Litif v. United States

BOUDIN, Circuit Judge. The United States seeks review of a judgment awarding $1.15 million to the family and the estate of Louis Litif (“the Litif estate”), resulting from Litif s murder by James “Whitey” Bulger in 1980. Litif v. United States, 682 F.Supp.2d 60 (D.Mass.2010). The Litif estate cross-appeals, seeking an increase in the damage award. This case, like its companion case decided by us today (“Davis-Hussey ”) and others already decided, 1 stems from *41 FBI agents’ sheltering of Bulger and his associate Stephen Flemmi over several decades. In the 1970s, Louis Litif worked as a bookmaker in the Boston area and was involved with Bulger’s Winter Hill gang. In September 1979, Litif was charged with murder in the death of James Matera. While out on bail, Litif secretly offered to cooperate with the Boston police, in exchange for leniency, in the investigation of an ongoing drug conspiracy involving Bulger and his gang. In March or April of 1980, Litifs attorney, Kevin Curry, met with Boston Police Detective Edward Walsh to discuss this offer. The meeting was also attended by FBI agent John Connolly. Unbeknownst to Litif and Curry, Bulger was a “top echelon” FBI informant, Connolly was his primary handler, and Connolly and others within the FBI were taking measures to protect Bulger from prosecution so that he could continue supplying information. Roughly three weeks after the meeting, Litif was murdered in South Boston. His body, repeatedly stabbed and shot, was discovered the day after the murder in the trunk of his car. At this early stage there was nothing directly connecting FBI agent actions to Litifs death. A news article at the time of Litifs death reported that “Litif had lost the ‘blessing’ of the so-called South Boston gang allegedly headed by James J. (Whitey) Bulger Jr.” Richard J. Connolly, Slaying Victim, Alienated Mob, Prober Says, Boston Globe, Apr. 15, 1980. But the Boston Globe stated that, “according to investigators, there was no evidence ... that he was killed for [losing Bulger’s support] or that he had been murdered by friends of Matera.” Id. The Boston Police Department appears to have stopped investigating Litifs murder in the summer of 1980. Throughout the 1990s, the Boston press and eventually the national news media began to report on alleged ties between Bulger, Flemmi and the FBI. First in news stories and then in judicial proceedings, it emerged that certain FBI agents allowed Bulger and Flemmi to engage in serious criminal activity while Bulger and Flemmi informed on other criminals (and paid at least some of the agents, including Connolly). United States v. Connolly, 341 F.3d 16, 23 (1st Cir.2003). Eventually, it was learned that in some instances Connolly had tipped off Bulger and Flemmi about cooperating witnesses against them. In 1997, the Boston Globe ran a front-page story about Litifs death. Shelley Murphy, In '80s, FBI saw Bulger as both informant and murder suspect, Boston Globe, Oct. 3, 1997 at Al. The article said that in 1982 Brian Halloran — a criminal associate of Bulger’s — gave the FBI an eyewitness account of Litifs murder implicating Bulger, shortly before Halloran was himself murdered, allegedly by Bulger. According to the article, Halloran claimed that Bulger murdered Litif because Litif had been freelancing as a drug dealer and because Litif had killed Matera without Bulger’s permission. Id. However, the Globe story identified its sources only as unnamed “law enforcement” officials and noted that the FBI was skeptical of Halloran’s reliability because he was facing murder charges at the time of his cooperation, was abusing drugs, and was unwilling to take a lie detector test. Murphy, In '80s, at B7. Halloran himself was killed in 1982. The Globe story also recounted that in 1980 a Boston Herald *42 reporter had investigated stories about Bulger’s involvement in the Litif murder and had been threatened by Bulger. The reporter acknowledged to the Globe that the encounter occurred, but refused to comment on the details. Id. In the late 1990s, District Judge Mark Wolf held the first extensive judicial proceedings (the “Salemme proceedings”) documenting the relationship between Bulger, Flemmi and the FBI. 2 In April 1998 testimony, FBI Agent John Morris implicated the FBI and, in particular, both himself and Connolly in covering up crimes by Bulger and Flemmi, Morris also testified that Connolly told Bulger and Flemmi of Halloran’s offer to cooperate with the FBI against them, which Morris suspected to be the cause of Halloran’s murder. Salemme, 91 F.Supp.2d at 209-10. Litifs murder was not a significant subject of testimony during the Salemme proceedings, although one Quincy police detective stated in his testimony that Bulger was suspected in Litifs murder, among others. This testimony was reported in the Boston Herald and Quincy Patriot Ledger. Ralph Ranalli, Cop: Whitey linked to IRA gun-running, Boston Herald, June 3, 1998; Robert Sears, Former Quincy detective recalls Bulger connection, Quincy Patriot Ledger, June 3, 1998. The police, however, never charged Bulger with Litifs murder. On August 5, 1998, the Boston Herald, citing “[a] local attorney who requested anonymity,” ran a story detailing the substance of Litifs offer to cooperate with the Boston police just weeks before his murder. Ralph Ranalli, Questions arise over agent’s link to murdered drug dealer, Boston Herald, Aug. 5, 1998. The anonymous attorney said that he told Detective Walsh, in the presence of Connolly, that Litif would implicate Bulger in exchange for leniency. The next day, the Herald reported Connolly’s denial that the meeting ever took place. Ralph Ranalli, Ex-FBI agent says slain dealer was informer, Boston Herald, Aug. 6, 1998. On September 15, 1999, Judge Wolf published his decision in United States v. Salemme, 91 F.Supp.2d 141 (D.Mass.1999). This lengthy decision provided the first detailed, publicly available legal document describing in great detail the relationship between the FBI, Bulger and Flemmi. Litifs murder was not discussed but Brian Halloran’s cooperation with the FBI and subsequent murder was canvassed in considerable detail. Id. at 208-13. The decision recounted and placed in context Morris’ testimony that Connolly had leaked Halloran’s name to Bulger who had then likely killed him. Id. The decision also contained summaries of evidence linking Bulger and Flemmi to the murders of Roger Wheeler, John Callahan, and John McIntyre. Salemme, 91 F.Supp.2d at 208-15. Further, the discussion of the McIntyre murder noted his offer of cooperation to law enforcement officials shortly before his disappearance. Id. at 213-15. Together with Morris’ testimony as to Halloran, the decision warranted at least a strong suspicion of systematic FBI leaks of informant names to Bulger followed by their death at Bulger’s hands. Litifs widow thereafter spoke with her family and engaged an attorney, and the Litif estate filed the administrative claim under the Federal Tort Claims Act (“FTCA”) on September 10, 2001, 28 *43 U.S.C. §§ 1346(b)(1), 2401(b), 2671 et seq.; when no action was taken by the government, the instant lawsuit was filed on September 9, 2002. 28 U.S.C. § 2675(a). After consolidation, discovery and pre-trial motions, the district court conducted a single twelve-day bench trial on the Litif, Davis and Hussey claims in July 2009. Thereafter, the district court found in favor of the Litif plaintiffs. It concluded that “Connolly leaked to Bulger Litifs willingness to incriminate Bulger.” Litif, 682 F.Supp.2d at 70. In reaching this conclusion, the court expressly relied on the trial testimony of Litifs former attorney, Kevin Curry, as to Litifs offer to cooperate; evidence of the relationship between Connolly and Detective Walsh; Connolly’s leaks to Bulger of information regarding other informants; and testimony (see note 6, below) from Bulger associates Kevin Weeks and Stephen Flemmi. Id. Having concluded that Connolly generated this leak, the district court ruled that “there [was] no question that Connolly had a duty of care, breached it, and by breaching it, created a foreseeable risk of injury to Litif.” Litif, 682 F.Supp.2d at 75. In particular, the district court ruled, Connolly breached “the general duty of care not to cause foreseeable harm to others through the criminal acts of a third person, a breach that caused Litifs death.” Id. The district judge then awarded the Litif plaintiffs damages totaling $1.15 million: $500,000 to Litifs daughter for loss of consortium, $250,000 to Litifs son for loss of consortium, $50,000 to Litifs wife for loss of consortium, and $350,000 to Litifs estate for Litifs conscious pain and suffering incident to his murder. The government filed the appeal now before us to challenge the judgment; the plaintiffs have cross-appealed to challenge as inadequate the award for Litifs conscious pain and suffering. The government contests liability on three grounds — that the administrative claim was filed after the statute of limitations had run; that there is insufficient admissible proof that the leak occurred and that Bulger killed Litif; and that the Litif estate failed to meet its burden for liability based on conscious pain and suffering. The limitations and the proof issues are connected but not identical so we address them separately in turn, reserving damage appeals on both sides for later discussion. As for the standard of review, the pure issues of law are reviewed de novo; findings of fact in a bench trial are reviewed for clear error. United States v. Padilla-Galarza, 351 F.3d 594, 597 n. 3 (1st Cir.2003). Of course, an individual claim, defense, or precondition (e.g., the statute of limitations) can turn on issues of law, fact, or law-application — or any combination of them; and on law-application some deference may be afforded to the district court’s judgment. Id. Within this framework, the standard as to particular issues is discussed below as needed. Statute of Limitations. The relevant statutory language of the FTCA provides that “[a] tort claim against the United States shall be forever barred unless it is presented in writing to the appropriate Federal agency within two years after such claim accrues.” 28 U.S.C. § 2401(b). The Litif plaintiffs filed their administrative claim on September 10, 2001. The government contends that their claim accrued prior to September 10, 1999, so that their lawsuit is now time-barred. The pertinent legal doctrine governing timeliness has now been repeated so often (see Davis-Hussey at n. 1) that a brief summary will suffice. Litifs murder occurred in 1980, but the “discovery rule”— *44 applied to FTCA claims, United States v. Kubrick, 444 U.S. 111, 100 S.Ct. 352, 62 L.Ed.2d 259 (1979) — delays accrual “where the fact or cause of an injury is unknown to (and perhaps unknowable by) a plaintiff for some time after the injury occurs.” Rakes v. United States, 442 F.3d 7, 19 (1st Cir.2006). This court has applied the discovery rule in a number of Bulger-related FTCA claims at this point; its applicability here is not in dispute. Under this rule, “a claim accrues under the FTCA once a plaintiff knows, or in the exercise of reasonable diligence should know, (1) of her injury and (2) sufficient facts to permit a reasonable person to believe that there is a causal connection between the government and her injury.” Skwira v. United States, 344 F.3d 64, 78 (1st Cir.2003), cert. denied, 542 U.S. 903, 124 S.Ct. 2836, 159 L.Ed.2d 267 (2004). The inquiry is an objective one — it includes not only what was actually known but what a reasonable person, once fairly prompted to investigate, would have discovered by diligent investigation. Rakes, 442 F.3d at 20, 23. 3 The district court ruled that whatever the plaintiffs may have suspected, they did not have actual or constructive knowledge of their claim prior to September 10, 1999. Litif, 682 F.Supp.2d at 79-80. Specifically, they did not know that Litif was an informant, that he was murdered by Bulger and Flemmi, that this occurred because he was threatening to incriminate them, and that Bulger and Flemmi knew these facts because Connolly so warned them. The court supported this conclusion by pointing to the lack, at that time, of identifiable first-hand witnesses or other admissible evidence to corroborate any of these facts. Id. The plaintiffs could not prove based on newspaper stories quoting anonymous sources that Bulger knew of Litif s offer to cooperate. The FBI was formally and informally denying involvement; Litif was not mentioned in any detail in the Salemme proceedings; and most of the helpful witnesses (e.g., Bulger, Connolly) would hardly, agree to incriminate themselves. The court noted that even today no one has ever been indicted for Litifs murder. Litif, 682 F.Supp.2d at 80. The government counters by saying that proof was unnecessary; prior to the cut-off date there were newspaper accounts suggesting, though falling far short of proving, that Bulger killed Litif; and evidence generated in the Salemme proceedings, reported before the decision, indicating that the FBI was protecting Bulger. This, according to the government, was (1) enough to require the filing of an administrative claim under an emboldening theory of liability, 4 and (2) therefore triggered the accrual period as to all related theories to support a liability claim. Even if we assume the second proposition to be a mechanical rule, but see Donahue v. United States, 634 F.3d 615, 625 *45 (1st Cir.2011) (citing Rakes, 442 F.3d at 21) — -we do not accept the first one. Whether the liability theory is emboldening or leak, prior to the Salemme decision, and arguably for some time after it, there was only press speculation — itself seemingly unknown to the plaintiffs — that Bulger killed Litif. Thus the plaintiffs lacked “sufficient facts to permit a reasonable person to believe that there is a causal connection between the government and [the] injury.” Skwira, 344 F.3d at 78. First, the plaintiffs denied being familiar with the early and relatively sparse newspaper coverage — two major articles and a few other isolated press mentions — connecting Bulger with Litifs death. The modest scale of coverage on this issue contrasts with our cases charging plaintiffs with knowledge regarding facts that have “achieve[d] a local notoriety great enough that the only practicable course is to attribute knowledge of them to people in a position to become familiar with them.” Rakes, 442 F.3d at 20. See also Donahue, 634 F.3d at 625-26; Rakes, 442 F.3d at 22-24; Callahan v. United States, 426 F.3d 444, 448-50, 452-54 (1st Cir.2005). Second, even assuming the articles were known, nothing available prior to September 10, 1999 would have yielded enough for a reasonable belief that Bulger killed Litif. See Rakes, 442 F.3d at 20. Halloran’s hearsay statement was discredited in the same article in which it appeared; Connolly himself refuted the next day the Boston Herald story that Connolly had been advised of Litifs offer to cooperate. There is little else beyond those articles, nor any indication that a diligent inquiry would have yielded anything more prior to September 10, 1999. Much of the evidence at trial — which the government says is still insufficient to establish liability — emerged after September 10, 1999. The September 15, 1999, Salemme decision assembled for the first time a well-documented story of the FBI’s protection of Bulger and Flemmi. It included a description of the evidence suggesting Halloran’s murder by Bulger and the prospect that this information had been leaked by Connolly along with discussion of other murders and players in the saga. The Halloran episode in turn offered helpful indirect evidence, later reinforced by additional examples, of a sequence similar to the one that caused Litifs death. Nor, prior to September 10, 1999, could the plaintiffs know of Bulger and Flemmi’s incriminatory conduct (described below) both before and after Weeks’ wedding; Kevin Weeks did not begin to cooperate with law enforcement until 2000 and his book containing substantially similar information was published in 2006. Kevin Curry’s trial testimony was also essential— both to the leak theory because it showed that Connolly was aware of Litifs offer to inform, and to substantiate the basic fact necessary to any theory supporting recovery that Bulger was the killer. The government also argues that our conclusion is at odds with prior Bulgerrelated cases in which FTCA plaintiffs have been barred under the discovery rule. However, in Donahue, the case most heavily relied upon, there was solid and well-publicized direct proof, more than two years before the Halloran plaintiffs filed their claims, that Morris had told Connolly of Halloran’s cooperation, that Connolly had told Bulger and Flemmi, and that Halloran’s death had followed. Donahue, 634 F.3d at 625-26. 5 *46 So, too, in Rakes. There, the plaintiffs were charged with knowledge, more than two years before their claim was filed, of Connolly’s corrupt relationship with Bulger, Rakes, 442 F.3d at 15-17, 22-24 (and thus the arguments for an emboldening theory); in addition, quite unlike the present case, they knew at the same time without any doubt that Bulger was the direct agent of the harm caused to them by the extortion, because it was Bulger who forced them to sign title of their store over to him. Id. at 12-14. Similarly, in Skwira, the debarred plaintiffs filed no timely claim within two years even after the following had emerged: a highly publicized investigation of suspicious deaths in the ward of a Veteran’s Administration Hospital in which Skwira died, the exhumation of Skwira’s body after a government request to the family citing “suspicions” about the deaths, and information passed to the family after an autopsy was performed confirming that the cause of death listed on the death certificate was inaccurate. Skwira, 344 F.3d at 80. To sum up, the legal “reasonable belief’ standard is not in dispute (save on one issue that we have assumed arguendo in favor of the government). Nor is the pattern of relevant facts as to what was or should have been known by the plaintiffs in this case the same as that in our earlier decisions. Applying the legal test to the facts of this case, our own evaluation, although not identical to that of the district judge, coincides with his conclusion, namely, that the reasonable belief standard did not require an administrative filing earlier than September 10, 2001. Causation. The substantive tort law of Massachusetts has been set forth in the Davis-Hussey decision, see Davis-Hussey at 53-57, and need not be repeated. In its application to the present case, the strongest claim for plaintiffs— both as to negligence and causation (but for and proximate) — was that Connolly leaked Litif s offer to inform to Bulger and that, as a direct and foreseeable consequence, Bulger killed Litif. The government contends that the admissible evidence for this conclusion is too thin. This is surely an uncomfortable position for the government since its own agents took statements from Halloran who told them that he witnessed Litif entering the bar on the night of his death and Bulger or Flemmi helping to carry out a body bag then loaded into a car trunk, where it was in fact found the next day. Nor is there anything implausible about that story. Nevertheless, we agree with the government that Halloran’s statements to Agent Montanari were hearsay under the Federal Rules of Evidence. The district court said the statements could still be considered, reasoning that federal law incorporates state competency rules in FTCA cases and that Massachusetts law provides: In any action or other civil judicial proceeding, a declaration of a deceased person shall not be inadmissible in evidence as hearsay or as private conversation between husband and wife, as the case may be, if the court finds that it was made in good faith and upon the personal knowledge of the declarant. *47 Mass. Gen. Laws ch. 233, § 65 (emphasis added). Litif, 682 F.Supp.2d at 65-66. Reading the underscored language as a rule of competency under Massachusetts law, the district court said that it rendered Halloran’s statement admissible even in the absence of a federally recognized hearsay exception. Id. at 67-68. Assuming arguendo that state competency law did govern FTCA claims, but see 27 Wright & Gold, Federal Practice & Procedure § 6007 (2d ed.2007), the Federal Rules of Evidence treat competency (Rule 601 et seq.) as quite distinct from hearsay (Rule 801 et seq,). So application of Rule 601 — the only invoked basis for importing state competency law — cannot be read to trump federal hearsay rules in a federal proceeding, whether the substantive claim rests on federal or state law. See Wright & Gold, § 4512; accord Fitzgerald v. Expressway Sewerage Constr., Inc., 177 F.3d 71, 74 (1st Cir.1999). Likely aware of such doubts, the district judge did not significantly rely on Hallo-ran’s hearsay testimony to prove that Bulger did kill Litif. That Bulger was responsible for the killing, one way or another, was amply supported by other evidence. Before saying anything about the “manner” of the murder — where the court did briefly mention Halloran’s statement — the district judge had already concluded: The Court infers that Connolly leaked to Bulger Litifs willingness to incriminate Bulger and that the murderous result of this leak was well within the foreseeable risk created.... Several pieces of evidence, taken together, mandate this conclusion: Curry’s testimony (which the Court finds credible); Connolly and Walsh’s relationship ...; Connolly’s leak of Castucci’s cooperation; Connolly’s later leaks of other informants’ status, leading them to a similar fate; ... and Bulger’s statements to Weeks and Flemmi, made shortly after the murder, that Litif was an informant. Litif, 682 F.Supp.2d at 70. In other words, there was a pattern of FBI leaks of informants to Bulger and Flemmi, which was well supported: Flemmi testified that Connolly leaked the names of between six and twelve informants to him. How many were then murdered was unclear but McIntyre, Halloran, and Castucci were apparently leak victims. And Connolly, as Curry testified, was present when Litifs plans to cooperate and incriminate Bulger were made known to the Boston Police, posing a serious danger to Connolly’s continued use of Bulger as an informant and a threat to expose Connolly’s role in covering up crimes. The inference that Connolly then told Bulger and Bulger killed Litif is supported by their respective motives, the timing of Curry’s meeting with Connolly in relation to Litifs death shortly thereafter, and remarks by Bulger to Weeks around the same period that in context can be taken to indicate knowledge by Bulger of Litifs fate. 6 All that was required for plaintiffs to prevail was to show that this was more likely than not. The district judge so found and his conclusion is unsurprising, does not rest solely on pattern, and is not even arguably clear error. Damages. Both parties appeal from the district court’s award of $350,000 for *48 Litif s conscious pain and suffering. The district judge concluded that Litif was stabbed dozens of times before he was shot in the back of the head, but nothing proved that the murder took more than a couple of minutes. Litif, 682 F.Supp.2d at 70-71 & n. 9. The government argues that the plaintiffs have not proved that Litif was conscious while he was stabbed, that is, that he could have been stabbed after being shot. This is possible but the district judge was right to deem it improbable: Bulger, as other evidence indicates (e.g., the strangling of Davis recounted in the companion case), was a cruel man; and stabbing a turncoat while alive and then ensuring his death with a bullet would in any event be more probable than stabs pointlessly inflicted after unconsciousness or death. The district judge certainly could find conscious pain and suffering shortly before Litif s death. The Litif plaintiffs ask that we increase the district court’s award for conscious pain and suffering, arguing that Litif was tortured over an extended period of time, with many of his stab wounds being to the liver (which is thought to be particularly painful). The plaintiffs also contend that Bulger had a motive for prolonged torture, namely, to determine what, if anything, Litif had already said to law enforcement officials. But this is conjecture: the district court found vicious stabbing but no proof that it lasted long, and this can hardly be deemed clear error. Plaintiffs say the award was too low even if Litifs agony was brief, but our discussion in Davis-Hussey forecloses this claim. See Davis-Hussey at 57-59. If published cases that involve stabbings are deemed important, we note two in that category: Miles v. Melrose, 882 F.2d 976, 980-81 (5th Cir.1989), aff'd, 498 U.S. 19, 111 S.Ct. 317, 112 L.Ed.2d 275 (1990) ($140,000 for pain and suffering when victim had been stabbed or cut at least 62 times); Givens v. Rochester City School Dist., 294 A.D.2d 898, 741 N.Y.S.2d 635, 636 (2002) (reducing $1,000,000 verdict to $300,000 for pain and suffering in a stabbing death). As in Davis-Hussey, we cannot conclude that the district court’s award was an abuse of discretion. The judgment of the district court is affirmed. Each side is to bear its own costs on this appeal. It is so ordered. Footnotes [1] The Davis and Hussey cases, Davis v. United States, 670 F.3d 48 (1st Cir.), ("Davis-Hussey"), decided today in a separate decision, were tried and decided together in the district *41 court and briefed and argued together in this court. Other civil cases involving similar claims are cited in Davis-Hussey, n. 1. [2] United States v. Salemme, 91 F.Supp.2d 141 (D.Mass.1999). The proceedings, which were initiated to determine Flemmi's claim of immunity in the criminal case against him as well as evidentiary issues, continued throughout 1998 and unearthed considerable evidence about the relationship of the FBI to its Boston-area informants. Id. at 163. [3] In some cases, the plaintiffs’ subjective knowledge was clear and was therefore held against them for claim accrual purposes. E.g., Skwira 344 F.3d at 80. Here, however, the plaintiffs all disclaim any knowledge — at least prior to the Salemme decision — of the news articles or other information linking Litif's death to Bulger and the FBI. [4] Our prior cases have distinguished between “emboldening” and "leak” theories of liability asserted by the plaintiffs. Roughly, the emboldening theory suggests that the FBI protection of Bulger and Flemmi enabled or emboldened them to commit further crimes for which the government bears responsibility. The leak theory focuses more specifically on Agent Connolly leaking confidential informants’ identities to Bulger and Flemmi. E.g. McIntyre v. United States, 367 F.3d 38, 53-54 (1st Cir.2004). [5] On April 22, 1998, FBI Agent John Morris had given sworn, sensational and well-publicized testimony in the Salemme proceedings that he had told Connolly of Halloran’s coop *46 eration with the FBI, that Connolly told Bulger and Flemmi about this cooperation, and that "he believed that Bulger and Flemmi may have killed Halloran.” Donahue, 634 F.3d at 626. And Flemmi had acknowledged that the FBI had tipped him off about Hallo-ran, which information was made public in a September 1, 1998 judicial order and widely reported the next day. Id. at 620, 626. [6] Weeks testified that shortly before his wedding on April 26, 1980, he talked to Bulger about his concerns regarding where to seat Litif at the wedding reception to prevent conflicts. Bulger told Weeks not to worry about it, that "he probably won't even show." At the reception, Bulger pointed to an empty chair at his table and said "[s]ay hi to Louie.” Flemmi then made wiping motions with his napkin in the area where Litif's head would be and said "he keeps on drinking but it keeps on leaking out of his head.”