Thanks for visiting! Welcome to a new way to research case law. You are viewing a free summary from Descrybe.ai. For citation and good law / bad law checking, legal issue analysis, and other advanced tools, explore our Legal Research Toolkit — not free, but close.
Glenn Bradford v. Richard Brown
Citations: 831 F.3d 902; 2016 U.S. App. LEXIS 14260; 2016 WL 4136957Docket: 15-3706
Court: Court of Appeals for the Seventh Circuit; August 4, 2016; Federal Appellate Court
Original Court Document: View Document
In 1993, Glenn Bradford was convicted of murder and arson in Indiana, receiving an 80-year prison sentence. He filed a federal habeas corpus suit in 2013, claiming he could prove his innocence and requesting a new trial. The district judge denied his request, leading to the appeal. Bradford, an Evansville police officer, had an extramarital affair with Tamara Lohr, who threatened to expose him after he ended the relationship. On the morning of the fire at Lohr’s house, shortly after Bradford completed a night shift, he reported the fire, claiming he attempted to rescue her but failed. Firefighters arrived shortly after Bradford's report and found Lohr’s body in a bed, with evidence suggesting the fire was intentionally set using gasoline. The investigation revealed Lohr had been stabbed to death before the fire was ignited, and her dog was also found stabbed. Police suspected Bradford, noting his attempt to conceal his whereabouts the night of the murder. He lacked an alibi for the time frame when Lohr was likely killed, as a hit-and-run accident he claimed to have responded to was corroborated by a fellow officer who did not see him there. Additionally, a police car resembling Bradford's was seen near Lohr's house around the time of the murder. Trial evidence included Bradford’s dubious alibi attempts and his actions to delete threatening emails from Lohr shortly after the fire. A bank camera captured him two blocks from Lohr’s house shortly before he reported the fire at 6:35 a.m. Testimony at trial debated whether the fire could have burned long enough for Bradford to set it and then report it, with conflicting expert opinions on the fire's duration. No gasoline was found on Bradford’s uniform, and the police did not investigate a potential suspect—a former jailer fired after Lohr accused him of sexual harassment—despite his suspicious behavior. However, there is no evidence linking this man to the arson or murder. Bradford was convicted of murder and arson, with his conviction upheld in Bradford v. State, 675 N.E.2d 296 (Ind. 1996), and his post-conviction relief request denied in Bradford v. State, 988 N.E.2d 1192 (Ind. App. 2013). He argues that he could not have been the arsonist because expert testimony indicated the fire must have burned for more than eight minutes, exceeding the maximum duration if he had set it. Initially, one expert estimated a burn time of 15 minutes, while a principal expert in post-conviction proceedings, Douglas Carpenter, asserted it was at least 30 minutes, based on an analysis of burned materials and carbon monoxide levels in a poodle's blood. Carpenter's claims relied on three factors: the time for the fire to burn through the bedroom door, the COHb accumulation in the poodle, and the mattress consumption time. However, Carpenter could not conclusively prove that the fire would have taken too long to burn through the door for Bradford to be the arsonist. He conducted simulations at a maximum temperature of 350°C but suggested that since none of the 12 window panes were broken, the ceiling temperature must have been below 280°C. This conclusion was based on a study regarding glass breakage at various temperatures, but he acknowledged the study’s limitations and could not verify that all windowpanes experienced the same temperature. Thus, his assumptions about temperature and burn duration lacked solid evidence. In a typical room fire, the temperature is highest at the ceiling, significantly lower near the floor. Carpenter's testimony that the door's top panel would take approximately 11 minutes to char through at a fire temperature of 600°C misrepresented his own findings, which were based on a heat flux corresponding to a ceiling temperature of 350°C. Carpenter did not test char-through times for temperatures above 350°C, despite the possibility that the fire in Lohr’s house was hotter. His estimate that the fire caused a 30-minute delay in elevating carbon monoxide levels (COHb) in a poodle was flawed, as it relied on human metrics and did not account for the dog’s smaller size or potential movement during the fire. It is established that the poodle was alive when the fire started, as the COHb levels indicated inhalation of CO from the fire. Carpenter's conclusions regarding the poodle’s COHb levels were inflated due to measurements taken outside closed doors, as the door to Lohr’s bedroom was partially open, allowing more CO exposure. He asserted that Lohr’s mattress was completely consumed before the door charred through, reasoning that if the fire had reached flashover conditions, the burning mattress would have interacted with the door, which did not happen. Carpenter used mass loss rates for flexible polyurethane foam to estimate the fire duration, concluding it lasted between 30 minutes and 2 hours, suggesting a start time between 4:30 and 6:00 a.m. However, the reliability of this estimate was weakened by the aforementioned errors. Lastly, two officers arriving three minutes after Bradford reported smoke from the eaves, contradicting Carpenter's assertion that such smoke could not occur until after the door had charred through. The document outlines the key details surrounding the timeline and evidence related to the arson and murder case involving Bradford. It asserts that if Carpenter's analysis is accurate, for Bradford to be the arsonist, the fire must have started, the door must have charred through, and smoke must have emerged from the eaves within three minutes of Bradford's arrival, as fire responders noted smoke upon their arrival three minutes later. The specifics of the char-through time and smoke movement remain uncertain, particularly if the fire exceeded 350°C. Two joggers reported seeing smoke from the house shortly after Bradford called emergency services, which Carpenter argues undermines Bradford's potential guilt since it suggests he could not have ignited the fire in under two minutes. However, this argument was not included in Bradford's briefs, and a state expert testified that smoke would likely be visible within one minute of ignition. Bradford contends that the evidence suggesting he could reach the house and start the fire within 65 seconds did not consider the time needed to carry a gas can, turn off electrical breakers, and stab a dog. It is noted that the location of the gas can is uncertain, and if Bradford had brought it inside previously, he could have used it without delay. Additionally, the time taken to stab a small dog would be minimal. The jury may reasonably conclude that 65 seconds was sufficient for Bradford to commit the acts attributed to the murderer-arsonist. Regarding his habeas corpus request for an evidentiary hearing, the District Court must evaluate the new evidence against the trial's guilt evidence. Bradford failed to present compelling evidence of his innocence. He also claims his trial counsel was ineffective for retaining fire expert Barker Davie, whose qualifications and testimony on fire duration, while potentially favorable, were questioned due to the applicability of his charring rate, which was criticized by the National Fire Protection Association for being based on limited conditions. The Indiana Court of Appeals upheld the conclusion that Bradford's trial counsel did not provide ineffective assistance by choosing Davie as a witness, as Davie was a qualified fire expert who provided a reasonable estimate of the fire's duration. The court noted that the defense counsel conducted a meaningful investigation and presented an expert on a pivotal issue, distinguishing the case from precedents of ineffective assistance. Bradford's claim regarding a juror visit to Lohr's house for timing an experiment was deemed lawful, as it involved common sense evaluation without demonstrable risk of a flawed verdict. The judgment denying habeas corpus is affirmed. In dissent, Judge Hamilton argued that Bradford presented compelling evidence of his innocence and his counsel's ineffective assistance regarding the fire's duration, criticizing the state and courts for failing to provide a plausible explanation for the physical evidence. Hamilton contended that the majority's critiques were untested and insufficient to deny relief, advocating for a new trial to thoroughly examine the evidence related to the timing of Bradford’s actions, the fire's characteristics, and fire behavior. Part I of Hamilton's dissent outlines critical facts about the murder and fire, while Part II addresses legal aspects, ineffective assistance of counsel, and the adequacy of expert testimony presented at trial versus post-conviction. The circumstantial nature of the case against Bradford is highlighted, noting the absence of direct evidence linking him to the crime despite his contentious relationship with the victim. The State's theory regarding the timing and circumstances of the murder of Lohr is undermined by significant contradictions. Bradford, who visited Lohr on the evening of August 1, 1992, left her home around 10:20 p.m., at which time she was alive and engaged in a phone call with her father until approximately 11:00 p.m. The following morning, Lohr was found dead from 21 stab wounds, and her bedroom was on fire. Bradford's documented activities showed he was on patrol from 10:30 p.m. until shortly after 6:30 a.m., with a critical gap in his log from 11:06 p.m. to 12:11 a.m., providing the only potential window for him to commit the murder. The State posited that Bradford set the fire shortly before firefighters arrived, but evidence indicates he left his shift at 6:30 a.m. and was seen passing a nearby ATM at 6:34:04 a.m., arriving at Lohr’s house at 6:35:09 a.m., just minutes before the fire was reported. Observations noted the fire produced thick, black smoke, and an empty gas can was found under Lohr's bed, but there were no traces of gasoline or blood on Bradford’s uniform, and he only smelled of smoke. The State's theory suggests that Bradford, while on a busy night shift, could have left to murder Lohr, stage the scene, and return to patrol without raising suspicion. This would require him to execute the murder and arson in a very short time frame while ensuring his return coincided with the discovery of the fire, raising further doubts about the plausibility of the State's narrative. The excerpt critiques the State's theory regarding Bradford's alleged timing and involvement in a fire that purportedly concealed a murder. It questions the rationale behind Bradford's actions, suggesting that if the fire was intended to hide a crime, it would have been more logical to ignite it at night when there was less chance of detection. The physical evidence from the fire is presented as a strong counterargument to the prosecution's claims. Investigators noted that the fire was confined to Lohr's bedroom, with minimal damage outside it, and the bedroom door remained mostly closed during the incident. This configuration limited smoke escape, contradicting the timeline proposed by the State, which required significant smoke production within 90 seconds of ignition. The analysis indicates that the damage inflicted on the mattress and door could not have occurred in the short time frame suggested by the prosecution. The absence of a plausible explanation for the fire's rapid escalation further undermines the State's case. Additionally, the excerpt highlights the critical role of fire duration in the circumstantial evidence against Bradford. If the fire lasted longer than eight and a half minutes, it would negate his culpability regardless of motive or other circumstantial factors. The failure of Bradford's legal team to engage a fire expert to testify about the duration of the fire is framed as ineffective assistance of counsel, contributing to an unreliable verdict. The section concludes by outlining the framework for further examination of expert evidence presented at trial and during post-conviction proceedings, as well as legal standards surrounding ineffective assistance claims. Jesse Storey, a fire investigator with the Evansville Fire Department, testified that the fire lasted approximately seven to ten minutes, aligning with the State's maximum time of eight-and-a-half minutes. He estimated that the door would char through in two to three minutes but did not provide a basis for this claim. Storey asserted that smoke would have been visible from outside within one minute of ignition, a point he did not substantiate, especially considering the nearly closed bedroom door. The defense's expert, Barker Davie, a forensic chemist, focused on fire behavior rather than the precise timing of events. He described the fire's initial intensity due to gasoline, followed by a decrease as oxygen was consumed, and a subsequent flare-up once the door charred through. Davie explained smoke's escape from the bedroom through the door crack and later through a hole after the door charred, eventually rising into the attic and becoming visible above the roof. However, Davie could not provide reliable quantitative estimates for the char-through time or when smoke would be visible. He referenced a general guideline indicating that wood chars at a rate of about one inch per forty-five minutes, applying this to estimate a char-through time of seven to ten minutes for the 5/16” thick door. Davie stated that seeing smoke within the first minute or two of the fire would be "highly improbable." His timing estimates relied on Gerald Johnson's testimony, who smelled smoke approximately seven and a half to ten minutes before Bradford's call at 6:35:09 a.m., suggesting the fire may have started between 6:26 and 6:29 a.m. Davie's testimony indicated that smoke would not have been noticeable for two to five minutes after the fire began, implying an earlier ignition time. During cross-examination, he estimated that the door frame could burn at a rate of one inch every forty-five minutes under constant direct flame, suggesting it would take approximately eleven minutes for complete charring. However, this opinion was based solely on the door frame, which was the only piece of physical evidence considered. His credibility was challenged when the State called Special Agent Donald Johnson, who criticized Davie's 'rule of thumb' as outdated, referencing NFPA Publication 921, which disavowed this standard. Johnson explained that char depth measurements should not be used to estimate fire duration, noting that charring rates vary significantly based on numerous factors, including fire intensity and wood characteristics. Johnson countered Davie's estimate by stating that the door would char through in three to four minutes, though he did not provide empirical support for this claim. Consequently, Davie's reliance on the discredited standard left Bradford without credible expert testimony regarding the fire's timing. The State emphasized this weakness in closing arguments, suggesting that Davie’s extensive testimony boiled down to an inconclusive assertion regarding whether the smoke observed by a witness preceded Bradford's arrival. Ultimately, Davie's lack of reliable expertise on fire duration failed to assist the defense effectively. Bradford was convicted of murder and arson, and his convictions were upheld on appeal. He later sought post-conviction relief, focusing on scientific evidence to argue that the fire could not have developed as the State claimed to support his guilt. Johnson testified that the smoke smelled like burning wood, which contradicted the expected odor from the foam mattress, the primary fuel source. He noted he was convinced the smoke originated from behind the house, not from Lohr's bedroom, which would have prompted him to alert authorities. Douglas Carpenter, a fire protection engineer, provided critical testimony that was missing in the original trial. He conducted a quantitative analysis of the fire's duration, explaining that flashover—where all combustible surfaces ignite—did not occur, as evidenced by the physical conditions observed. Carpenter detailed how fire conditions were "ventilation limited," indicating that the fire was constrained by a lack of oxygen rather than fuel. He highlighted that the nearly closed bedroom door restricted oxygen replenishment and minimized smoke escape before char-through occurred. This meant there was no substantial smoke buildup in the house or attic prior to the door's charring, preventing a visible smoke column. After the door charred through, however, smoke could escape more freely, leading to accumulation in the attic and the heavy plume observed by witnesses at 6:36 a.m. Carpenter also challenged the State's experts' estimate of two to four minutes for the door to char through, instead conducting experiments that showed it took an average of 11 minutes and 33 seconds at a heat flux of 20 kW/m², which exceeded the maximum duration of the fire according to the State's theory. Carpenter needed to establish the actual heat flux from the fire in Lohr's bedroom to substantiate his findings. Carpenter's analysis of the fire's dynamics involved evaluating the bedroom window and door. He referenced the NFPA 921 standard, indicating a char-through rate of 0.4 inches per hour at 750°F (400°C). The bedroom door, at 0.3 inches thick, could not have reached 750°F, as evidenced by the absence of glass breakage in the twelve-pane window, which only showed possible cracking in one pane. Based on studies of glass breakage probabilities, he estimated the temperature to be around 280°C (540°F) with a heat flux of 16 kW/m², implying a char-through time of approximately 30 minutes. Carpenter examined the fully consumed mattress, a key fuel source, estimating it contained about 20 kg of polyurethane foam. He noted that the fire was ventilation-limited due to the nearly closed bedroom door, which delayed the introduction of more oxygen. Utilizing the CFAST model, he predicted a two-hour burn duration for the mattress with a slightly ajar door, allowing him to infer the fire was likely set between 4:30 a.m. and 6:00 a.m., well before Bradford could have been involved. Independent testimony corroborated Carpenter's findings, noting a visible smoke column at 6:36 a.m. Additionally, the analysis included the criteria for ineffective assistance of counsel, establishing that Bradford demonstrated both deficient performance by his counsel and resulting prejudice, as defined in Strickland v. Washington. Deficient performance is characterized by representation that falls below an objective standard of reasonableness. Debate exists regarding the reliability of Carpenter’s analysis, which claimed that the concentration of carbon monoxide in Lohr’s dog’s blood indicated the fire burned longer than eight minutes. The court refrains from relying on this analysis due to its controversial nature. When evaluating counsel’s performance, courts must adopt a deferential approach, recognizing that strategic decisions may influence actions taken. Establishing a claim for ineffective assistance of counsel is challenging unless an isolated error is egregious and prejudicial. In this case, the selection of expert witness Davie was deemed a reasonable strategic choice by Indiana courts, given his good reputation within the fire investigation community. However, Davie failed to provide a credible opinion on a crucial issue: whether it was feasible for Bradford to have set the fire. Davie’s limited analysis overlooked significant physical evidence, such as intact windows and a fully consumed mattress, and relied on an outdated method discredited prior to the trial. The prosecutor effectively demonstrated Davie’s inadequacy during closing arguments. Other circuits have recognized deficient performance when counsel neglects to present expert evidence on critical issues, which undermines the defense's theory. The failure to challenge the prosecution's expert finding allowed the prosecution to dominate the narrative, thereby jeopardizing the defense's position. This case underscores the essential need for expert consultation in challenging the State’s claims regarding arson. In Rogers v. Israel, the court emphasized that effective representation in criminal cases may necessitate counsel to present expert testimony that counters the state's expert evidence. For an error to be classified as ineffective assistance, it must be egregious and prejudicial, meaning that the omitted evidence should have been obviously better than the defense pursued at trial. Scientific testimony that could exonerate a defendant qualifies under this standard, provided it was known at the time of the trial. The state trial court concluded that the expert opinion from Carpenter was not "new" evidence since it merely offered a different perspective on previously available evidence. The court determined that the theories and calculations Carpenter used were discoverable with due diligence during the trial. The Indiana Court of Appeals supported this conclusion, reinforcing that the defense’s failure to utilize Carpenter's expertise was a significant oversight against the weak testimony presented by the state's witnesses. To succeed in a habeas corpus claim, Bradford must demonstrate that it is reasonably likely that counsel’s failure to present Carpenter’s analysis would have led to a different trial outcome. The standard requires a substantial likelihood of a different result, not merely a conceivable one, and does not necessitate that the chances of acquittal exceed 50 percent. Here, the lack of reliable scientific evidence on fire duration presented at the trial, coupled with the critical timing established by a bank camera, suggests that if Carpenter’s analysis had been introduced, it could have significantly impacted the jury's decision. Bradford notes the challenges in obtaining federal habeas relief based on ineffective counsel claims. Despite scrutiny, Carpenter's scientific analysis remained intact during cross-examination in state courts, supporting Bradford's assertion that his conviction is undermined by this evidence. The prosecutor aimed to demonstrate that an alternative expert could have conducted a comparable analysis during the trial, which supports the argument against the reliability of Davie's findings. The prosecution raised issues regarding the thickness of window glass, the similarity of door samples in char-through experiments, and the dimensions of Lohr's mattress; however, Carpenter effectively countered these challenges, asserting the reliability of his estimates and indicating that minor variable changes would not affect the overall conclusion. The state trial and appellate courts avoided substantial engagement with Carpenter's analysis. In this federal appeal, a court makes an initial attempt to evaluate Carpenter's work on its merits, but the majority's new criticisms do not significantly undermine his primary conclusion. Additionally, there is a discussion about whether the federal constitutional guarantee of due process could support a claim for relief based solely on actual innocence, as referenced in House v. Bell and Herrera v. Collins, although the threshold for such a claim is considered extremely high. The petitioner must demonstrate a strong likelihood of innocence, which the author believes Bradford has achieved, albeit with some reservations due to the absence of expert rebuttal to Carpenter's analysis. The majority critiques Carpenter’s probability assessment regarding glass breakage, which the author acknowledges is flawed. Specifically, a one in twelve chance of a glass pane breaking does not guarantee that one will break in twelve trials, similar to coin toss probabilities. Carpenter's failure to explain the uniform exposure of all panes to heat, uncertainties about the glass thickness, and the lack of challenges from the prosecutor regarding the probability of window panes remaining intact are noted. Carpenter’s report suggests that the probability of glass breakage increases significantly at temperatures around 275°C, and by 350°C, the likelihood that none of the twelve panes would break becomes exceedingly low. Additionally, established literature indicates that windowpanes typically crack at temperatures between 150-200°C, yet Carpenter’s findings show no signs of cracking in the front bedroom window. Even with corrections to Carpenter's probability analysis regarding glass, the State fails to clarify why all front windowpanes remained intact if the fire was intense enough to char through the door rapidly. Many critiques of Carpenter's work were not raised during his post-conviction cross-examination, limiting his ability to defend his conclusions. Although the majority criticizes him for confusing heat flux levels in his testimony, this mistake was not present in his report, which clearly states that his char-through time estimate relied on intact glass. The lack of questioning from the prosecutor on this point prevented a thorough discussion of Carpenter's assumptions. While one critique regarding the type of glass was addressed, Carpenter's explanations suggest he could have effectively rebutted other critiques if given the opportunity. Although Carpenter's evidence is not flawless and relies on certain assumptions due to the nature of fire behavior, it remains the only empirical analysis of the fire's duration, based on actual tests rather than conjecture. His findings are quantitative and testable, indicating that the fire would have had to be significantly more intense than the State's theory allows, as it could not have caused the observed damage within the alleged burn time. Furthermore, the rapid formation of a heavy smoke plume visible from a distance contradicts the State's timeline, implying that char-through times must be measured in seconds rather than minutes. The analysis indicates that the fire's ventilation contradicts expert opinions, suggesting it could have reached flashover, which it did not. Dr. John D. DeHaan's report highlights significant fire damage to the victim, Lohr, inconsistent with a short fire duration of five to seven minutes; instead, it aligns more with exposure to flames for approximately 20 minutes or longer, albeit potentially reduced in intensity due to ventilation constraints. Despite minor errors in Carpenter’s analysis, it remains the only scientifically and empirically grounded assessment of the fire's duration, supporting the conclusion that Bradford may be exonerated. The State has the opportunity to challenge Carpenter's findings in a retrial, potentially presenting new expert testimony on the fire's characteristics. The court's view that Carpenter's analysis was accessible during the original trial implies ineffective assistance of counsel for not pursuing critical expert evidence. Alternatively, this case is suitable for an actual innocence claim for habeas corpus. A writ should be issued for Bradford’s release or retrial.