the bruising or the abrasions. The bruising on her abdomen may have been related to the pancreatic laceration, however.

The pancreatic laceration “was [the] initiating factor” that triggered “the cascade that led to . . . death.” Dr. Kesha posited that after Jamila’s pancreas was lacerated, she developed sepsis (“an infection in her blood”), and “[t]here was also necrosis or death of her bowel or large colon and she had pneumonia. So the lacerated pancreas caused the sepsis, which caused the pneumonia, which killed the decedent.” Jamila could have developed pneumonia absent the pancreatic laceration, but Dr. Kesha had no reason to think that she would have. Jamila’s multiple injuries demonstrated that she was the victim of child abuse, in Dr. Kesha’s estimation. He ruled her death a homicide.

Dr. Kesha denied that the abrasions or bruises he delineated were rashes, and he saw no evidence of eczema. He admitted that when he first examined Jamila’s brain, he did not appreciate that it was malformed. Only after reading Jamila’s medical record and examining the autopsy photographs did Dr. Kesha realize that there were “developmental issues.” Those issues did not impact Jamila’s pancreatic function or the cause of her death, he stated.

Dr. Ljuvisa Dragovic, the chief forensic pathologist and chief medical examiner for Oakland County, testified only on Thomas’s behalf. He stated that Gordon’s attorney, John McWilliams, had retained him to review Jamila’s medical records and Dr. Kesha’s autopsy report. Like Dr. Kesha, Dr. Dragovic concluded that bronchial pneumonia caused Jamila’s death. His opinion departed from Dr. Kesha’s in several important ways, however. In Dr. Dragovic’s view, Jamila did not have sepsis. Rather, she “had bronchial pneumonia and that’s it, so there is no sense to conclude that the child died of sepsis.” Dragovic conceded that Jamila “might have sustained blunt force trauma in some point in time to her belly,” but she died of pneumonia, and not from the pancreatic laceration. Dr. Dragovic expressed strong disagreement with Dr. Kesha’s opinion that the pancreatic laceration led to the pneumonia, terming it “a quantum leap.” If Jamila truly had sepsis, Dragovic opined, she would not have been able to walk or run around, and may have been unconscious.

Further, Dr. Dragovic disagreed that the pancreatic injury was inflicted by a blunt object such as a fist. In his view, the source had to have been a smaller, “very focused” object such as a stick, a broomstick, or a tricycle handlebar. Had it been a fist, neighboring anatomical structures,

-5-

such as the liver, also would have sustained injury. Jamila could “have been pushed against something by someone or ran into something . . . that is being held or some exposing type of profile or structure, it’s hard enough to render this type of injury.” While Jamila could not have injured her own pancreas, another child could have done it during play or a confrontation. Alternatively, she may have run into something while being chased by someone. He emphasized that had he performed the autopsy, he would not have called Jamila a victim of child abuse.

Dr. Dragovic also took issue with Dr. Kesha’s determination that all of the marks on Jamila’s body represented bruising or trauma. If a pathologist wants “to be sure about the nature of the lesion you have to take a microscopic section, that’s the main tool in any pathologist’s evaluation, tissue sample.” Dr. Kesha did not create any microscopic slides of the bruises. Although Dr. Dragovic’s testimony on this point is somewhat unclear, he suggested that Jamila’s skin conditions could account for some of the marks found at autopsy:

Q. What is your opinion as to the scars or discolorations throughout her abdomen or chest area?

A. The discolorations that we refer to in medicine is a hyperpigmentation in heavily pigmented areas of the skin are actually healed drier diseases or conditions of the skin or injuries of the skin. So the scar is something that occurs up for healing. . . . So those are the categories that are attributable to the appearance of the body of the deceased daughter.

Dr. Dragovic conceded that one of the “lesions” on Jamila’s abdomen “overlies” the general area where the head of the pancreas meets the body of the pancreas. He explained that he could not call it a “bruise:” “This particular lesion, and I have to call it lesion because that’s what it is, it’s a change in the skin. I cannot know exactly how it was generated because clearly there was some pressure, something that allowed the blood vessels to break down and create this bruise appearance.” Although the lesion was consistent with someone doing CPR compressions in that area, the pancreatic laceration beneath it was not “fresh.” A small amount of bleeding near the pancreas may have been the result of the resuscitative efforts.

The prosecutor cross-examined Dr. Dragovic extensively regarding other cases in which he had testified against the Wayne County medical examiner’s office, and brought out that he had been turned down for the position of Wayne County Medical Examiner many years earlier. Dr. Dragovic was combative, as was the prosecutor.

E. THE REMAINING EVIDENCE, AND THE VERDICTS

Gordon’s sister, Lakisha Thaxton, testified that she and Gordon’s father frequently babysat for Jamila before Gordon became involved with Thomas. Thaxton recalled that Jamila had rashes on her body at various times. She last saw Jamila a week or two before the child’s death, and recounted that the child had “rashes or bumps” on her legs and arms that did not concern her. At that time, Jamila, had no bruises or “open marks on her face.” Although the prosecution contended that Gordon had deliberately isolated Jamila during the last weeks of Jamila’s life, Thaxton testified that Gordon had asked her to babysit on the day Jamila died.

-6-

Gordon and Thomas both testified before their respective juries. Thomas vehemently denied ever harming Jamila, and Gordon insisted that she was unaware of any abuse. The juries were also provided with extensive medical records and records generated by personnel with the Early On program. None of the Early On evaluators testified at the trial. Their records established that Jamila was severely developmentally delayed. When she died at the age of 3 years and 4 months, Jamila was nonverbal, could not feed herself, could not follow directions or make her needs known, was not toilet trained, and functioned at the intellectual level of an 11 month old. Nevertheless, the evidence supported that Jamila was able to walk and run.

Ultimately, the two juries rendered somewhat inconsistent verdicts. Thomas’s jury convicted him of involuntary manslaughter, MCL 750.321 (negligent failure to perform a legal duty), and assaulting, resisting or obstructing a police officer (arising from his attempt to flee from arrest at the hospital), but acquitted him of felony murder, MCL 750.316(1)(b), and first- degree child abuse, MCL 750.136b(2). The trial court sentenced Thomas to 57 months to 15 years’ imprisonment for involuntary manslaughter.

Gordon’s jury convicted her first-degree child abuse as an aider and abettor, MCL 750.136b(2), and involuntary manslaughter, MCL 750.321 (negligent failure to perform a legal duty). She was acquitted of felony murder. The trial court sentenced Gordon to 14 to 25 years’ imprisonment on the first-degree child abuse conviction and seven to 15 years’ imprisonment on the involuntary manslaughter conviction.

Both defendants now appeal.

II. DEFENDANT THOMAS

A. SUFFICIENCY OF THE EVIDENCE

Thomas challenges the sufficiency of the evidence underlying his involuntary manslaughter conviction. We review challenges to the sufficiency of the evidence de novo. People v Meissner, 294 Mich App 438, 452; 812 NW2d 37 (2011). “Even though unpreserved, a question of sufficiency of evidence may be reviewed de novo by this Court in a criminal case.” People v Osby, 291 Mich App 412, 415; 804 NW2d 903 (2011).

The crime of involuntary manslaughter is not defined by statute.2 Under the common law:

[I]nvoluntary manslaughter is the unintentional killing of another, without malice, during the commission of an unlawful act not amounting to a felony and not naturally tending to cause great bodily harm; or during the commission of some

2 MCL 750.321 provides: “Any person who shall commit the crime of manslaughter shall be guilty of a felony punishable by imprisonment in the state prison, not more than 15 years or by fine of not more than 7,500 dollars, or both, at the discretion of the court.”

-7-

lawful act, negligently performed; or in the negligent omission to perform a legal duty. [People v Mendoza, 468 Mich 527, 536; 664 NW2d 685 (2003).]

1. LEGAL DUTY

Thomas first challenges the sufficiency of the evidence supporting that he negligently omitted to perform a legal duty. Thomas claims that he merely served as Jamila’s occasional babysitter, and therefore did not have a duty, or even the right, to seek medical assistance for her.

“The law recognizes that under some circumstances the omission of a duty owed by one individual to another, where such omission results in the death of the one to whom the duty is owing, will make the other chargeable with manslaughter.” People v Beardsley, 150 Mich 206, 209; 113 NW 1128 (1907). This rule “is always based on the proposition that the duty neglected must be a legal duty, and not a mere moral obligation.” Id. The Supreme Court catalogued the legal relationships giving rise to a legal duty as follows:

If a person who sustains to another the legal relation of protector, as husband to wife, parent to child, master to seaman, etc., knowing such person to be in peril, willfully and negligently fails to make such reasonable and proper efforts to rescue him as he might have done, without jeopardizing his own life, or the lives of others, he is guilty of manslaughter at least, if by reason of his omission of duty the dependent person dies.

So one who from domestic relationship, public duty, voluntary choice, or otherwise, has the custody and care of a human being, helpless either from imprisonment, infancy, sickness, age, imbecility, or other incapacity of mind or body is bound to execute the charge with proper diligence, and will be held guilty of manslaughter, if by culpable negligence he lets the helpless creature die. [Id. at 209-210 (quotation marks and citation omitted, emphasis added).]

As examples of omissions that could support liability for involuntary manslaughter, Beardsley invoked cases involving a husband who allowed his intoxicated wife to die in the snow, a husband who neglected to provide clothing and shelter for his “insane wife,” and a mother who exposed “her infant child without protection.” Id. at 210. In contrast, the Court agreed with the holding in a Vermont case that the employers of “a maid servant” who lived in their home had no duty to supply the maid servant’s child with “proper care.” Id. at 210-211.

Applying these principles, the Beardsley Court reversed the defendant’s manslaughter conviction. The victim in that case was an adult woman who spent a night or two drinking with the defendant while the defendant’s wife was “temporarily absent from the city.” Id. at 207. She paid a third person to purchase some morphine tablets for her use. Id. The victim unsuccessfully attempted to conceal her ingestion of the morphine from the defendant, and he unsuccessfully tried to prevent her from taking it. Id. at 208. When she passed into a stupor, the defendant “was too intoxicated to be of any assistance.” A third person took her to someone else’s room, where she died. Id. In setting aside the defendant’s manslaughter conviction the Supreme Court held that the defendant had no duty to prevent the victim’s death. The Court explained:

-8-

[T]he deceased was a woman past 30 years of age. She had been twice married. She was accustomed to visiting saloons and to the use of intoxicants. She previously had made assignations with this man in Detroit at least twice. There is no evidence or claim from this record that any duress, fraud, or deceit had been practiced upon her. On the contrary it appears that she went upon this carouse with respondent voluntarily and so continued to remain with him . . . .

. . . Respondent had assumed either in fact or by implication no care or control over his companion. Had this been a case where two men under like circumstances had voluntarily gone on a debauch together, and one had attempted suicide, no one would claim that this doctrine of legal duty could be invoked to hold the other criminally responsible for omitting to make effort to rescue his companion. How can the fact that in this case one of the parties was a woman change the principle of law applicable to it? Deriving and applying the law in this case from the principle of decided cases, we do not find that such legal duty as is contended for existed in fact or by implication on the part of respondent towards the deceased, the omission of which involved criminal liability. [Id. at 214-215 (emphasis added).]

The defendant may have had a moral duty to intervene, the Court opined, but he did not have a legal one. Id. at 215.

Thomas asserts that as Jamila’s “occasional babysitter” he had no legal duty to obtain medical care for her “absent express consent from her parent.” Beardsley supports that Thomas’s “domestic relationship” with Jamila, coupled with her “helpless[ness],” could give rise to a duty. More specifically, the in loco parentis doctrine imposes the legal duty that applies in this case and is fully consistent with the duty principles set forth in Beardsley. Recall that in Beardsley, the Supreme Court opined that a person who “has the custody and care” of a human being unable to care for herself due to “infancy [or] age . . . is bound to execute the charge with proper diligence[.]” The in loco parentis doctrine stands for the same proposition.

“The intimate relationship between parent and child can surely arise outside genetic bonds.” Hush v Devilbiss Co, 77 Mich App 639, 647; 259 NW2d 170 (1977). This Court explained in Hush that whether an adult stands in loco parentis to a child depends, in part, on the adult’s intent:

Legally, the assumption of in loco parentis status1 is a question of intent. Intent to assume parental status can be inferred from the acts and declarations of the parties. Some factors to consider are the age of the child; the degree to which the child is dependent on the person claiming to be standing in loco parentis; the amount of support, if any, provided; the extent to which duties commonly associated with parenthood are exercised. Providing financial support may be an unimportant consideration in light of the relative situations of the person standing in loco parentis and the real parents. In some cases, while providing financial support may be unnecessary, the benefits flowing to the child by reason of someone’s assumption of the other infinitely various parental duties may be most essential.

-9-

_______________________________________________________________ 1 The various definitions of the term “in loco parentis” were compiled in 67 CJS Parent and Child, § 71, pp 803-804:

The term ‘in loco parentis’ has been defined as in the place of a parent; instead of a parent; charged, factitiously, with a parent’s rights, duties, and responsibilities; more specifically, the relationship which a person assumes toward a child not his own, holding the child out to the world as a member of his family toward whom he owes the discharge of parental duties. It has been said that the accepted definition of a person in loco parentis is one who means to put himself in the situation of a lawful parent to the child with respect to the office and duty of making provision for it; one assuming the parental character and discharging parental duties; a person standing in loco parentis to a child is one who has put himself in the situation of a lawful parent by assuming the obligations incident to the parental relation, without going through the formalities necessary to a legal adoption.

Distinctions. One standing in loco parentis has been distinguished from a brother or sister, cousin, or natural parent. Loco parentis differs from adoption in that the one is temporary and the other permanent in nature. (Footnotes omitted.) [Id. at 649 (citations omitted).]

________________________________________________________________

In People v Thomas, 85 Mich App 618, 624; 272 NW2d 157 (1978), this Court applied the in loco parentis doctrine in a criminal case, holding that a person who has “voluntarily assumed a parental function” bears a legal duty to obtain necessary medical attention for one not otherwise able to obtain it on his or her own. The defendant in Thomas was “the work coordinator” at a “religious practical training school” who had obtained permission from the victim’s parents to discipline their mentally ill son “if necessary.” Id. at 621. After a severe beating administered in the guise of discipline, the victim could not walk and gradually became ill with flu-like symptoms. The defendant failed to arrange for any medical intervention, and the young man died. Id. at 622. Medical evidence linked the beating to the death. Id. at 623. This Court upheld the defendant’s conviction for involuntary manslaughter premised on his failure to perform a legal duty, explaining:

Defendant was a supervisor of Oak Haven, stood in a position of authority over the victim and, by talking with the victim’s parents and obtaining their permission to discipline the decedent, he directly and voluntarily assumed a parental function, and stood in a position of loco parentis to the decedent. Under such circumstances, defendant’s beating of the victim coupled with his failure to provide medical attention, when decedent was unable to obtain same himself, violated defendant’s legal duty to care for the victim. The elements of involuntary manslaughter were adequately established. [Id. at 624.]

-10-

In Beardsley’s parlance, Thomas’s legal duty arose “by implication . . . from the facts and circumstances of the case.” He voluntarily undertook a parental function to care for Jamila in her mother’s absence, acting as her sole caretaker during portions of the day. Despite his protestations that he had not caused Jamila’s head injury, he acknowledged that Jamila’s condition began to deteriorate after the head injury occurred and that her mother had decided against medical consultation. Thomas actively participated in “treating” Jamila with ice, cocoa butter and Pedialyte, acting as a parent toward an ill and injured child.

This evidence substantiated that Thomas intentionally and voluntarily undertook a duty of care, and had more than a moral obligation to seek medical care for the child. That Gordon failed to provide necessary care does not excuse Thomas from his independent obligation, as one standing in loco parentis, to react appropriately to the child’s symptoms of serious illness. Further, Thomas was not a mere “babysitter” for Jamila. He cohabitated with her mother, cared for her on a regular basis, and voluntarily addressed her medical needs. These facts suffice to give rise to a legal duty of care for an obviously ill child.

Nor did the prosecutor neglect to introduce sufficient evidence regarding Thomas’s legal duty. The evidence substantiated that Thomas voluntarily assumed the obligations of a parent; he admitted that when Gordon was at work, it was his “responsibility to see that [the children] were cared for[.]” He “cooked and . . . cleaned” for them, changed Jamila’s clothes when she vomited on them, and treated Jamila’s head wound by icing it when it occurred under his watch. Thomas admitted during his interrogation that he “should have” taken Jamila to the hospital. In the fourth hour of questioning he stated: “I shoulda just took her to the hospital” when she did not get better after she had hit her head. He continued: “I made a bad judgment call . . . on not taking her to the hospital.” He explained that ever since she hit her head, “she’s been going down.” Accordingly, we reject Thomas’s arguments that he had no legal duty, or that the prosecution failed to present evidence substantiating it.

2. CAUSATION

Thomas further contends that insufficient evidence supported that his omission caused Jamila’s death. This claim also lacks merit.

Dr. Kesha testified that Jamila’s pancreatic injury would have caused symptoms of illness obvious to a caretaker. He explained that “depending on the age of the child and the development of the child,” she would exhibit “fever, can have abnormal bowel movement, abnormal urination, pain, be drowsy, lethargic, wants to sleep all the time, wants to be held all the time, can have many different symptoms.” In response to the prosecutor’s questioning, Dr. Kesha stated that a child with a pancreatic injury “might vomit or she might not.” If Jamila had received treatment after the first pancreatic injury, Dr. Kesha opined, she would have survived.

The trial evidence substantiated that Jamila had been vomiting since sustaining her head injury, and did not appear to be as energetic as she normally was. Thomas admitted that the child had been sick “that last couple of weeks.” In combination with Dr. Kesha’s testimony, this evidence sufficed to demonstrate a causal relationship between Thomas’s omission and Jamila’s death.

-11-

B. OV 10

Thomas further contends that the trial court improperly assessed 10 points for offense variable 10.

Under the sentencing guidelines, the circuit court’s factual determinations are reviewed for clear error and must be supported by a preponderance of the evidence. Whether the facts, as found, are adequate to satisfy the scoring conditions prescribed by statute, i.e., the application of the facts to the law, is a question of statutory interpretation, which an appellate court reviews de novo. [People v Hardy, 494 Mich 430, 438; 835 NW2d 340 (2013).]

MCL 777.40 provides for the scoring of OV 10 as follows:

(1) [OV] 10 is exploitation of a vulnerable victim. Score [OV] 10 by determining which of the following apply and by assigning the number of points attributable to the one that has the highest number of points:

(a) Predatory conduct was involved.... 15 points

(b) The offender exploited a victim’s physical disability, mental disability, youth or agedness, or a domestic relationship, or the offender abused his or her authority status........ 10 points

(c) The offender exploited a victim by his or her difference in size or strength, or both, or exploited a victim who was intoxicated, under the influence of drugs, asleep, or unconscious...... 5 points

(d) The offender did not exploit a victim’s vulnerability.......... 0 points

(2) The mere existence of 1 or more factors described in subsection (1) does not automatically equate with victim vulnerability.

(3) As used in this section:

***

(b) “Exploit” means to manipulate a victim for selfish or unethical purposes.

(c) “Vulnerability” means the readily apparent susceptibility of a victim to injury, physical restraint, persuasion, or temptation.

The text of OV 10 permits a score of 10 points when “[t]he offender exploited a victim’s . . . youth or agedness[.]” MCL 777.40(1)(b). MCL 777.40(3)(b) defines “exploit” as “to manipulate a victim for selfish or unethical purposes.” “Accordingly, to merit a score of 10 points for OV 10, a defendant must have manipulated a young victim for a selfish or unethical purpose and the victim’s vulnerability must have been readily apparent.” People v Needham, 299 Mich App 251, 255; 829 NW2d 329 (2013).

-12-

Thomas admitted that Jamila had been vomiting during the last week of her life and “[s]he just wasn’t running around busy like she usually be.” During his interview, Thomas had no excuse for failing to obtain medical care for the child, other than that doing so was her mother’s responsibility. Even if Thomas did not inflict any of Jamila’s injuries, he selfishly neglected to meaningfully attend to Jamila’s obvious medical needs. The trial court did not err in scoring OV 10 at 10 points. Accordingly, we affirm both his convictions and his sentences.

III. DEFENDANT GORDON

Gordon contends that her trial counsel’s performance was constitutionally deficient, leading to her convictions. Gordon preserved this issue by successfully moving in this Court for remand to conduct a hearing on this issue. People v Armisted, 295 Mich App 32, 46; 811 NW2d 47 (2011), citing People v Ginther, 390 Mich 436; 212 NW2d 922 (1973).

“Whether a person has been denied the effective assistance of counsel is a mixed question of fact and constitutional law.” People v LeBlanc, 465 Mich 575, 579; 640 NW2d 246 (2002). Findings of fact are reviewed for clear error, while determinations of constitutional law are reviewed de novo. Id. “ ‘[I]t has long been recognized that the right to counsel is the right to the effective assistance of counsel.’ ” United States v Cronic, 466 US 648, 654; 104 S Ct 2039; 80 L Ed 2d 657 (1984), quoting McMann v Richardson, 397 US 759, 771 n 14; 90 S Ct 1441; 25 L Ed 2d 763 (1970). In Strickland v Washington, 466 US 668, 687; 104 S Ct 2052; 80 L Ed 2d 674 (1984), the United States Supreme Court set forth the now familiar formula for an ineffective assistance claim: “First, the defendant must show that counsel’s performance was deficient. . . . Second, the defendant must show that the deficient performance prejudiced the defense.” To establish the first component, a defendant must show that counsel’s performance fell below “an objective standard of reasonableness” under “prevailing professional norms.” People v Solmonson, 261 Mich App 657, 663; 683 NW2d 761 (2004). With respect to the prejudice aspect, the defendant must demonstrate “a reasonable probability that but for counsel’s . . . errors,” the result of the proceedings would have differed. Id. at 663-664. “A reasonable probability is a probability sufficient to undermine confidence in the outcome.” Strickland, 466 US at 694. The defendant must overcome the strong presumptions that his “counsel’s conduct falls within the wide range of reasonable professional assistance,” and that counsel’s actions represented sound trial strategy. Id. at 689.

A defense counsel possesses “wide discretion in matters of trial strategy.” People v Odom, 276 Mich App 407, 415; 740 NW2d 557 (2007). And decisions regarding what witnesses to present are generally considered strategic, supporting relief only when the defendant is denied a substantial defense based on a witness’s absence. People v Payne, 285 Mich App 181, 190; 774 NW2d 714 (2009). This Court may not “substitute our judgment for that of counsel on matters of trial strategy, nor will we use the benefit of hindsight when assessing counsel’s competence.” Id. (quotation marks and citation omitted). However, “even deliberate trial tactics may constitute ineffective assistance of counsel if they fall outside of the wide range of professionally competent assistance.” Martin v Rose, 744 F2d 1245, 1249 (CA 6, 1984) (quotation marks and citation omitted).

-13-

A. GINTHER HEARING

Pursuant to this Court’s order, the trial court conducted a Ginther hearing. Gordon’s trial attorney, John McWilliams, testified that his theory of the defense was that Jamila died due to “natural causes from a congenital birth defect.” The “congenital birth defect” was the child’s underlying brain malformation. McWilliams constructed his causation hypothesis from two facts: Jamila’s brain abnormality and her “failure to thrive” diagnosis at the time the brain abnormality was diagnosed. He postulated “that with the missing 30 percent of a person’s brain that that could, in fact, cause a downward cascading condition of the various systems of a human body, which would and possibly [did] account for the fact that there was a diagnosis by Children’s Hospital that the child was failing to thrive.”

When Jamila was admitted to CHM and the “failure to thrive” diagnosis rendered, McWilliams continued, the treatment was to “pack[] that child with calories.” Yet she lost weight, he observed, “because [she] was failing to thrive.” He concluded from these facts that Jamila’s failure to thrive “was the cause and effect of the 30 percent [loss of brain matter] and the downward cascade of a child who was born with a congenital birth defect, unfortunately.” He elaborated:

These downward cascading symptoms that eventually led to, in my estimation, the breakdown of various not only body functions, but system functions within the child’s body, which included nutrition, failure to thrive, and I thought better explained why the child, when seen in the emergency [room] on the night in question, the child is there deceased, and that had all of these bruises on the interior and back side of the child, and why the child had pneumonia, that all of that was in conjunction with what had previously happened.

No expert witness testified in support of McWilliams’ theory that Jamila’s 2014 pancreatic laceration was related to the 2013 “failure to thrive” diagnosis or to her brain malformation. More importantly, McWilliams knew his theory was scientifically bankrupt before he decided to employ it, but he rested Gordon’s defense on it nonetheless.

McWilliams conceded at the Ginther hearing that he unsuccessfully consulted several medical experts in his search for validation of his idea that Jamila’s brain disorder and failure to thrive caused her death. In addition to Dr. Dragovic, he contacted Dr. Werner Spitz3 and at least one medical malpractice lawyer at the Charfoos & Christensen law firm, which specializes in plaintiffs’ medical malpractice work:

I also showed the morgue pictures and the Autopsy Report of [to?] Dr. Spitz.

3 McWilliams saw Dr. Spitz and another well-known forensic pathologist, Dr. Michael Baden, at a conference. He showed them the autopsy report and the photographs and they told him: “this child was abused.”

-14-

I also submitted the file to the Charfoos & Christensen law firm, well- known for childbirth defects as litigators in these kinds of cases.

And none of them would agree with me that other than - - and also the fact that there was a lacerated pancreas on this particular child, which everybody, including all of the literature that I researched all over the place, could - - no lacerated pancreas could not happen in any child, except for blunt force trauma, improper blunt force trauma, except what I believed was the proper theory that the pancreas received this laceration because of the medical condition, the breakdown cascading of being one of the functions of nutrition, that that caused the laceration, which I thought Dr. Dragovic was going to back me up on.[4]

McWilliams testified that his own research led him to conclude that the pancreas cannot be injured except through trauma. Nevertheless, he believed that “because of a birth defect” and his personal experience with acid reflux, he could explain Jamila’s injury as nontraumatic:

And [I] did a great deal of research, and especially through Dr. Werner Spitz’ Medicolegal book, his textbook, and also looking through many journals, and journal articles with regard to specifically trying to determine how it was otherwise that a pancreas, which is placed uniquely in a human body between - - I’m not exactly sure between what and what - - but it is so deeply embedded in the human body that unless there is some type of either intentional or major accidental injury, trauma, force, that usually a pancreas is not otherwise damaged, except possibly - - it’s my belief - - because of a birth defect, which has a cascading effect not only upon the lungs and not only upon the whatever, blood flow and things like that, but also upon the nutritional value of the child.

And nutrition has to do with the pancreas. The pancreas has to do with breaking down the intake material from the mouth from eating. And it contains many acid enzymes which break down food, which is very volatile because I, you know, from time to time I get this thing called “acid reflux”.

I think that stuff is battery acid in my throat or stomach. Where the heck did it come from? I find out it comes partially from your pancreas. That stuff is really - - and I thought to be truthful and fully disclosing that that, in fact, could have been part of what caused the lacerated pancreas, as opposed to blunt force trauma.[5]

4 The Charfoos attorneys also concluded that the child had been abused, McWilliams volunteered. 5 McWilliams’ flawed understanding of medical facts permeated both Gordon’s trial and the Ginther hearing. McWilliams auditioned his idea that the pancreas “has to do with breaking down the intake material from the mouth when eating” when he cross-examined Dr. Kesha. Dr. Kesha told him that the pancreas secretes enzymes but the enzymes are not used in digestion:

-15-

On cross-examination, the prosecuting attorney summarized these theories with McWilliams:

Q. Now, let me just recap the theories. Your theory was the child died based on the congenital birth defect, which deteriorated from the point of birth until the child eventually succumbed to it partially due to the breakdown of the pancreas and that this was all based on a congenital birth defect?

A. Correct, sir.

McWilliams expressed confidence that the hundreds of pages of medical records introduced into evidence proved him right.

Despite having told Gordon’s jury in his opening statement that the medical examiner for an “adjoining county” would testify as a defense expert witness, McWilliams elected against calling Dr. Dragovic.6 McWilliams justified this decision in several ways. First, he claimed that his causation theory was more likely to lead to a defense verdict than would Dr. Dragovic’s testimony. Second, McWilliams was aware of Dr. Dragovic’s “personal animosity” toward the Wayne County Medical Examiner’s Office and believed that this would negatively impact his Q. . . . Probably a common thing that happens to people is this thing called acid reflux?

A. That’s correct.

Q. I call it battery acid because that’s how bad it burns my throat when I get it. And I have the feeling that something like that has to do with the emanation of that material that emanates from the pancreas, is that true or not?

A. No, the acid is produced by the stomach.

Q. Okay. But some of the enzymes which produce that has come from the pancreas originally, would you agree with me?

A. No.

In fact, the pancreas secretes insulin into the duodenum, not the mouth. Enzymes produced in the mouth and stomach create the “stuff” McWilliams called “acid reflux.” Basic research would have readily revealed to McWilliams that his pancreas/acid reflux theory had no merit whatsoever. See Acid Reflux, available at <http://patients.gi.org/topics/acid-reflux/> (accessed March 29, 2017). 6 McWilliams’ opening included his explanation of his natural-cause of death theory. He advised the jury that they would learn that Jamila had “congenital” birth defects “which caused a vital internal organ to self-destruct, became infected, injecting and collapsing the health of other organs resulting in the child’s death. In other words, natural cause of death.” Her “pancreas was part and parcel and caused by . . . 36 months of congenital birth condition, not blunt force trauma,” he asserted. None of these contentions were validated during the trial.

-16-

cross-examination. Third, Dr. Dragovic “had no explanation as to why and how all of these horrible bruises were on this child.” And fourth, McWilliams believed that he had scored points with the jury that would lead him to victory: Gordon had provided “very fine testimony” and Dr. Kesha had agreed that the child’s brain anomaly “would, in fact, have some effect upon the various aspects of the human development and the system of a human being, which included the pancreas[.]” McWilliams reasoned:

I gained knowledge and information from [Dr. Dragovic], which was extremely helpful, which was the key to other cases, which was the key in my belief to the defense of Ms. Gordon, okay, which I then went on and developed as a strategy and a theory of the defense of Ms. Gordon and I believe also the vindication of not only Ms. Gordon, but the vindication of Mr. Thomas, her codefendant, her boyfriend.

• • * *

. . . I did not call him because three days before the trial I received a letter from Dr. Dragovic. I had anticipated that he would agree with me and reinforce the theory of the defense which I had had. And when I read his letter, I looked at it and thought to myself, as the trial lawyer working the strategy and in conjunction with many conversations with Ms. Gordon, that I had the better strategy.

McWilliams claimed that he discussed this approach with Gordon, and predicted that Dr. Dragovic would get into a “rout” with the prosecutor on the witness stand. The actual events proved him right, McWilliams asserted.

Appellate counsel also questioned McWilliams regarding his decision not to call Rose Bonaci, who worked in the Early On program, despite that she was present in the courthouse and prepared to testify on Gordon’s behalf. Bonaci’s report, prepared for the Detroit Public Schools 10 days before Jamila’s death, did not reference abuse or any signs of abuse. McWilliams testified that he decided against calling Bonaci because had he done so, he believed that the prosecutor would have introduced photos of Gordon and Thomas’s home that were “very unflattering.” McWilliams understood that Jamila had been fully clothed when Bonaci evaluated her, eliminating any chance that Bonaci would have seen the child’s bruises.

Gordon was the only other witness who testified at the Ginther hearing. She claimed that McWilliams did not tell her that he had decided against calling Bonaci and Dr. Dragovic until after Gordon had testified, when he “whispered” it to her at counsel table. Further, Bonaci had never visited the home in which Gordon lived at the time of Jamila’s death; any home visits occurred at a previous apartment. The last evaluation occurred at the school.

The trial court concluded that given McWilliams’ concerns about Dr. Dragovic’s demeanor, Dr. Dragovic’s failure to explain the bruises, and the positive things about Gordon that appeared in the medical and school records introduced in evidence, “I can’t find that it was not sound trial strategy” to forego calling him as a witness. The trial court continued, “It may be that in hindsight maybe calling them might have possibly made a difference, but it might not

-17-

have either.” Gordon’s own statement that “she didn’t want to take her daughter to the doctor because she didn’t want CPS called in” likely had an impact on the jury, the court observed.

B. COUNSEL’S PERFORMANCE WAS DEFICIENT

People v Ackley, 497 Mich 381, 383; 870 NW2d 858 (2015), also involved a child’s death due to alleged abuse. In that case, as here, “expert testimony was critical to explain whether the cause of death was intentional[.]” Counsel in Ackley consulted an expert witness, Dr. Hunter, who advised that he could not help the defense, but suggested another physician who would “ ‘give you your best shot.’ ” Id. at 386. Counsel never consulted that expert, or any other. Instead, he determined to present an accident defense through cross-examination of the prosecution experts. Id. at 386-387. After the defendant was convicted of felony murder and first-degree child abuse, appellate counsel consulted an expert witness (Dr. Werner Spitz) whose affidavit supported that the child’s injury was accidental. Id. The trial court granted the defendant a new trial, finding counsel’s failure to consult with a different medical expert objectively unreasonable. Id. at 387. This Court reversed, holding that counsel’s “decision not to consult a second expert constituted trial strategy[,]” and that no prejudice had resulted. People v Ackley, unpublished opinion per curiam of the Court of Appeals, issued April 22, 2014 (Docket No 318303), unpub op at 4.

The Supreme Court reversed, explaining that “a court cannot insulate the review of counsel’s performance by calling it trial strategy; counsel’s strategy must be sound, and the decisions as to it objectively reasonable.” Id. at 388-389 (quotation marks and citation omitted). Although the selection of an expert is a “paradigmatic example” of trial strategy, the Court determined that the record in Ackley “betray[ed] no objectively reasonable explanation for counsel’s decision to confine his pursuit of expert assistance to Hunter, a self-proclaimed opponent of the very defense theory counsel was to employ at trial, despite Hunter’s referral to at least one other expert who could provide qualified and suitable assistance to the defendant.” Id. at 390-391 (emphasis in original, citation omitted). The Supreme Court unanimously determined that Ackley’s counsel had performed ineffectively, summarizing:

In this case involving such “substantial contradiction in a given area of expertise,” counsel’s failure to engage “expert testimony rebutting the state’s expert testimony” and to become “versed in [the] technical subject matter” most critical to the case resulted in two things: a defense theory without objective, expert testimonial support, and a defense counsel insufficiently equipped to challenge the prosecution’s experts because he possessed only Dr. Hunter’s reluctant and admittedly ill-suited input as his guide. Knott v Mabry, 671 F2d 1208, 1212-1213 (CA 8, 1982). This “constitute[d] a constitutional flaw in the representation” of the defendant, not reasonable strategy. Id. at 1213. [Ackley, 497 Mich at 392.]

Arguably, McWilliams’ decision not to present expert testimony represents more egregious ineffectiveness than that of Ackley’s counsel, as McWilliams had actually retained expert support for an exculpatory theory and the expert appeared at the trial ready to testify. Against that backdrop, McWilliams’ strategic decision to forego the expert and instead present an unsupportable medical theory through his argument and cross-examination constituted

-18-

ineffective performance. As discussed in greater detail below, McWilliams deprived Gordon of a substantial defense because his “strategy” actually amounted to no defense at all.

In defending his strategy at the Ginther hearing, McWilliams asserted that he capably advanced an exculpatory defense despite having no witness support for it, and the trial court apparently agreed. But our review of McWilliams’ cross-examination of Dr. Kesha reveals that the trial court clearly erred in this regard, as McWilliams made no evidentiary headway whatsoever in advancing his personal hypothesis that Jamila’s death and her pancreatic laceration evolved naturally from her “congenital brain anomaly.”

Dr. Kesha firmly rejected that Jamila had “a congenital birth defect.” Nor did Dr. Kesha concede that the “failure to thrive” diagnosis made at CHM had anything to do with the cause of the child’s death:

Q. Okay. Did you see that the child was suffering from some type of generalized congenital birth defect?

A. No.

Q. No, you didn’t see that?

A. I didn’t see any congenital birth defect, no.

Q. Well, you saw a description of a condition severally [sic] developmentally disabled; didn’t you see that?

A. Yes.

Q. Okay. And would that be any - - within the realm of being a congenital birth defect?

A. It’s not considered a birth defect.

Q. Okay.

A. A cleft palate, a cleft lip is a birth defect.

Q. Okay, I understand that. But you did see that there was some problem with regard to this child’s brain functioning or brain makeup; isn’t that true?

A. That’s correct.

• • * *

Q. . . . [T]he parents of this child came to Children’s Hospital in April 2013, something was not functioning right with regard to this child’s digestive process?

A. I don’t know that.

-19-

Q. Okay. It could have though; isn’t that true?

A. It could have.

Q. Right. And that may be one of the reasons why the child was failing to thrive; isn’t that true?

A. That’s true.

Q. Especially if the child was demonstrating flu-like symptoms, vomiting, diarrhea, temperature, fever, throwing up?

A. Yeah, but it would have to be over a long period of time, not sporadic.

Q. Not sporadic?

A. Not sporadic.

• • * *

Q. . . . [I] just need your frank opinion, should you have gone back and taken a look at the Children’s Hospital records at the time that you did this autopsy?

A. It wouldn’t have changed the cause and manner of death.

Q. Okay. I understand that. Because you at the time of making the opinion as to cause and manner of death are only looking at that child as she presented on the day and at the time you made that autopsy; isn’t that true?

A. No.

Q. It’s not?

A. No.

Q. Why not?

A. We have to look at medical records to see any progression or any reason that can account - - anything that can change or make a cause and manner of death.

These excerpts are representative of McWilliams’ unsuccessful efforts to elicit admissions from Dr. Kesha consistent with this theory that Jamila’s brain abnormality was a “congenital defect,” that somehow damaged her pancreas. Viewed indulgently, his questions may have laid the groundwork for his theory. But the answers offered no support for it. His

-20-

cross-examination yielded no concessions or actual evidence that Jamila’s death was natural rather than the result of abuse.7

Furthermore, the jury was twice instructed that counsel’s statements and questions are not evidence, and that only the testimony of witnesses and exhibits admitted in evidence could be considered as such. See M Crim JI 2.5 (“Evidence includes only the sworn testimony of the witnesses, the exhibits admitted into evidence, and anything else I tell you to consider as evidence.”); M Crim JI 2.7 (“The questions the lawyers ask the witnesses are not evidence. Only the answers are evidence. You should not think that something is true just because one of the lawyers asks questions that assume or suggest that it is.”), and M Crim JI 3.5(5) (“The lawyers’ statements and arguments [and any commentary] are not evidence. They are only meant to help you understand the evidence and each side’s legal theories. You should only accept things the lawyers say that are supported by the evidence or by your own common sense and general knowledge. The lawyers’ questions to the witnesses [your questions to the witnesses,] and my questions to the witnesses are also not evidence. You should consider these questions only as they give meaning to the witnesses’ answers.”). The jurors are presumed to have followed their instructions. People v Graves, 458 Mich 476, 486; 581 NW2d 229 (1998). Lacking a witness to corroborate his theory of the case, McWilliams had no actual evidence to support his selected defense, as Dr. Kesha never testified in conformity with McWilliams’ medical theory. Thus, the uncorroborated “defense” McWilliams offered the jury was no defense at all.

McWilliams accurately predicted Dr. Dragovic and the prosecuting attorney’s quarrel during cross-examination regarding Dragovic’s history with the Wayne County Medical Examiner’s Office. McWilliams claimed he had seen this mutual hostility unfold in other trials. The prosecutor argues that McWilliams’ knowledge that Dragovic would be a “problematic” witness justifies his decision not to call him. But this reasoning begs the questions: why did McWilliams consult Dragovic in the first place if he knew that Dragovic’s cross-examination would be treacherous, and why did he promise the jury that Dragovic would testify?

Despite Dr. Dragovic’s scrappiness, as an experienced pathologist he would have provided Gordon with an actual defense to the charge that she had aided and abetted Thomas’s first-degree child abuse and to the charge of felony murder, with child abuse in the first degree serving as the underlying felony. In his report and his testimony, Dr. Dragovic offered that Jamila had not been the victim of child abuse, and that the pancreatic injury did not cause her

7 In his closing argument, McWilliams admitted that his cross-examination had been less than convincing: I wish I could have done a better job of cross examining the medical physician from the medical examiner’s office to demonstrate to you the link between the child losing 30 percent or missing some 30 percent of its brainwave [sic] and a pancreas that was defective, did not mature, was not creating the necessary things to do the digestive system.

-21-

death. His expert testimony was truly exculpatory, in contrast to McWilliams’ unsupported theory.8

Dr. Dragovic’s report put McWilliams on notice that Dragovic could be helpful to the defense in multiple respects. The prosecution relied on an alleged link—sepsis—between the pancreatic injury and the pneumonia to prove that Thomas had fatally abused the child. Dragovic opined that Jamila had not died of sepsis caused by her pancreatic laceration: “The professed dubious patho-physiology of alleged ‘sepsis’ as a plausible cause of bilateral bronchopneumonia of this severely mentally retarded child is without scientific basis and is devoid of common sense.” In his report (and at the trial) he termed Dr. Kesha’s opinion connecting the fatal pneumonia with the pancreatic injury “a quantum leap.” Although he conceded that the child had sustained trauma to her abdomen resulting in the pancreatic laceration, Dr. Dragovic’s report and his testimony supported that this injury could have occurred in an innocent manner, such as a “toddler running into” or being pushed into an object. Further, at the trial, he forcefully contradicted Dr. Kesha’s testimony that the injury could have been inflicted with a fist, explaining that a punch necessarily would have damaged nearby organs. He raised the possibility that Jamila’s head wounds were self-inflicted, stating in his report: “Children with severe mental retardation are known for their propensity to have bouts of head-banging.” (Emphasis in original.) Dr. Dragovic also opined that had he done the autopsy, he would not have called Jamila a victim of child abuse, and that Dr. Kesha had not properly documented that the lesions Dr. Kesha called bruises were actually bruises.

Perhaps equally helpful to the defense, Dr. Dragovic also called into question the competency of Dr. Kesha’s autopsy report. Dr. Dragovic pointed out that when he performed the autopsy, Dr. Kesha did not realize that Jamila’s brain was malformed; Dr. Kesha admitted this rather striking oversight at the trial.9 Further, Dr. Dragovic is a far more experienced pathologist than is Dr. Kesha. Despite his antagonism toward Wayne County, he has served as the Oakland County Medical Examiner for many years, and has a much broader range of experience than did Dr. Kesha, who finished his residency training in 2012.

By rejecting Dr. Dragovic’s testimony and proceeding to defend Gordon with his made- up, evidence-free theory, McWilliams fought a duel with an unloaded gun, and with the bullets in his pocket. McWilliams’ ineffectiveness is particularly glaring in light of the fact that he had promised the jury in his opening statement that it would hear from an expert witness who disagreed with Dr. Kesha.

8 By casting doubt on whether Jamila had been the victim of child abuse, Dr. Dragovic’s report and testimony rebutted the first element of an aiding and abetting charge: that the crime was committed by another person. 9 Dr. Dragovic’s report states: “Despite the severe mental retardation and the obvious microcephaly with the documented brain weight of 843 grams, (i.e. about 30% of the brain tissue missing because it never developed), this child’s brain is deemed without abnormality.”

-22-

Several courts of appeal have held that a defense attorney’s decision to break a promise made during opening statement may qualify as ineffective assistance of counsel. In United States ex rel Hampton v Leibach, 347 F3d 219, 257 (CA 7, 2003), the Court of Appeals for the Seventh Circuit found counsel ineffective for having “reneged on his promises” to present evidence through certain witnesses “without explaining to the jury why he did so.” “Turnabouts of this sort may be justified when ‘unexpected developments . . . warrant . . . changes in previously announced trial strategies,’ ” the Seventh Circuit explained, but “when the failure to present the promised testimony cannot be chalked up to unforeseeable events, the attorney’s broken promise may be unreasonable, for ‘little is more damaging than to fail to produce important evidence that had been promised in an opening.’ ” Id. (citations omitted, alterations in original). See also Johnson v State, 145 SW3d 97, 119 (Tenn Crim App, 2004) (“In the instant case, trial counsel failed to keep their promises to the jury, without a reasonable basis for such departure. All potential pitfalls regarding Dr. Schacht’s testimony were known and considered by counsel prior to trial.”).

We are unpersuaded by McWilliams’ claim that Dragovic’s report supported his decision against calling Dragovic at trial. McWilliams received Dragovic’s report several days before the trial. He knew what Dr. Dragovic had to say, which was that pneumonia caused Jamila’s death and was unrelated to the pancreatic laceration, and that Dr. Kesha’s autopsy was incorrect in other respects. Nothing changed during the trial, other than McWilliams’ failure to elicit any concessions supporting his theory from Dr. Kesha. Nor did McWilliams ever explain what had happened to the expert he promised to produce. But McWilliams’ broken promise, standing alone, would not require us to reverse Gordon’s conviction.

McWilliams’ strategic decision to rest Gordon’s defense on a scientific theory that lacked any evidentiary support was fundamentally unsound strategy and constitutionally ineffective, given that an expert was prepared to testify that the child’s death was not the result of child abuse. The Supreme Court’s holding in Ackley is instructive:

Given the centrality of expert testimony to the prosecution’s proofs and the highly contested nature of the underlying medical issue, counsel committed exactly that kind of error by failing to consult an expert who could meaningfully assist him in advancing his theory of defense and in countering the prosecution’s theory of guilt. [Ackley, 497 Mich at 393-394.]

Here, the jury heard only the prosecution’s evidence that a violently inflicted kick or punch to Jamila’s abdomen lacerated her pancreas, which led to pneumonia and her death. Dr. Dragovic disagreed that a punch or a kick were ever delivered, given that no other organs were injured, and that the laceration led to pneumonia. We conclude that McWilliams’ decision not to present Dr. Dragovic’s testimony fell below an objective standard of reasonableness. Strickland, 466 US at 688.

C. PREJUDICE

But for McWilliams’ deficient performance there is a reasonable probability that the outcome of Gordon’s trial would have been different.

-23-

Gordon’s conviction for first-degree child abuse turned on whether the jury believed that she had aided and abetted Thomas’s physical abuse of the child. Her jury heard only the testimony of Dr. Kesha, which supported that Thomas had intentionally harmed Jamila by delivering a blow that lacerated her pancreas.10 Dr. Dragovic cast considerable doubt on whether Jamila’s pancreatic injury had been intentionally inflicted. He offered innocent explanations for the pancreatic tear, and testified that as an experienced pathologist he would not have deemed her a victim of child abuse. This evidence “would have provided the jury with another viable and impartial perspective on the facts of the case while contradicting the prosecution’s theory[.]” Ackley, 497 Mich at 394.

No direct evidence supported that Thomas had abused the child, or that Gordon had aided and abetted him. While the circumstantial evidence suggested that Gordon was grossly negligent in failing to obtain medical attention for Jamila, that evidence supported only her conviction for involuntary manslaughter. To prove that Gordon aided and abetted Thomas in abusing the child, the prosecution had to convince the jury that Thomas had, in fact, intentionally or knowingly perpetrated the abuse. As McWilliams recognized, creating a reasonable doubt as to whether Jamila was abused was the key to Gordon’s defense of this charge. The only evidence McWilliams employed was Gordon’s own testimony. And her credibility was seriously damaged by her admission that due to her intoxication, she had neglected to take Jamila to the doctor for her head injury.

“Had an impartial, scientifically trained expert corroborated the defendant’s theory, the defendant’s account of the child’s death would not have existed in a vacuum of [her] own self- interest.” Id. Gordon testified that she did not perceive the lesions on Jamila’s body as bruises. Dragovic supported this perception, suggesting that other conditions could account for the changes in the appearance of Jamila’s skin, and that Dr. Kesha had failed to properly document that the lesions were actually bruises by studying them microscopically. Gordon also professed her disbelief that Thomas had intentionally abused her daughter; again, this testimony would have been corroborated and bolstered by Dr. Dragovic. As in Ackley, “expert testimony was not only integral to the prosecution’s ability to supply a narrative of the defendant’s guilt, it was likewise integral to the defendant’s ability to counter that narrative and supply his own.” Id. In other words, without Dr. Dragovic’s support for her defense that she had not aided or abetted intentional abuse of the child, Gordon had no defense at all. To satisfy Strickland’s second prong she need not show that Dr. Dragovic’s “more likely than not” would have produced an acquittal, 466 US at 693, only that there exists “a probability sufficient to undermine confidence” in the verdict. Id. at 694.

We know that another jury saw the same documentary evidence and heard the same testimony—other than Gordon’s but including Dragovic’s—and acquitted Thomas on the charge of first-degree child abuse. In a sense, that jury serves as a control for Strickland’s second prong. Evaluating whether a reasonable probability of a different outcome exists is always a speculative

10 Under MCL 750.136b(2), “A person is guilty of child abuse in the first degree if the person knowingly or intentionally causes serious physical or serious mental harm to a child.”

-24-

endeavor, but here, irrefutable evidence exists that a jury exposed to Dr. Dragovic’s views rendered a different verdict.

In summary, the crux of the prosecution’s case was Dr. Kesha’s testimony that Jamila’s pancreatic laceration was deliberately inflicted, a necessary element of first-degree child abuse. Dr. Dragovic countered this evidence with his testimony that the laceration could have occurred accidentally, and that Dr. Kesha’s theory of how the injury occurred was not scientifically valid. A jury that heard both Drs. Kesha and Dragovic apparently rejected Dr. Kesha’s view. In light of their verdict, Dragovic’s absence during Gordon’s trial raises a reasonable probability of her acquittal and undermines confidence in the jury’s guilty verdict.11

D. ROSE BONACI

We need not consider whether McWilliams performed ineffectively by failing to call Bonaci as a witness, because her testimony is not reasonably likely to have altered the outcome. Bonaci last saw Jamila before the child had sustained her head wound, and likely before the serious abuse occurred.

We affirm in Docket No. 329448, but vacate defendant Gordon’s convictions and sentences and remand for a new trial in Docket No. 329449. We do not retain jurisdiction.

/s/ Peter D. O'Connell /s/ Elizabeth L. Gleicher /s/ Mark T. Boonstra

11 Although Dr. Dragovic’s testimony most readily applied to the child abuse charge, counsel’s constitutionally deficient performance requires vacation of both convictions. “Unless the accused receives the effective assistance of counsel, ‘a serious risk of injustice infects the trial itself.’ ” Cronic, 466 US at 656, quoting Cuyler v Sullivan, 446 US 335, 343; 100 S Ct 1708; 64 L Ed 2d 333 (1980). See also People v Dixon, 263 Mich App 393; 688 NW2d 308 (2004) (BANDSTRA, J. concurring and SCHUETTE, J. concurring in part and dissenting in part).

-25-