Full opinion text
MEMORANDUM AND ORDER RE: MOTION TO VACATE CONVICTION GERTNER, District Judge. TABLE OF CONTENTS I.INTRODUCTION..........................................................91 II. FINDINGS OF FACT ......................................................95 A. Background............................................................95 B. Pre-Trial..............................................................97 C. Trial..................................................................99 1. Domingos’ Testimony: Building Construction and Burn Patterns..........99 2. Canine Evidence...................................................101 3. Dragan: Accelerant Sample.........................................103 4. Myers’ Testimony: Thermal Imaging.................................103 5. Other Evidence....................................................104 6. Defense Case......................................................104 a. Burn Patterns.................................................104 b. Thermal Imaging..............................................105 c. Laboratory Analysis............................................105 7. Government’s Rebuttal.............................................105 D. Closing...............................................................105 E. Verdict...............................................................106 F. Habeas Evidentiary Hearing............................................106 1. Testimony Re: Counsel’s Performance................................106 2.Testimony Re: Prejudice ...........................................107 a. Cause-and-origin Testimony.....................................107 (1) Basement Theory ..........................................107 (2) Building Construction and Burn Patterns......................108 (3) Thermal Imaging...........................................109 b. Arson Evidence................................................109 (1) Canine Evidence...........................................109 (2) Laboratory Analysis........................................Ill III. ANALYSIS...............................................................Ill A. Defense Counsel’s Performance Was Constitutionally Deficient..............Ill 1. Counsel Erred in Failing to Move for a Daubert Hearing on any Expert Testimony................................................112 2. Failure to Challenge Canine Evidence Was Error......................115 a. Failure to Move for Daubert Hearing.............................115 b. Failure to Object to Lynch’s Testimony...........................119 3. Failure to Request a Daubert Hearing on the Laboratory Test and Drugan’s Testimony Was Error....................................120 4. Failure to Request Daubert Hearing on Cause-and-Origin Evidence Was Error......................................................121 B. Counsel’s Deficient Performance Prejudiced Hebshie.......................122 1. Is There a Reasonable Probability the Court Would Have Granted a Daubert Hearing?................................................122 2. Is There a Reasonable Probability That as a Result of the Daubert Hearing, the Court Would Have Excluded or Limited Evidence?.....123 a. Laboratory Accelerant Sample Test..............................123 b. Canine Evidence...............................................124 c. Cause-and-Origin Testimony (Domingos and Miller)................124 3. Is There a Reasonable Probability That Exclusion or Limitation of the Expert Evidence Would Have Undermined Confidence in the Verdict? ........................................................127 I. INTRODUCTION Petitioner James Hebshie (“Hebshie”) was convicted of arson and mail fraud in June 2006 for an April 2001 fire in a commercial building in Taunton, Massachusetts. At the time of the fire, Hebshie was leasing space in the building for his convenience store, Main Street Lottery & News Store. He was sentenced to a mandatory fifteen years in prison, all the while proclaiming his innocence. After exhausting his appeals, Hebshie filed a petition for a writ of habeas corpus pursuant to 28 U.S.C. § 2255 based on the ineffective assistance of his trial counsel. His petition raises a number of grounds but its principal focus is the way counsel dealt with scientific testimony on arson. The government had to prove that a crime had been committed at all — that the fire was intentionally set, rather than accidental — and that Hebshie was the perpetrator. To show arson, the government relied on two kinds of evidence — first, the “cause-and-origin” forensics expert who testified about where and how the fire started, and second, the testimony of the handler of an “accelerant-detection dog” and a laboratory technician to prove that it was incendiary in nature. To show that Hebshie was the perpetrator, the government demonstrated that Hebshie leased space in the building for his convenience store, that he owed roughly $5,000 to lottery authorities, that he was trying to sell the store, and that he sought to recover on an insurance policy. The arson evidence had substantial problems, as this decision shows; the evidence of Hebshie’s motive to burn down his store was less than overwhelming. Two Supreme Court cases help frame this decision: the very rigorous standards for determining when counsel’s performance is ineffective described in Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984), and its progeny — and the special requirements for scientific testimony under Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993) and for technical testimony under Kumho Tire Co. v. Carmichael, 526 U.S. 137, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999). Under Strickland, Hebshie must prove (1) that his counsel’s performance was deficient and (2) that he suffered prejudice as a result. 466 U.S. at 687, 104 S.Ct. 2052. To be deficient, an attorney’s conduct must fall below an “objective standard of reasonableness” established by “prevailing professional norms.” Id. at 687-88, 104 S.Ct. 2052. To demonstrate prejudice, Hebshie must show a “reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different.” Id. at 694, 104 S.Ct. 2052. While the Strickland standards are notoriously difficult to apply to the usual strategic decisions of trial, scientific and expert evidence raises fundamentally different questions, and impose a different set of obligations. By 2006, when Hebshie’s trial took place, a number of articles in legal journals and cases cast a critical eye on the scientific reliability of arson evidence, methodologies, and techniques. Ordinarily competent counsel would have understood that men and women had been convicted, sentenced, perhaps even executed, on the basis of flawed arson evidence and taken appropriate steps to litigate the issues using all the tools available. But the case for relief is far stronger here than in the run-of-the-mill arson case; it does not depend upon that literature, those cases or even the drumbeat of concern about arson prosecutions. Hebshie’s trial counsel had been specifically warned about deficiencies in this case by predecessor counsel and by the experts retained by them. Even more significant, when the Court pointedly inquired during the trial, not once, but three times, whether there would be a challenge to the arson expert testimony, the laboratory analysis, or the canine handler’s testimony, even when the Court suggested it was willing to suspend the trial in order to conduct such a hearing, counsel declined. Trial counsel never challenged the validity or reliability of the testimony of Sergeant David Domingos (“Domingos”), the government’s “cause-and-origin” arson expert; John Dragan (“Dragan”), the laboratory technician; or Sergeant Douglas Lynch (“Lynch”), the accelerant-detection dog handler, either prior to or during the trial, much less called for any limitations on their testimony even when there were substantial grounds to do so. They never objected to testimony that went far beyond the science on which it was allegedly based. Not a single pretrial motion or proposed jury instruction was filed on expert issues at all. Nor did counsel object when the government fundamentally mischaracterized critical testimony from the canine handler in its closing. (In fact, Hebshie’s counsel legitimized the government’s experts when he addressed the jury; the testimony was “impressive,” he said; he did not “quarrel” with it.) While one might imagine a strategic reason for not seeking a Daubert hearing under some circumstances, or for not objecting to certain testimony, absolutely no legitimate reason was argued here. In fact, trial counsels’ “explanations” (offered at the evidentiary hearing on the habeas petition) were, in a word, incoherent. The most significant problems should have been abundantly clear — problems with the core evidence of arson, namely, the canine evidence and the laboratory analysis. Once Billy the dog sniffed a so-called accelerant in the left side of Hebshie’s store, the area to which she had been led (the very area that Domingos concluded was where the fire started), government experts never checked other areas for “accelerants.” They took no control or comparison samples from anywhere else, as the scientific method and arson investigation standards suggest. That single sample was then tested and found to be “light petroleum distillate.” “Light petroleum distillate,” however, is a category so broad that a host of entirely benign substances fit within it, especially in this case. Hebshie’s store was a convenience store, after all; light petroleum distillate could be found in a number of the goods he offered for sale (lighter fluid or lighters, for example). In addition, it was a substance generated by the heat of the fire itself, a product of pyrolysis of other materials in the store (like carpet glue). In fact, the laboratory test was only probative of arson if one area tested positive while others did not, or if the test disclosed a chemical that would not normally be present at the scene. Nevertheless, the sample was never analyzed further. Counsel’s errors in these areas — the canine and laboratory evidence — covered both pretrial and trial conduct. With respect to the laboratory evidence, a Daubert/Kumho Tire hearing would have allowed the trial court to screen whether minimum scientific standards were met— not in the abstract, but in the particular context of this case. Likewise, with respect to canine evidence, which may be admitted to show how the dog assisted investigators in the selection of samples for testing, a hearing would have allowed the court to monitor how far the testimony could lawfully range, and what were its appropriate limits. To be sure, holding a hearing does not guarantee that evidence will be excluded, but as to the canine and laboratory evidence, exclusion (or in the case of the dog, strict limitation) was more than a “reasonable probability.” It was likely. Counsel’s errors pretrial in dealing with the arson testimony, in any case, were compounded by their trial errors, especially with respect to Billy, the accelerantdetection dog. It is not an understatement to say that Lynch, the dog handler, was permitted to testify to an almost mystical account of Billy’s powers and her unique olfactory capabilities. He presented unsubstantiated claims about the dog’s accuracy. He was allowed to go on at great length about his emotional relationship with the dog and his entirely subjective ability to interpret her face, what she thought, intended, and the “strength” of the alert she gave in this case. Finally, Lynch was permitted to testify that the dog did not alert to anything else on the premises, as if the dog had been allowed to range widely on the fire scene (she was not), and as if the dog’s failure to alert had evidential value (it does not). The latter theme was repeated in the government’s closing. The government argued: Billy knew what she was smelling for. She indicated an alert on one place and one place alone. It was a strong alert. If it was the carpet that was setting her off, then she would have triggered everywhere else around her.... That was the one place where she alerted to. (Italics supplied). Trial Tr. Vol. 6,171: 5-13. Counsel never objected or moved to strike the testimony or the closing. To be sure, cross-examination disclosed that Billy was only taken to one area on the fire scene, and that only one sample was taken. But the problems with Lynch’s testimony and the government’s closing were fundamental ones. As the Court held in Daubert, some testimony may be so problematic that the usual trial techniques are just not enough to prevent a jury from giving it far more credence than it deserves. See Daubert, 509 U.S. at 596-97, 113 S.Ct. 2786. The testimony should not reach the jury at all. Here, the scientific literature cast doubt on the significance of the dog’s failure to alert (false negatives) and even raised concerns about canine “proficiency” testing, concerns counsel never raised. Nor did counsel argue about the extent to which Lynch’s testimony (regarding the strength of the alert, the dog’s unique prowess, Lynch’s ability to “read him”) coupled with the government’s closing, effectively turned Billy’s alert into substantive evidence of guilt, just what the law and literature caution against. In fact, even if counsel had not moved for a pretrial Daubert hearing before trial, nothing justified standing mute during Lynch’s testimony or the government’s closing. Billy, like the traditional Ouija board, was simply allowed to point to Hebshie as an arsonist. Hebshie also argues counsel erred in connection with the eause-and-origin testimony. For example, petitioner claims the government experts failed to follow the scientific method in the way the scene was preserved. A reasonable hypothesis, according to petitioner’s experts, was that the fire began in the basement rather than in Hebshie’s store, or that even if it had started there, that substances regularly found in a convenience store had accidentally ignited. Yet Domingos failed to preserve the scene in a way that enabled others, including defense experts — not to mention peer reviewers — to evaluate his conclusions. He failed to take photographs of the basement or even mention that he had been there in his investigation reports. In effect, few, if any, investigative materials involving places other than where Domingos concluded the fire had begun were collected before the building was razed to the ground and further investigation was impossible. And these failures were especially significant in the light of the challenges to Domingos’ “interpretation” of the fire patterns raised at the habeas evidentiary hearing. Still, there was no pretrial motion for a Daubert hearing, and no motion for a voir dire during trial, even when the court invited counsel to do so. But the cause-and-origin challenge stands on a different footing than the canine and laboratory evidence. Whether a Daubert challenge to cause-and-origin would have succeeded on these grounds is a closer question. Counsel not only cross-examined Domingos; they offered a rebuttal expert raising the basement hypothesis, and challenging the premises of the government’s cause-and-origin theory. While the defense was hamstrung by limitations in the government’s investigation, still, as the government notes, there was not a complete failure to raise many — although not all — of the salient points highlighted in the habeas hearing. Domingos’ testimony was troubling, to be sure, but not nearly so troubling as the laboratory analysis or the canine evidence. Under the circumstances, I need not resolve whether there was a “reasonable probability” that I would have excluded the cause-and-origin testimony had the issues been teed up before me because the exclusion of the canine and laboratory evidence is so clear. Under the “prevailing professional norms,” reasonably competent counsel should have moved for a Daubert/Kumho Tire hearing before trial on all of the expert testimony — a) on the accelerant laboratory analysis based on the investigator’s failure to use a comparison or control sample and not test beyond the generic finding of “light petroleum distillate”; b) on the canine evidence, highlighting problems with proficiency testing and emphasizing the limited scope of the testimony; and, c) on the expert cause-and-origin testimony, when the expert’s proposed testimony was scientifically flawed. If counsel had requested such a hearing, there is more than a “reasonable probability” that it would have been granted, that the laboratory analysis and the canine evidence would have been excluded, or severely limited, at the very least. Moreover, quite apart from what counsel ought to have done before the trial began, reasonably competent counsel should have at least objected — strenuously—to Lynch’s testimony about Billy when it ranged far, far, beyond its limited purpose, and to the government’s closing argument, which dramatically overstated Billy’s significance. If, during trial, counsel had done so, the objection would have been sustained. The prejudice is clear: Without Billy’s “testimony” or the laboratory analysis, there is a more than “reasonable probability” that the outcome of the trial would have been different. While the cause-and-origin testimony purports to identify where the fire began, the canine evidence and the laboratory results are essential to prove that the fire was an arson, not an accident. Without it, there is simply no crime. In short, I find that due to counsel’s failures, the very danger that Daubert and Kumho Tire sought to avoid occurred: Questionable theorizing about arson, about Billy’s mystical prowess, and the generic laboratory results, were presented as “science” to the jury, and as a result, Hebshie was convicted. Hebshie’s § 2255 petition is GRANTED. II. FINDINGS OF FACT I will first describe the background of events chronologically, then the evidence adduced at trial on direct and cross-examination, and finally, the evidence adduced at the evidentiary hearing on Hebshie’s § 2255 petition: A. Background On Saturday, April 21, 2001, a fire broke out in a two-story commercial building at 32-34 Main Street in Taunton, Massachusetts. Hebshie leased space on the first floor for his convenience store, Main Street Lottery & News Store. Two other businesses, Ro An Jewelers and Downtown Sports Card Shop, were also located at 32-34 Main Street. On the day of the fire, Hebshie left his store at 1:37 p.m., setting the security alarm as he left. According to the alarm company, a motion detector alerted within the store at 1:44 p.m. At roughly the same time, a police officer noticed smoke coming from the building. Lieutenant Todd Myers (“Myers”), a firefighter with the Taunton Fire Department, was the first to enter. Since visibility was poor, Myers, along with other firefighters, crawled into the smoke-filled building towards an “orange glow” at the back of the convenience store-described as the office area, but also an area near what appeared to be the basement stairs. Trial Tr. vol. 2, 15:1-3, 19:1-14. After about ten minutes, Myers reported that the fire had been extinguished. Id. at 20:16-21:11. In short order, however, firefighters noticed smoke in the back of the convenience store and on the second floor of the building, as well. Id. at 9:1-9. After several hours and multiple alarms, they managed to put out the fire, but the building sustained substantial damage and partially collapsed. Domingos, who worked for the Massachusetts State Police Fire Marshal’s Investigative Unit led the “cause-and-origin” investigation. He arrived at around 2:00 p.m. on the day of the fire. Trial Tr. vol. 3, 19:1-3. He directed that statements be taken from witnesses and firefighters. Id. at 21:13-22:13, 23:9-12. Around dusk, he tried to crawl into the convenience store but debris from the second floor that had fallen during fire suppression efforts blocked his entrance. Id. at 23:1-25:11. He returned the next day, April 22. Domingos concluded very early on that the fire had started in the left-hand wall in Hebshie’s store initially because he believed it was the area of greatest damage and because of his interpretation of certain “burn patterns.” He took photographs, including of the area where the basement stairs were, but not any of the basement itself. While he testified that he did in fact go into the basement and discounted it as the origin of the fire, his report makes absolutely no mention of the basement’s condition, his observations, or that he entered the area at all. The stairwell to the basement was not even in the government’s rendering of the store’s floor plan. In fact, based on the investigatory materials from Domingos, the defense would have had no way of knowing that anyone even entered the basement. Domingos called on an “accelerant-detection” canine to assist in his determination of whether the fire was arson or an accident. “Accelerant-detection” canines are supposed to be used to help investigators narrow the search area for ignitable liquids. Once a dog alerts to an area or areas, samples are taken and tested. While the dog is described as an “accelerant-detection” dog, in fact what the dog is trained to alert to is actually a class of chemicals, some of which can be used as an accelerant in an arson, but some of which also have entirely benign purposes and are part of the area’s ambient environment. Moreover, the dog’s “results” depend upon two additional factors — how usual or unusual it is for that kind of chemical to be on the scene at all and whether the “accelerant’s” location was affected by the fire suppression efforts, i.e. the fire hoses or the excavation, or whether it pre-dated the fire. The dog’s handler, Sergeant Lynch of the Massachusetts State Police, brought “Billy” the dog only to an area that was safe for her, also taking into account the area Domingos wanted to evaluate. Trial Tr. vol. 3, 178:19-23. The safe area was the left side of the store, the area that Domingos had specifically cleared and where Domingos believed that the fire had started. In fact, this was the most intact area of the convenience store; virtually everywhere else was filled with debris. Billy alerted Lynch to one spot along the wall. Lynch took a single sample for laboratory analysis from that spot. Neither he nor Domingos directed the dog to any other areas in the building-like the basement-apart from the convenience store’s left side to see whether she would register the same kind of “alert.” Trial Tr. vol. 4, 27:8-16. Nor did Lynch bother to take a sample from other areas in the store to act as a control or a comparison sample. Taking control samples, he noted, just was not part of his routine. And in this case, he did not do so in part because of his subjective assessment of the “type” of alert that Billy gave, and because Domingos had not asked him to. (Significantly, Domingos denies directing him to take only one sample.) The State Police Crime laboratory took that one sample and identified the general class of chemicals to which it belonged— light petroleum distillate — but not the specific chemical. No further tests were requested, tests that might have narrowed the field and distinguished the sample from the usual products in the convenience store that also contained “accelerants,” or the products of pyrolysis. On Sunday night, April 22, 2001, Domingos left the scene and released it to the insurance company. The building was demolished soon thereafter. B. Pre-Trial On May 29, 2002, Hebshie was indicted for arson, mail fraud, and use of fire to commit a felony. Robert Muse, and then his son, Peter Muse, were appointed to represent him. In order to help prepare the defense’s case, Peter Muse hired two forensic experts — John Titus (“Titus”), a fire protection engineer, and Michael Higgins (“Higgins”), President of K-Chem Laboratories. Evid. Hr’g Tr. 215:7-12. In March 2005, Hebshie replaced Muse with retained counsel, John T. Spinale, who in turn hired his son, Jay Spinale, as co-counsel. The trial began over a year later, on June 12, 2006. The Spinales kept Titus on as a consultant and expert witness but did not retain Higgins. Id. at 132:2-12. Titus (testifying at the evidentiary hearing) was adamant that he had expressly warned the Spinales about flaws in the government’s eause-and-origin evidence, the canine evidence, and the laboratory analysis, over and over again, both before and during the trial. Id. at 269:12-24. The Canine evidence was especially problematic: First, a handler has to interpret the dog’s reactions. Id. at 263:9-264:14. Second, simply because a dog “alerts” to accelerant does not resolve “whether or not the presence of that chemical has anything to do with incendiary intent or [is] simply part of the ambient environment for that place.” Id. at 262:14-263:5. In fact, Titus reiterated that the latter — the ambient environment of the convenience store — would include “thousands” of materials to which the dog might alert. Id. at 266:16-19. Titus also flagged the prejudicial use of the term “accelerant,” as in “accelerant-detection dog,” which makes the conclusion of arson, rather than an accidental fire, all the more likely, even when the substances alerted to may be entirely benign. Id. at 265:13-266:5. In addition to the canine evidence, Titus highlighted the problems with the testing in this case — first, the fact that the laboratory identified only the general category of accelerant, light petroleum distillate, but not the specific chemical within that category, and second, that there were no comparison or control samples taken from any place other than where Domingos suspected the fire to have begun. He again warned them that the laboratory test was probative only if it identified an accelerant that was not present elsewhere in the store. Id. 268:15-269:6. Spinale’s files contained a laboratory analysis from K-Chem Laboratories, an analysis that Hebshie’s prior attorney had commissioned, which drew the same conclusions. Despite these warnings, the Spinales never filed any pretrial motions (except motions to obtain incidental relief from Hebshie’s pretrial release). There were no motions in limine, no motions directed to the expert testimony whatsoever. They never requested a Daubert hearing to challenge the bona fides of the Domingos investigation or the canine evidence, never moved in limine to exclude the laboratory analysis because of the generality of its conclusions and the lack of a control sample. They did not use the expert from K-Chem Laboratories that Muse had hired or the data from the insurance file that they had been given. Nor were there any explanations for these failures' — much less strategic ones. Spinale admitted at the habeas evidentiary hearing that he had just “assumed” that Muse had decided not to pursue a Daubert hearing concerning Domingos’ investigation. Id. at 151:23-152:11. Muse in fact testified that he too had expressly highlighted problems with the government’s forensic case to the Spinales and encouraged them to challenge the evidence pretrial. Id. at 217:20-219:13. Whatever the Spinales concluded at the outset of their representation concerning scientific evidence, they never reconsidered it during the year and three months after they took over from Muse. Nor did they reconsider it, even after the Court asked them three times during the trial whether they intended to challenge the cause-and-origin evidence, the canine evidence, and Domingos testimony, and indicated that the Court was prepared to suspend the trial for a hearing. C. Trial At trial, the government’s witnesses included: Myers, the first-in firefighter; Domingos, the cause-and-origin investigator; Lynch, the canine handler; Drugan, the police laboratory technician who tested the carpet sample to which Billy the dog had alerted; and a rebuttal witness, Wayne Miller (“Miller”), a former agent of the Bureau of Alcohol, Tobacco, Firearms and Explosives (“ATF”). The defense put on Titus, the expert whom the Muses had found. The government’s theory was that Hebshie used an accelerant to burn down his store in order to collect on his $30,000 insurance policy. The only evidence of financial exigency on the record was that Hebshie had lost his license to sell lottery tickets a few weeks before because he owed $5,000, and that his efforts to sell the store the week before the fire had been delayed because the putative purchaser had a death in the family. The government claimed that Hebshie set the fire along the left-hand wall in his store. The defense’s theory was that the fire had started accidentally in the basement. But because of the limited scope of the Domingos investigation, Titus had little proof — no photographs of the basement itself or comparison samples from elsewhere in the convenience store (and no opportunity to do an independent evaluation of the scene). Indeed, the government seized on the limitations of Titus’ testimony in its closing, limitations for which it was entirely responsible: “Mr. Titus, ladies and gentlemen, came to no conclusion about what the cause of the fire was.... [A]t no point did he say this is where the fire started.” Trial Tr. vol. 6, 172:10-15. The government’s case on cause-and-origin and arson was based on four areas of evidence: (1) building construction and Domingos’ interpretation of burn patterns; (2) the canine evidence; (3) the laboratory testing of the accelerant sample; and, (4) Myers, the first-in firefighter’s account. 1. Domingos’ Testimony: Building Construction and Burn Patterns Domingos’ job was to interpret what would have looked like a “pile of rubble” to the “untrained eye” the day after the fire had been extinguished. Trial Tr. vol. 3, 82: 4-7. He concluded almost immediately that the fire had started along the left-hand wall in Hebshie’s store, moved several feet to the pipe through the wall and through that to the second floor. Id. at 40:13-41:16. He discounted the fact that, in between the extinguishing of the fire on Saturday and his return on Sunday, part of the building had collapsed, leaving the site different than the night before. Id. at 31: 15-19. That change, he concluded, did not matter because the collapsed area “was not specifically related to our first noticed area of origin from the first-in fire department.” Id. at 32:6-13 (italics supplied). Domingos began by examining the exterior of the building. Id. at 32:14-20. Then, he walked through the sports memorabilia store and the jewelry store and ruled them out as possible areas of origin. Id. at 33:8-36:25. He next entered the convenience store, directing the excavator operator to clear out some of the “roof material and timber material” that had dropped into the scene, clearing the rest with hand tools. Id. at 37:25-39:22. Domingos’ observations confirmed his theory of the fire’s origin. He pointed to the “heavy damage and charring,” id. at 40:13, a “low burn or area that was damaged close to the bottom of the floor level,” id. at 41:5-11, and a “V” pattern on the wall above the hypothesized origin of the fire, id. at 4L14-20. Domingos ruled out the basement as the origin of the fire (although as noted above, nothing in any of his reports reflected that he had even been there). Spinale cross-examined Domingos. Domingos agreed that the area he had identified as the most damaged was in fact an area where sections of the wall were most intact, id. at 65:19-66:3, where even the flimsy paneling had not been burned through, id. at 94:6-95:5, nor had a stack of papers, id. at 95:6-15, or nearby plastic bags, id. at 96:3-97:2. He also acknowledged that the most damaged area may not be the area where the fire started because the burn pattern may be affected by other factors, id. at 84:19-85:12, which could also lead to false identification of cause-and-origin. For example, “V” patterns can be caused by factors having nothing to do with the fire’s origins-ventilation or the fact that some materials burn more easily. In fact, he agreed that there was no real “V” in the picture at all. Id. at 89:9-91:4. He conceded that there was significant burning in the stairwell area leading to the basement, id. at 102:16-25. And as to Billy, he agreed the dog was permitted to go only where her handler deemed it safe for her. Id. at 82:4-83:9. Domingos discounted the theory that the fire was an electrical fire starting in the basement, even though, as noted, his written report did not mention the basement at all. Id. at 107:22-108:19. He agreed that there was a “V” pattern on the electrical outlet on the wall created by the fire coming in through the outlet, id. at 92:20-24, but concluded that the pattern meant only that the fire had gotten into the wall, not that the source of the fire was an electrical malfunction, id. at 93:6-14. He noted that the “burnt toast” smell that several witnesses reported an hour before the fire could have been an electrical problem, id. at 107:15-21. Most important, Domingos acknowledged that control samples were not taken, id. at 98:14-16, even though he conceded they were “advisable,” and that their purpose is to make certain whatever “accelerant” is found is not part of the ambient environment, id. at 98:23-99:1, 119:15-20. He agreed that many household products have accelerants. Id. at 97:14-17. He admitted that he taught his students to take control samples, id. at 99:3-10, and that he had done so in the past, id. at 99:25-100:3. In fact, he acknowledged he did not know why samples were not taken in this case. Id. at 120:5-7. When Domingos was asked why he took no control samples, he said only, ‘You’ll have to ask Trooper Lynch that question.” Id. at 120:7. Trooper Lynch, as described below, points the finger at Domingos. 2. Canine Evidence Lynch testified at great length about Billy, the accelerant-detection dog. He emphasized both the dog’s unique powers and his singular ability to read her reactions, moods, and sounds. Before he described Billy’s prowess, in particular, he went on about his experiences with other dogs, including dogs that were not involved in arson investigations, like a narcotics-sniffing dog and explosive-sniffing dogs. He talked about Amor, his first dog, a narcotics-sniffing dog, imported from Germany because Germany “had a very solid, proven product.” Trial Tr. vol. 3, 143:8-16. He worked with Amor until the dog “died on me in 1995.” Id. at 114:19. In fact, he described the special olfactory qualities of all of the dogs used for these investigations. Then he focused on Billy, and made his emotional attachment to her abundantly clear. (She had died before the trial. He lived with her day and night.) Id. at 146:17-25. He underscored his ability to communicate with her. Trial Tr. vol. 4, 12:2-8. He could “read her face,” Trial Tr. vol. 3,150:14-16; he knew “her personality,” “the way her eyes shifted,” “the way her ears shifted when she located stuff.” Id. at 150:20-22. He could interpret her mood even in a dark room, just from the sounds she made. Id. at 15L17-22. And there were times when she was “visibly more confident in her alerting on something than on other occasions,” id. at 151:5-25, when she gave particularly strong alerts. Then he vouched for Billy’s accuracy. She had been “certified,” although the basis of the certification was not clear. Id. at 158:5-20, 159:13-15. And when counsel asked, “Over the course of your working with Billy, do you know how accurate she was,” he categorically stated that she was 97 % accurate. Id. at 156:7-9. In fact, he noted that if Billy were wrong it was the handler’s error not hers. Id. at 156:16-20. He based the latter conclusion on the fact that he had “spoken to lab techs .... in both Connecticut and Mass.” Trial Tr. vol. 3, 156: 13-14. No reports or tests were introduced. Lynch would only take Billy to areas that were safe. Significantly, the area to which Billy was directed was an area that Domingos directed the excavator to clear (aided by the hand shovels of Domingos and others) and precisely where Domingos concluded the fire had started. Billy was employed just “in the cleared area on the left side of the store,” indeed the more “intact portion” of the store. Trial Tr. vol. 4, 26:21-22, 27:8-16. Lynch stated that when Billy passed the spot near where Domingos had identified a “V-pattern,” the dog reacted in a way that Lynch interpreted to be an “alert” that accelerants were present. He characterized this as a “strong alert.” Trial Tr. vol. 3, 152:5-8. The spot was the same place that Domingos observed “plastic appeared to have dripped down to the floor,” id. at 97:3-5, which he later acknowledged could be mistaken for an accelerant. (See supra n. 15.) He testified — without objection— that the dog did not alert to anything else on the premises. Trial Tr. vol. 4, 13:7. Lynch cut a single sample from the carpet in the area where Billy alerted. Id. at 18:2-3. He did not take any other samples from nearby spots to act as control or comparison samples. He did not do it in this case because of the type of alert the dog gave and because the principal fire investigator had not requested comparison samples. Trial Tr. vol. 3, 169:23-170:13, 20. Although counsel cross-examined Lynch, challenging the fact that only one sample was taken, and that Billy was limited to certain areas, he never objected to the breadth of the testimony, its emotional quality, or the irrelevant references to other dogs. He never challenged the 97% accuracy figure or addressed the scientific literature on false negatives — when the dog failed to alert. Indeed, Spinale did not object, even after the Court asked: THE COURT: I wanted to make sure I didn’t miss anything here, there was no challenge to the canine evidence, to the dog? MR. SPINALE, JR.: No. Id. at 12L15-18. After Lynch’s testimony, the Court inquired again as to whether there would be a challenge to the fact that only one sample had been taken: THE COURT: So there’s a single sample taken from where this witness said the dog had identified the accelerant? PROSECUTOR: Yes. THE COURT: That’s stipulated to, so you don’t have a counter on that? Id. at 181:17-22. 3. Drugan: Accelerant Sample The single carpet sample from the convenience store was sent to the Massachusetts State Police Crime Laboratory on April 23. Drugan, a police laboratory technician and supervisor, testified that, using a gas chromotograph flame ionization detector, he classified the sample as a light petroleum distillate. Trial Tr. vol. 4, 44:5-25. He did not identify the substance beyond that. The government submitted Drugan’s lab report as an exhibit. Gov’t Trial Ex. 14. Spinale cross-examined the expert about the lack of a comparison sample. Drugan admitted that using a comparison sample is the “preferred practice.” Id. at 51:5-6, 19-21. Spinale never challenged the reliability or admissibility of the test itself. Indeed, Spinale stipulated to the lab results. Trial Tr. vol. 3,181:14-16. 4. Myers’ Testimony: Thermal Imaging Myers, who was the first-in firefighter, testified to his observations as well as interpretation of observations made with a thermal imaging camera brought in by another company of firefighters after he believed the fire had been extinguished. The camera showed white “hot spots” in all four walls, “pretty much around the place depending upon what you shined it on, pretty much 360 degrees,” Trial Tr. vol. 1, 106:16-18, which according to Myers, meant just that the walls “had been heated” from the flames in other locations, but that the fire had not spread into the walls. Trial Tr. vol. 2, 21:22-22:3. While Myers believed that these results were consistent with an extinguished fire, in fact, the fire continued. Id. at 9:5-9. Spinale cross-examined Myers, suggesting that visibility was poor, that he had seen “the glow” of the fire in an area of the office through a door to an area near the basement stairs. Id. at 27:22-28:12. Spinale, however, never challenged Myers’ testimony concerning the meaning of “hot spots,” relying on his expert, Titus, to give a different interpretation of what those “hot spots” meant, namely, evidence of fire in the walls, consistent with a basement fire traveling internally through the building. Trial Tr. vol. 6, 22:14-22. 5. Other Evidence In addition to evidence concerning cause-and-origin, the government offered the testimony of Deputy Chief Nastri, who was the ranking officer in charge of the scene. His testimony included a reference to the fact that one of the responding firefighters had had a heart attack during the fire suppression efforts, which he survived; this testimony was not objected to. Dave Lambert, the owner of Ro An jewelers testified to a “burnt toast smell” an hour before the fire started (which Domingos conceded could be the smell of an electrical fire). Domingos testified concerning Hebshie’s statements, that he had not stated that he had been doing painting in his store (paint cans being a possible source of fire), although there had been paint cans in the basement, that Hebshie began his day on the morning of April 21, leaving at 1:30 p.m., and setting the alarm with a code that identified him as the last person in the store. Trial Tr. vol. 3, 57:23-58:25. In addition, Hebshie told Domingos that lottery officials had removed his license to sell lottery tickets because he owed $3,500, id. at 60:3-6, and that a potential sale of the business did not go through because there had been a death in the family of the purchaser, id. at 60:9-12. David Lampert, the owner of the Ro An Jewelers, testified that on the day of the fire Hebshie “pretty much said that he’s sorry that the whole building went down and that you never think that a fire could take down a building like that.” Trial Tr. vol. 2, 133:1-4. 6. Defense Case Titus, testifying for the defense, disagreed with Domingos’ interpretations of the burn patterns and his theory of the fire’s progression. He opined that the evidence pointed toward the basement as the fire’s source. a. Burn Patterns Titus pointed out that the store’s left-hand wall was not the area of greatest damage. The three other walls in the store collapsed; the left-hand wall was the only one left standing, the most “intact area,” as Lynch noted. Trial Tr. vol. 4, 26:21. That the supposed area of origin remained standing suggested that the fire did not in fact start there. Trial Tr. vol. 6, 49:25-50:5. In addition, Titus pointed to two burn patterns suggesting the fire began in the basement. First, the burn pattern on the wall just above the heating grate aligned almost exactly with the outline of the grate. Id. at 33:16-34:1; Gov’t Trial Ex. 1-14 (suggesting that fire came up through the grate and burned there for some period of time). Id. at 36:10-23. This would also explain why there was significant damage along that part of the wall. He also pointed to a burned-through V-pattern on the far side of the left-hand wall as evidence that the fire broke out into the store from inside the walls. See Gov’t Trial Ex. 1-10; Trial Tr. vol. 6, 31:16-32:1. b. Thermal Imaging As Titus explained at trial, the presence of the hot spots on all four walls and the continuation of the fire appeared to be more consistent with the defense’s theory that the fire began in the basement. The white color on the thermal imaging camera represented high heat (that is, a burning fire). If the walls were warm from only residual heat, as Myers indicated, they would not have shown up so brightly on the camera. Trial Tr. vol. 6, 21:21-22:22. Further, hot spots on all four walls (as opposed to one or two), suggested that the fire was coming up from below, from the basement. c. Laboratory Analysis Titus also testified that a control sample was an “essential” part of the scientific method and was necessary in order to validate results. Id. at 56:8-21. Without any evidence about the basement’s condition, photographs, samples, or even notes, Titus was able to argue only some inconsistencies in the government’s theory. He could not present much proof of his own. 7. Government’s Rebuttal Wayne Miller, the government’s rebuttal witness at trial and a former ATF agent, testified that, based on the materials he reviewed, he did not see anything to suggest the fire started in the basement. Id. at 127:6-9. Like Titus, Miller was able to review only the photographs and notes the investigators produced. He had no original information. Miller admitted that Domingos should have documented the state of the basement. Id. at 138:10-23. D. Closing The government placed special emphasis on Billy the dog, implying that she had alerted to one space to the exclusion of all the others — which was not the case. Billy had not been shown “all other” areas: Billy knew what she was smelling for. She indicated an alert on one place and one place alone.... If it was the carpet that was setting her off, then she would have triggered everywhere else around her.... She triggered on one place, it wasn’t on the heating grate, it was right underneath those newspapers. She didn’t trigger on anywhere near the counter, she didn’t trigger down the stairwell. That was the one place where she alerted to. Id. at 171:5-13. Spinale did not object to the closing. And the government reiterated the 97% accuracy figure for Billy: “They send [the sample] in to the lab and like 97 percent of the other times that she’s ever done this she was right again.” Id. at 13-15. Again, Spinale stood silent. When it came time for Spinale to address the jury, he conceded Domingos’ qualifications (“a very qualified person ... I don’t quarrel with any of his testimony ....”), id. at 188:11-13, that Lynch and his dog were “impressive,” id. at 189:10, that there really was not much difference between the experts for each side, id. at 198:1-5, and that it was just about common sense, id. at 198:10. E. Verdict The jury found Hebshie guilty on all counts. He was sentenced to a mandatory fifteen years in prison. Hebshie appealed, and on December 4, 2008, the First Circuit affirmed his conviction and sentence. United States v. Hebshie, 549 F.3d 30 (1st Cir.2008). F. Habeas Evidentiary Hearing Both parties presented testimony at an evidentiary hearing held over four days to address: (1) whether defense counsel’s performance was deficient; and (2) if so, whether the deficient performance resulted in prejudice. See Strickland, 466 U.S. at 687, 104 S.Ct. 2052. As to the first prong, Hebshie presented testimony from his trial attorneys, John T. and Jay Spinale; his previous attorney, Peter Muse; and an evidence professor, Professor Jane Moriarty to identify any possible justifications for the Spinales’ failure to move for a Daubert hearing, establish the standard for competent counsel in an arson case, and determine what the Spinales knew about the deficiencies in the government’s investigation. As to the second prong, Hebshie presented testimony from John Lentini (“Lentini”), a forensics arson expert; Titus, the defense expert at trial; and Daniel Cronin (“Cronin”), the insurance investigator, to demonstrate what a Daubert hearing would have shown. The government called Domingos, the lead fire investigator at the scene who also testified at trial, and Michael Marquardt (“Marquardt”), a special agent with the ATF. 1. Testimony Re: Counsel’s Performance John T. Spinale had the “primary responsibility” for the case; his son, Jay Spinale, helped him in analyzing the scientific evidence and expert witnesses. Evid. Hr’g Tr. 137:18, 170:23-24, 174:6-8. Both testified that they had spent many hours researching and preparing Hebshie’s defense but were unable to give specifics. John T. Spinale did remember, however, that he and his son “relied very heavily” on the National Fire Protection Association’s “Guide for Fire and Explosion Investigations” or NFPA 921. Id. 137:1-8, see infra note 39. John T. Spinale’s strategy at trial was to attack the quality of the investigation and the validity of the government’s theory about the cause of the fire, and to rebut the motive evidence. Evid. Hr’g. Tr. 129:6-130:10, 131:11-18. When asked why he did not bring a Daubert challenge to the cause-and-origin testimony and the arson evidence, Spinale stated that he “assumed” predecessor counsel Muse had decided not to pursue a Daubert hearing. But he did not believe he had ever spoken with Muse about it. Muse categorically contradicted him. Muse never told Spinale that he had already ruled out filing a Daubert motion or suggested not filing one. Id. at 218:24-219:4. In fact, he was “certain” he had discussed challenging the government’s experts with the Spinales. Id. at 219:1-4. Titus underscored Muse’s warnings. He warned Spinale about the misuse of canine testimony, the lack of control samples, and the lack of adequate investigation of alternative hypotheses. Indeed, Spinale conceded there was no strategic reason for the failure to try to exclude or object to the expert testimony: Q. (Hebshie’s habeas counsel) And is it your testimony that if you had had a basis in law or fact to exclude the government’s expert testimony, you would have done so? A. (John Spinale) Certainly I would have objected to or moved to exclude testimony if I thought there was a basis for excluding it. Q. (Hebshie’s habeas counsel) So there was no strategic reason to not do so; am I right? A. (John Spinale) We did not strategize to allow that in. Id. at 168:8-15. When Jay Spinale was asked why he had not moved for a Daubert hearing on canine evidence or objected to it, he stated that his research had shown it to be admissible. Id. at 178:17-179:6. But when asked about why he had foregone a Daubert hearing on the cause-and-origin testimony and the accelerant test, he equivocated. At one point, he stated that he thought the government’s experts made the wrong conclusions but not that their testimony was unreliable, id. at 198:21, 199:2; and at another point he stated that he believed that the arson evidence was not the core of the government’s case, id. at 199:18-200:11. I do not credit Jay Spinale’s testimony at all. He admitted that he had taken pains not to prepare for the evidentiary hearing in any way; he had reviewed nothing. Id. 204:3-22. 2. Testimony Re: Prejudice a. Cause-and-origin Testimony (1) Basement Theory Cronin, the insurance company investigator, called Domingos to discuss the fire. Domingos told him that it had started in the convenience store. Evid. Hr’g Tr. Ex. 8 at 1; Evid. Hr’g Tr. 234:3-5. Domingos, nevertheless, asked Cronin to call him when he went to the scene because Domingos wanted to go into the basement. When Cronin did so, Domingos did not show up. Evid. Hr’g Tr. 234:11-235:21; see also Evid. Hr’g Ex. 8, at 1 (“Custody of the fire scene had been released; however, Domingoes [sic] requested a call when the scene exam was scheduled, as he wanted to get into the basement.”). Cronin testified that it was his usual practice to document the conditions of the basement, but here, he was unable to take a single photograph. Evid. Hr’g Tr. 237:3-12. The building was then demolished; the basement was inaccessible. Cronin did, however, take one photograph of the stairs leading down to the basement. See Evid. Hr’g Ex. 13. In fact, it was a photograph with a fuller view of the stairwell. Spinale never asked for any of the photographs from the insurance investigation. Titus testified that he had never seen the insurance photographs until the evidentiary hearing. And the photographs were significant. They showed that the lintel, the bar by the basement staircase that supports the floor, was charred and had significant soot deposits on it, indicating that fire came out of the basement. Lentini, Hebshie’s second arson expert, testified that the two photographs of the basement stairs (Domingos’ and Cronin’s) strongly suggested that the fire began in the basement. See Gov’t Trial Ex. 12G; Evid. Hr’g Ex. 13. Burning along the lintel and along the top of a ladder in the staircase indicated fire had come up from the basement. Evid. Hr’g Tr. 110:8-18. If the fire had started on the first floor, then the stairs down to the basement would not show the signs of fire damage that they did. Domingos reiterated his view that he did not bother to take photographs in the basement because there were no signs of fire there. Id. at 380:2-381:2. He conceded that he never mentioned his observations of the basement in any report and that the basement stairwell was absent from the government’s floor plan of the store introduced at trial. Agent Marquardt, a long-time ATF agent and fire investigator, testified for the government, discounting the basement hypothesis based on Myers’ (the first-in firefighter) observations. If the fire had begun in the basement, it would have continued to burn because Myers did not extinguish any flames in the basement. Id. at 464:13-15. Marquardt also testified that if there had been fire in the basement, “heavy heat and smoke” would have been “pouring up through that stairwell into that back area,” yet Myers did not report seeing any fire back there. Id. at 465:4-7. Further, the “floor should have collapsed into the basement.” Id. at 466:15-17. (2) Building Construction and Burn Patterns Lentini, the defense expert, testified that, in fact, the left-hand wall was not the area of greatest damage, precisely because the walls in this area were still intact. Id. at 50:21-24. Indeed, he asserted that the government experts’ theory of the fire’s movement was scientifically impossible. If the fire entered the pipe chase, it would go up, where it had a clear space not five feet to the left through the wall’s vertical studs, as Domingos suggested. Id. at 36:-3-13. Lentini also testified that Domingos’ theory and testimony with respect to the movement of the fire was based on a faulty understanding of fire dynamics. Id. at 90:2-9. First, the wall paneling next to the supposed point of origin was quarter-inch plywood paneling, which burns extremely easily. See Gov’t Trial Ex. 1-14; Evid. Hr’g Ex. 4, 5. If the fire had started there, the paneling would have been burned all the way through. Instead, it was only partially burned. Evid. Hr’g. Tr. 36:23-37:10. Second, the government experts concluded that the fire had started on the first floor and then burned upward, based on the “V-pattern.” According to Lentini, there was not much of a ‘V-pattern;” the fire did not interact at all with a shelf just above the alleged point of origin. See Gov’t Trial Ex. 1-14; Evid. Hr’g Ex. 4, 5. Typically, a fire would burn upward and then spread out when it hit the shelf; photographs suggested that that did not occur. Evid. Hr’g Tr. 37:10-15. The government’s theory would have required the fire to burn up near the alleged point of origin, bypass the shelf, then fall down again near the pipe chase. According to Lentini, it was far more likely that the fire started in the basement, came up through the wall, and then broke out of the wall into the store, a conclusion he contended was reflected in the V-pattern where the wall had burned. Id. at 38:19-39:11. Gov’t Trial Ex. 1-10; Evid. Hr’g Ex. 4. (3) Thermal Imaging Lentini disagreed with Myers’ interpretation of the thermal imaging “hot spots.” If the fire began along the left-hand wall of the store, the thermal image camera would show a “bright spot where the flame was,” then the image would darken as it moved farther from that point. Evid. Hr’g. Tr. 47:2-7. Instead, there were white spots all around. The only way for all four walls to show hot spots would be if the entire room were on fire, an observation consistent with a basement fire traveling through the walls. Id. at 46:5-47:7; see also id. at 289:9-12 (Titus testifying) (“[T]he only way for the fire to get into the walls and generate those hot spots is to be able to move up from a lower area where it has access to those channels between the wall studs.”). b. Arson Evidence (1) Canine Evidence To the Spinales, Titus indicated great concerns about the use of accelerant-detection canines in the scientific community, id. at 264:1-14. Titus directed their attention to NFPA 921. Id. at 259:25-260:7. NFPA 921 circumscribes the use of canines; they are meant simply to be tools to help investigators narrow the search area for ignitable liquids. NFPA 921 § 14.5.3.5 (“Canine ignitable liquid detection should be used in conjunction with, and not in place of, the other fire investigation ... methods .... ”); id. (“The proper objective of the use of canine/handler teams is to assist with the selection of samples that have a higher probability of laboratory confirmation than samples selected without the canine’s assistance.”). Titus gave the Spinales an article from the Journal of Forensic Sciences that highlighted the varying levels of reliability in accelerant detection depending on the substance in question and the canine handler. Evid. Hr’g. Tr. 259:1-7.; Evid. Hr’g Ex. 12 (Michael E. Kurtz et al., Effect of Background Interference on Accelerant Detection by Canines, 41 J. Forensic Sci. 868 (1996)). Titus told them that concerns arose from the fact that the handler has to interpret the dog’s actions. Evid. Hr’g Tr. 262:5-266:23. And he especially underscored the prejudice of allowing the testimony to be described as “accelerant detecting.” What investigators refer to as “accelerants” actually represent a wide range of common and frequently benign materials. In addition, such chemicals can be created by the breaking down of materials during a fire, such as decomposing carpet and other adhesives. “Unlike explosive-or drug-detecting dogs, these canines are trained to detect substances that are common to our everyday environment.... [Mjerely detecting such quantities is of limited evidential value.” NFPA 921 § 14.5.3.5. For these reasons, NFPA 921 requires not just laboratory corroboration, but also comparison samples. I'd And it emphasized concerns about the canine’s reliability — false positives and false negatives. Id. at § 14.5.35; see also, Evid. Hr’g. Exh. 12. (2) Laboratory Analysis According to petitioner’s experts, the lack of a comparison sample did not just call into question the investigator’s conclusions; it fundamentally undermined its validity, a “critical error,” according to Lentini. Evid. Hr’g Tr. 64:22-24. He pointed out that the convenience store had items that would also have tested as “light petroleum distillate.” He listed building materials and glue as an example, but noted particular concern that a can of Zippo or Ronson lighter fluid might have been knocked over by the high powered hoses and spread throughout the store with the water. Id. at 48:7-16. Lentini also pointed out that NFPA 921 emphasized the need for comparison samples in such a case as the one at bar. NFPA 921 § 14.5.3.4 provides: The collection of comparison samples is especially important in the collection of materials that are believed to contain liquid or solid accelerants. For example, the comparison sample for physical evidence consisting of a piece of carpeting believed to contain a liquid accelerant would be a piece of the same carpeting that does not contain any of the liquid accelerant. Comparison samples allow the laboratory to evaluate the possible contributions of volatile pyrolysis products to the analysis and also to estimate the flammability properties of the normal fuel present. Lentini noted that there was carpet and flooring away from the alleged area of origin that could have been collected. Evid. Hr’g Tr. 45:19-21. In fact, he also pointed out that the carpet