Blinzler v. Marriott International, Inc.

SELYA, Circuit Judge. These cross-appeals require us to wend our way through a maze of unusual facts and subtly nuanced legal issues. After exploring a little-charted frontier of tort law, we reverse the district court’s direction of judgment notwithstanding the verdict and reinstate the jury’s award on the plaintiffs claim for negligent infliction of emotional distress. In all other respects, we affirm the rulings of the lower court. I. BACKGROUND This litigation arises out of the tragic demise of James Blinzler, husband of the plaintiff Gloria Blinzler. The course of events leading to James Blinzler’s death began on November 13, 1992, when the Blinzlers checked into a Somerset, New Jersey, hotel operated by the defendant Marriott International, Inc. (Marriott). Shortly after 8:30 p.m. the decedent, relaxing in his room, experienced difficulty in breathing. Sensing danger, he ingested nitroglycerin (he had suffered heart attacks before) while his wife called the hotel PBX operator and requested an ambulance. The operator received the SOS no later than 8:35 p.m. and agreed to honor it. She promptly told the hotel’s security officer and the manager on duty about the medical emergency. Though the defendant steadfastly maintains that the operator also called an ambulance then and there, the record, read hospitably to the verdict, see Cumpiano v. Banco Santander P.R., 902 F.2d 148, 151 (1st Cir.1990), indicates that she did not place this critical call until some fourteen minutes after receiving the plaintiffs entreaty. The ambulance arrived at 9:02 p.m. In the meantime the plaintiff watched her husband’s condition deteriorate: he collapsed on the bed, vomited while supine, and apparently stopped breathing. During this horrific hiatus, the plaintiff troce asked hotel personnel whether an ambulance had been summoned when the emergency first arose. She was twice falsely reassured (whether in honest error is not clear) that one had been called. When the paramedics arrived on the scene, they could not locate a pulse and discovered that the decedent’s airway was blocked. Re- *1151 suscitative efforts restored the decedent’s heart to a normal rhythm and he was transported celeritously to a nearby hospital. Doctors diagnosed the heart attack as a “very small myocardial infarction.” Nevertheless, the brain damage resulting from a prolonged period of asystole without cardiopulmonary resuscitation led to James Blin-zler’s death three days later. II. PROCEEDINGS BELOW, ISSUES ON APPEAL, AND RULES OF DECISION Invoking diversity jurisdiction, 28 U.S.C. § 1332 (1994), the plaintiff sued Marriott in Rhode Island’s federal district court for wrongful death (count 1), loss of consortium (count 2), and negligent infliction of emotional distress (count 3). She alleged in substance that the hotel failed to summon an ambulance in a timely fashion and that this carelessness proximately caused both her own damages and her husband’s death. The jury agreed, awarding $200,000 for wrongful death, $50,000 for loss of consortium, and $200,000 for emotional distress. Addressing a variety of post-trial motions, the district judge upheld the verdict, on the first two counts, but granted judgment for the defendant on the third count. Both sides appeal. The cross-appeals raise several issues. Two are in the forefront. The centerpiece of the defendant’s appeal is the assertion that the evidence did not forge a causal link between the failure promptly to summon an ambulance and the ensuing death. In contrast, the plaintiffs appeal hinges on the district court’s extirpation of the jury verdict on her claim for negligent infliction of emotional distress. Because the defendant’s contention that the plaintiff failed as a matter of law to prove causation involves an across-the-board challenge to the jury verdict as a whole, we deal first with that issue. We then mull the plaintiffs contention that the lower court erroneously forecast emergent New Jersey law on bystander liability and therefore erred in setting aside the verdict on count 3. Finally, we address the defendant’s remaining assignments of error. Under the principles of Erie R.R. Co. v. Tompkins, 304 U.S. 64, 78, 58 S.Ct. 817, 822, 82 L.Ed. 1188 (1938), state law (here, the law of New Jersey) supplies the substantive rules of decision in this diversity case. Since New Jersey law is less than explicit on one key issue that concerns us, we pause to comment briefly on the role of a federal court in adjudicating controversies controlled by state law. In its barest essence, borrowing state law demands nothing more than interpreting and applying the rules of substantive law enunciated by the state’s highest judicial authority, or, on questions to which that tribunal has not responded, making an informed prophecy of what the court would do in the same situation. 1 See Moores v. Greenberg, 834 F.2d 1105, 1112 (1st Cir.1987). In the latter instance, we seek guidance in analogous state court decisions, persuasive adjudications by courts of sister states, learned treatises, and public policy considerations identified in state decisional law. See Ryan v. Royal Ins. Co., 916 F.2d 731, 734-35 (1st Cir.1990); Kathios v. General Motors Corp., 862 F.2d 944, 949 (1st Cir.1988). As long as these signposts are legible, our task is to ascertain the rule the state court would most likely follow under the circumstances, even if our independent judgment on the question might differ. See Moores, 834 F.2d at 1107 n. 3. III. CAUSATION The defendant challenges the entire verdict on the basis that the plaintiff provided insufficient evidence from which a reasonable jury could conclude that the belated call constituted a proximate cause of the ensuing death. Under New Jersey law the plaintiff bears the burden of proving that the defendant’s conduct comprised “a substantial factor in producing the harm” of which the plaintiff complains. Francis v. United Jersey Bank, 87 N.J. 15, 432 A.2d 814, 829 (1981). When the questioned con- *1152 duet is an omission — the defendant’s failure to act rather than the defendant’s maladroit performance of an affirmative act — this rule is easier to state than to apply. In the last analysis, it can rarely (if ever) be said with absolute certainty that harm would not have befallen the victim if the omitted action had been taken. . One species of omission that occurs from time to time involves the generic charge that, had the defendant done some particular act, the plaintiff (or, as here, the plaintiff’s decedent) would have had a better chance to ward off threatened harm. In these so-called “loss of chance” cases New Jersey law instructs that the plaintiff can carry her burden by showing a “substantial possibility” that the harm would have been averted had the defendant acted in a non-negligent manner. Hake v. Manchester Township, 98 N.J. 302, 486 A.2d 836, 839 (1985); see also Olah v. Slobodian, 119 N.J. 119, 574 A.2d 411, 417-19 (1990) (discussing Hake) 2 Transposed to the rescue context, this rule renders a defendant’s omission actionable if the plaintiff can show that the omission “negated a substantial possibility that prompt rescue efforts would have been successful.” Hake, 486 A.2d at 839. Under these authorities, the question here reduces to whether the evidence, viewed in the light most congenial to the plaintiff, supports a finding that the defendant’s failure promptly to call an ambulance negated a substantial possibility that James Blinzler would have survived. We think that this question warrants an affirmative answer. The plaintiff submitted evidence that she beseeched the defendant to summon help at 8:35 p.m.; that an ambulance crew was available and free to respond at that time; and that the defendant agreed to place the call but then neglected to do so. The defendant actually made the call at 8:49 p.m. (some fourteen minutes later) and the ambulance reached the scene at 9:02 p.m. (an elapsed time of thirteen minutes). The juxy heard opinion evidence from a renowned cardiologist that serious brain damage (and, hence, death) would have been forestalled had the paramedics reached the premises ten minutes earlier. On this record, we believe that a reasonable jury could conclude that the defendant’s omission negated a substantial possibility that the rescue efforts would have succeeded. Put another way, a reasonable jury could find (as this jury apparently did) that the ambulance likely would have arrived fourteen minutes earlier had it been summoned immediately; that the course of treatment would have been accelerated by a like amount of time; and that, but for Marriott’s negligence James Blinzler would have survived. The defendant tries to parry this thrust in two ways. One initiative involves assembling a string of cases (mostly of hoary origin) in which courts have rejected plaintiffs’ claims of negligence for failure to rescue. See, e.g., Foss v. Pacific Tel. & Tel. Co., 26 Wash.2d 92, 173 P.2d 144, 149 (1946); Whitehead v. Carolina Tel. & Tel. Co., 190 N.C. 197, 129 S.E. 602, 605 (1925); Volquardsen v. Iowa Tel. Co., 148 Iowa 77, 126 N.W. 928, 930 (1910); Lebanon, L. & L. Tel. Co. v. Lanham Lumber Co., 131 Ky. 718, 115 S.W. 824, 826 (1909). These cases — all of which involve fire damage coupled with some alleged negligence on the part of a telephone company in respect to a telephone - call meant to summon the fire department — provide little guidance. In those cases, unlike here, the plaintiffs did not proffer evidence that, had the call gone through, the rescuers (there, the firefighters) could have reached the scene in time to prevent the harm (there, the rapid spread of a conflagration that had already started). See, e.g., Foss, 173 P.2d at 149; Lebanon, 115 S.W. at 826. And perhaps more importantly, each of those cases draw on Lebanon for the legal standard of causation — a standard that differs materially from New Jersey’s standard. See Lebanon, 115 S.W. at 826 (stating that “it must be established with certainty that but for their negligence the fire would *1153 have been confined” as the plaintiff contends) (emphasis supplied). This second point is aptly illustrated by the one entry in Marriott’s string citation that does not involve a burning building: Hardy v. Southwestern Bell Tel. Co., 910 P.2d 1024 (Okla.1996). To understand Hardy, it is necessary to note that, in McKellips v. St. Francis Hasp., Inc., 741 P.2d 467, 475-77 (Okla.1987), the Oklahoma Supreme Court held that the causation standard of the Restatement (Second) of Torts § 328 (under which a plaintiff may prove negligence in a loss of chance case by showing that the defendant’s omission “increase[d] the risk” of harm), applied in medical malpractice cases. Hardy — a case in which the plaintiff alleged that the telephone company’s negligent operation of a 911 service prevented him from summoning rescue assistance and thereby proximately caused his wife’s death — postdated McKellips. The Oklahoma Supreme Court there considered extending the causation standard of Restatement § 323 to loss of. chance claims outside the medical malpractice context. See Hardy, 910 P.2d at 1025. It declined to do so. See id. at 1030. Hardy, fairly read, confirms the distinction between proof of causation in loss of chance cases under the traditional test (to which Oklahoma adheres in cases not involving medical malpractice) and under more modem standards that focus instead on whether a defendant’s conduct has significantly increased particular risks. As we have explained, New Jersey’s “substantial possibility” standard applies to loss of chance cases in general, 3 and it is at a minimum as liberal as the “increased risk” standard endorsed by section 323 of the Restatement. See Olah, 574 A.2d at 419 (suggesting that whether the plaintiff “has a substantial possibility of avoiding harm would ordinarily be subsumed in the jury’s determination whether a defendant’s deviation increased the risk of harm”) (internal quotations omitted). Since Hardy apparently would have stated a claim had the Oklahoma court applied the more lenient “increased risk” standard, see Hardy, 910 P.2d at 1030, Marriott’s flagship case actually supports a finding of causation under New Jersey law. Marriott’s second attempt to scuttle the finding of causation features its lament that, the plaintiff did not prove that the same traffic conditions which were extant at and after 8:49 p.m. were also extant at and after 8:35 p.m. This lament can scarcely be taken seriously. Juries have the power to draw reasonable inferences from established facts. It is well within a jury’s ordinary competence to conclude that traffic conditions for an emergency vehicle do not change dramatically in a fourteen minute period that is well outside rush hour. The defendant’s suggestion that a highway accident, or a diluvian tempest, or some other freak occurrence, later abated, might have delayed the ambulance if it began its run at 8:35 p.m. rather than at 8:49 p.m. is equally jejune. It is the plaintiffs burden to prove her ease by a preponderance of the evidence, not beyond all conceivable doubt. In the absence of some reason to suspect changed conditions — and there is no evidence of any actual change here — the jury’s inference that the ambulance would have arrived in roughly the same elapsed portal-to-portal time is unimpugnable. See Levesque v. Anchor Motor Freight, Inc., 832 F.2d 702, 704 (1st Cir.1987) (explaining that the “perhapses” that dot a factbound trial record typically “are for fact-finders to resolve — not for judges imperiously to dictate”); see also W. Page Keeton et al., Prosser & Keeton on Torts § 41, at 269 (5th ed.1984) (noting that a plaintiff does not have to negate entirely the possibility that the defendant’s conduct was not a contributing cause of the harm). Silhouetted against this legal backdrop, the evidence of record, visualized most favorably to the plaintiff, see Cumpiano, 902 F.2d at 151, suffices to ground a finding that, had the defendant hailed an ambulance immediately upon request, there was at least a significant possibility that James Blinzler’s death could have been prevented. Accordingly, we are not at liberty under New Jersey law to dis *1154 turb the jury’s conclusion that Marriott’s negligence constituted a substantial factor in the ensuing death. IV. BYSTANDER LIABILITY The most vexing issue in this case involves the plaintiffs claim of negligent infliction of emotional distress. This claim is based on the injury that she experienced while watching her husband suffer as the beleaguered couple awaited the ambulance’s overdue arrival. We start this segment of our analysis with a discussion of the doctrine of bystander liability as it has evolved in New Jersey, then shift our attention to an open question that the district court found to be dispositive, and, finally, apply the doctrine as we understand it to the idiosyncratic facts of this case. A. General Principles of Bystander Liability. American courts first recognized bystander liability in the landmark case of Dillon v. Legg, 68 Cal.2d 728, 69 Cal.Rptr. 72, 441 P.2d 912 (1968). Drawing in part on precedents from English common law, the court ruled that a mother could recover for emotional and physical injuries suffered “from witnessing the [negligent] infliction of death or injury to her child.” Id. at 914. The Dillon court implicitly suggested that any bystander should be able to recover for all objectively foreseeable injuries. See id. at 920-21. To help jurists navigate the reefs and shoals of foreseeability, the court attempted to elucidate guidelines based on Dillon’s factual scenario. See id. at 920. Twelve years later, New Jersey embraced bystander liability in Portee v. Jaffee, 84 N.J. 88, 417 A.2d 521 (1980). The state supreme court did not clasp Dillon uncritically to its bosom, but, rather, abjured a tunnel-vision focus on foreseeability, fearing that it would open the door to claims of emotional distress advanced on behalf of any onlooker to any negligently caused event. 4 See id. at 527 (cautioning against institutionalizing “an unreasonably excessive measure of liability”); see also Carey v. Lovett, 132 N.J. 44, 622 A.2d 1279, 1286 (1993) (suggesting that treating foreseeability as a sole talisman would render it difficult to differentiate between legitimate and fraudulent claims); Prosser & Keeton, supra, § 54, at 366 (warning that forcing defendants to pay for the “lacerated feelings” of every bystander would be “an entirely unreasonable burden on human activity”). In an effort to furnish a condign remedy for deserving injuries while at the same time avoiding “speculative results or punitive liability,” Portee, 417 A.2d at 526, New Jersey transmogrified the Dillon guidelines into prerequisites of a cause of action for bystander liability, see id. at 528. The Portee court concluded that a bystander plaintiff should be permitted to recover under New Jersey law only if she could prove (1) the death or serious injury of another (caused by the defendant’s negligence); (2) an intimate familial relationship with the victim; (3) her observation of the victim at the time of the injury or immediately thereafter; and (4) severe emotional distress resulting from the observation. See id. Subsequent decisions have cut plaintiffs some slack (but not very much) in their efforts to fulfill this quadrat of requirements. See, e.g., Dunphy v. Gregor, 136 N.J. 99, 642 A.2d 372, 377-78 (1994) (holding that unmarried cohabitants may enjoy an intimate familial relationship); Frame v. Kothari, 115 N.J. 638, 560 A.2d 675, 678 (1989) (explaining that a plaintiff may recover without actually seeing the injury so long as it is “susceptible to immediate sensory perception” and the plaintiff observes the victim immediately after the injury is inflicted). These four elements serve a critical function in keeping bystander liability within reasonable bounds. First, they furnish a set of guideposts that help to identify and define a range of claims that are presumptively valid while excluding other claims that society simply cannot afford to honor. See Dunphy, 642 *1155 A.2d at 377 (noting that the elements of bystander liability “structure the kind of ‘particularized foreseeability 5 that ensures that the class is winnowed ... and that limitless liability is avoided”). Second — and relatedly — they combine to define narrowly the emotional interest that the law protects. See Carey, 622 A.2d at 1286; accord Thing v. La Chusa, 48 Cal.3d 644, 257 Cal.Rptr. 865, 880, 771 P.2d 814, 829 (1989). While “[t]he law should find more than pity for one who is stricken by seeing that a loved one has been critically injured or killed,” Portee, 417 A.2d at 526, the elements of the bystander liability tort frankly recognize that it is not the law’s province to shield people from all anxieties. Since the ordinary slings and arrows of human existence inevitably produce stress and strain, “only the most profound emotional interests should receive vindication for their negligent injury.” Id. The common thread that runs through these cases is that emotional anguish is a natural, perhaps omnipresent, reaction whenever one is forced to watch a loved one suffer, and therefore should not be compen-sable in the absence of special circumstances. In an effort to hold the line, New Jersey law decrees that bystanders may recover in tort only' for the particularly exquisite anguish that occurs when they personally observe trauma strike a loved one like a bolt from the blue. See Frame, 560 A.2d at 679 (explaining that bystander liability is supposed to remedy the “harm of seeing a healthy victim one moment and a severely injured one the next”); Portee, 417 A.2d at 527 (“Discovering the death or serious injury of an intimate family member will always be expected to threaten one’s emotional welfare. Ordinarily, however, only a witness at the scene of the accident causing death or serious injury will suffer a traumatic sense of loss that may destroy his sense of security and cause severe emotional distress.”). Thus, there can be no recovery unless the close relation’s helpless watching arises in the context of a sudden, unexpected, and accidental injury. B. The Fork in the Road. The issue before us is whether the plaintiff’s asserted injury falls within the narrow range of bystander liability claims that are actionable under New Jersey law. The district court decided that it did not. The court relied primarily on a series of bystander liability/medical malpractice cases in which the New Jersey Supreme Court placed a gloss on its earlier decisions and indicated that a plaintiff must witness the actual act of malpractice and appreciate its effect on the patient in order to bring herself within the class of persons entitled to recover. See Carey, 622 A.2d at 1288 (declaring that a plaintiff must “contemporaneously observe the malpractice and its effects on the victim”); Frame, 560 A,2d at 681 (“In an appropriate case, if a family member witnesses the physician’s malpractice, observes the effect of the malpractice on the patient, and immediately connects the malpractice with the injury, that may be sufficient to allow recovery for the family member’s emotional distress.”); see also Gendek v. Poblete, 139 N.J. 291, 654 A.2d 970, 975 (1995) (rejecting a claim on the ground that the bystander did not “immediately conneet[] any act of malpractice” with the victim’s death). Here, concededly, Mrs. Blinzler cannot satisfy this added requirement; she neither “witnessed” the negligence (which comprised the hotel operator’s failure immediately to call an ambulance and which occurred six floors beneath the Blinzlers’ room) nor fully appreciated at the time that the negligence was hindering needed assistance (indeed, the defendant’s misrepresentations, if believed, concealed the very fact of the negligence). Thus, to decide this case we must determine whether the Gendek-Carey-Frame gloss applies only to bystander liability/medical malpractice claims (as the plaintiff contends) or whether it applies to all bystander liability claims (as the defendant contends and as the lower court concluded). Although the answer to the question is by no means free from doubt, we think that the district court took the wrong fork in the road. As an initial matter, the New Jersey Supreme Court has never imposed the added requirement that a plaintiff witness the negligent act and contemporaneously connect it to the injury of a loved one in any case outside the medical malpractice context, and the mal *1156 practice cases in which the requirement has been imposed strongly suggest that it is restricted to that milieu. See Gendek, 654 A.2d at 974 (describing the requirement as “added” and “special” in that it is “imposed to establish an indirect claim for emotional distress arising from medical malpractice”); Carey, 622 A.2d at 1286 (“With medical-malpractice claims, we have required that claimants observe contemporaneously the act of malpractice and the resultant injury”) (emphasis supplied). What is more, crafting a special set of rules for bystander liability/medical malpractice cases is not in any way an unprecedented flight of fancy; other courts that recognize bystander liability claims in general sometimes treat such claims more restrictively in the medical malpractice setting, even, on occasion, barring them outright. See, e.g., Maloney v. Conroy, 208 Conn. 392, 545 A.2d 1059, 1063-64 (1988); Wilson v. Galt, 100 N.M. 227, 668 P.2d 1104, 1110 (1983). We note, too, that the added requirement designed by the New Jersey Supreme Court for use in connection with bystander liability/medical malpractice claims is grounded in a set of policy considerations that do not seem to apply to other bystander liability claims. For one thing, the unique emotional interest that fuels the doctrine of bystander liability is unaffected in most cases of medical malpractice for the harm caused by, say, misdiagnosis usually does not manifest itself until days, weeks, months, or years have elapsed, and even then, the misdiagnosis rarely culminates in a single spontaneous and shocking event. See Frame, 560 A.2d at 678-79 (explaining that in the typical malpractice ease “[g]rief over the gradual deterioration of a loved one, as profound as that grief may be, often does not arise from a sudden injury,” but, rather, under circumstances in which the family members have had the “time to make an emotional adjustment”). It is largely for this reason that bystander liability must be even more “narrowly circumscribed in the context of a medical misdiagnosis or failure to act.” Gendek, 654 A.2d at 975-76. New Jersey chose to accomplish this circumscription by limiting bystander liability in the medical malpractice arena to those situations in which the putative plaintiff observes both the act of malpractice and its immediate effects, and appreciates the interrelationship. See, e.g., Frame, 560 A.2d at 681. That rationale loses force outside the medical malpractice context. For another thing, the added requirement applicable to bystander liability in the medical malpractice context reflects societal concerns about the impact of expanded liability on the delivery of health care. See Gendek, 654 A.2d at 975 (“In considering the standards that govern an appropriate duty of care and limitations of liability in [the health care] setting, we must be especially mindful of the principles of sound public policy that are informed by perceptions of fairness and balance.”); Carey, 622 A.2d at 1286 (voicing uneasiness about the “effects of the expansion of liability on the medical profession and society,” and specifically noting sharp increases in malpractice insurance premiums); Frame, 560 A.2d at 681 (emphasizing that the special refinement of bystander liability for medical malpractice cases is at least partly driven by a desire to avoid “unreasonably burdening the practice of medicine”). This group of situation-specific policy concerns is best addressed by “narrowly circumscribing]” bystander liability in the medical malpractice setting so as to minimize any “adverse effect on the practice of medicine or on the availability of medical services.” Frame, 560 A.2d at 681. Once again, this reasoning loses force outside the medical malpractice context. The language of the New Jersey cases and the distinctive nature of the policy considerations that underlie the added requirement mark the genesis of our belief that, when the opportunity arises, the New Jersey Supreme Court will not engraft this health-care-specific requirement upon the body of cases that lie beyond the medical malpractice arena. New Jersey has already expressed its view of general public policy concerns with respect to expanded liability for run-of-the-mine accidents by conferring a right of recovery on bystanders and defining the elements of the tort. See Dunphy, 642 A.2d at 377; Portee, 417 A.2d at 528. We think it is no accident that in superimposing the added requirement *1157 on bystander liability/medieal malpractice cases, the state supreme court has been scrupulously careful not to imply a broader sweep. Because we believe that this specificity is purposeful rather than serendipitous, we hold that the added requirement imposed by the Gendek-Carey-Frame line of cases is applicable only to causes of action that, at bottom, charge health-care providers with malpractice. The district court, therefore, took the wrong fork in the road. C. Applying the Principles. Once we put the added requirement to one side, the only question that remains open under this rubric is whether the jury lacked evidence satisfactory to support a finding that the plaintiffs injury fell within the standard parameters of bystander liability that obtain in New Jersey vis-a-vis suits arising outside the medical malpractice context. We think the evidence sufficed. Intimate relationship and third-party injury (i.e., a spouse’s death) are not in dispute, and the record contains adequate proof of severe emotional distress. The seminal New Jersey case suggests that, in addition to these three elements, a plaintiff need only show that she “observ[ed] the death ... while it oc-eur[red].” Portee, 417 A.2d at 527; see also supra p. 1154 (recounting the four elements of the tort under New Jersey law). This last element — firsthand observation — corresponds to the distinct emotional interest that is infringed when an individual witnesses a “shocking event” and “see[s] a healthy [family member] one moment and a severely injured one the next.” Frame, 560 A.2d at 679. We appreciate that things are not always what they seem and that it may be overly simplistic to say that in New Jersey firsthand observation of a suddenly inflicted injury to a loved one invariably gives rise to the unique emotional interest that underlies bystander liability. Arguably, it is not merely the observation of the injury but the perception that it is accidental or otherwise unwarranted that threatens a “plaintiffs basic emotional security,” Portee, 417 A.2d at 521, and thus paves the way for bystander liability. See id. at 528 (noting that it is the “shock and fright” attendant to observing the “accidental death” of an intimate relation that infringes the narrowly defined interest in emotional security). Frame makes this point most clearly, albeit in dictum: Everyone is subject to injury, disease, and death. Common experience teaches that the injury or death of one member of a family often produces severe emotional distress in another family member. A threshold problem is separating the grief that attends that distress when no one is at fault from the added stress attributable to the fact that the injury or death was produced by the negligent act of another. Id. at 677. And while the Portee elements have not yet been formally modified in this respect, 5 we think it is not unlikely that New Jersey will move in this direction. Cf. Thing, 771 P.2d at 829 (tightening the elements of a bystander liability action under California law to require that the plaintiff be “present at the scene of the injury-causing event” and be “then aware that it is causing the injury to the victim”). But there are two reasons why we need not cross this bridge today. 1. The evidence here clearly satisfies the Portee requirements simpliciter. The plaintiff witnessed a sudden and shocking event when she watched her husband of forty-two years undergo excruciating chest pain, vomit, struggle to catch his breath, asphyxiate, lose consciousness, and ultimately die. Because she “witness[ed] the victim when the injury [was] inflicted,” Frame, 560 A.2d at 678, recovery would seem appropriate under a formal incantation of the Portee elements. 2. The law of the case doctrine eliminates any potential problem as to the precise dimensions of Portee. At trial’s end, the district court charged the jury that “the plaintiff must be present at the scene of the event *1158 and be aware that the victim is being injured.” The defendant’s counsel objected generally to the court’s decision to instruct the jury at all on count 3 (asseverating that New Jersey law requires the plaintiff actually to witness the negligent act) but he did not object in any other, more specific respect to the district court’s formulation of the basic elements of the tort. Thus, even if New Jersey might in an appropriate ease impose some intermediate limitation going beyond Portee but stopping short of mandating that the plaintiff witness the negligent act, the defense formulated no such intermediate position at the jury-instruction stage. In other words, the content of the instruction stands as the law of the case with respect to the unembellished contours of a cause of action for bystander liability. See Quinones-Pacheco v. American Airlines, Inc., 979 F.2d 1, 4 n. 3 (1st Cir.1992); Milone v. Moceri Family, Inc., 847 F.2d 35, 38-39 (1st Cir.1988). Auid as we have already pointed out, the plaintiffs proof, measured against the language of the trial court’s instruction, suffices to create a jury question. Even if we assume arguendo that the New Jersey Supreme Court would augment the elements of a non-medieal-malpractiee cause of action for bystander liability along the lines exemplified by Thing, the verdict might well be sustainable. From the evidence adduced at trial, the jury rationally could find that during the incident proper the plaintiff twice asked the manager whether the ambulance had been called. Though she was (erroneously) assured that the call had been made punctually, she asked the manager yet again at the hospital (receiving the same misinformation), and then checked with the hotel three days later (after her husband had perished). This type of evidence arguably could support an illation that the plaintiff suspected all along that a delay attributable to the defendant was causing injury to her husband. Watching the event while suspecting that her husband’s suffering was being unnecessarily prolonged and worrying that prospects for his rescue were diminishing would appear to be the kind of distinct emotional harm for which bystander liability would lie under the premise of Thing. See, e.g., Bloom v. DuBois Regional Med. Ctr., 409 Pa.Super. 83, 597 A.2d 671, 683 (1991). V. OTHER ISSUES The defendant raises a salmagundi of other issues in connection with its appeal. None of its asseverations is persuasive. Only three warrant discussion. A. The Evidentiary Rulings. The defendant argues that it is entitled to a new trial because the district court erred in certain evidentiary rulings. Its chief complaint concerns the admission of evidence relating to the destruction of the so-called Xeta report (a printout that catalogues all outgoing calls from the hotel’s PBX operator) for November 13, 1992. The defendant destroyed this telephone log approximately thirty days after the incident. Had the report been preserved, it would have pinpointed the very moment that the operator first placed the call for emergency assistance. During the trial, the plaintiff sought to show that the defendant had destroyed this evidence. The defendant objected, contending that it discarded the Xeta report in the ordinary course of business, pursuant to established practice, and not as part of an effort to inter unfavorable evidence. The district court overruled the objection and permitted the plaintiff to introduce evidence at trial of the existence and subsequent destruction of the Xeta report, leaving the defendant’s explanation to the jury. We review the district court’s rulings admitting or excluding evidence for abuse of discretion. See Veranda Beach Club Ltd. Partnership v. Western Sur. Co., 936 F.2d 1364, 1373 (1st Cir.1991); United States v. Nazzaro, 889 F.2d 1158, 1168 (1st Cir.1989). We see none in this instance. When a document relevant to an issue in a case is destroyed, the trier of fact sometimes may infer that the party who obliterated it did so out of a realization that the contents were unfavorable. See Nationwide Check Corp. v. Forest Hills Distributors, Inc., 692 F.2d 214, 217 (1st Cir.1982); see also 2 Wigmore on Evidence § 285, at 192 (James H. Chadbourn rev. ed. 1979). *1159 Before such an inference may be drawn, there must be a sufficient foundational showing that the party who destroyed the document had notice both of the potential claim and of the document’s potential relevance. See Nation-Wide, 692 F.2d at 218. Even then, the adverse inference is permissive, not mandatory. If, for example, the factfinder believes that the documents were destroyed accidentally or for an innocent reason, then the factfinder is free to reject the inference. See, e.g., Jackson v. Harvard Univ., 900 F.2d 464, 469 (1st Cir.), cert. denied, 498 U.S. 848, 111 S.Ct. 137, 112 L.Ed.2d 104 (1990); Anderson v. Cryovac, Inc., 862 F.2d 910, 925-26 (1st Cir.1988). In this case, the defendant contends that there was no direct evidence to show that it discarded the Xeta report for any ulterior reason. This is true as far as it goes — but it does not go very far. The proponent of a “missing document” inference need not offer direct evidence of a coverup to set the stage for the adverse inference. Circumstantial evidence will suffice. See, e.g., Brown & Williamson Tobacco Corp. v. Jacobson, 827 F.2d 1119, 1134 (7th Cir.1987), cert. denied, 485 U.S. 993, 108 S.Ct. 1302, 99 L.Ed.2d 512 (1988). We do not believe that the district court abused its considerable discretion in deciding that the totality of the circumstances here rendered such an inference plausible. A reasonable factfinder could easily conclude that Marriott was on notice all along that the Xeta report for November 13, 1992 was relevant to likely litigation. Although no suit had yet been begun when the defendant destroyed the document, it knew of both James Blinzler’s death and the plaintiffs persistent attempts — including at least one attempt after Blinzler died — to discover when the call for emergency aid had been placed. This knowledge gave the defendant ample reason to preserve the report in anticipation of a legal action. When the evidence indicates that a party is aware of circumstances that are likely to give rise to future litigation and yet destroys potentially relevant records without particularized inquiry, a factfinder may reasonably infer that the party probably did so because the records would harm its case. See Vodusek v. Bayliner Marine Corp., 71 F.3d 148, 156 (4th Cir.1995); Partington v. Broyhill Furn. Indus., Inc., 999 F.2d 269, 272 (7th Cir.1993); Nation-Wide, 692 F.2d at 219. In the circumstances at bar, the trial court acted within its discretion in admitting the Xeta report. The, defendant also chastises the court for admitting evidence of another missing record. The security officer’s log for November 13, 1992 could not be located, and the judge permitted evidence of that fact to go to the jury. Once again, the ruling camiot be faulted. The defendant had no good explanation for the missing log, and the jury was entitled to infer that the defendant destroyed it in bad faith. To cinch matters, these two pieces of evidence had a synergistic effect. We think it would be proper for a reasonable factfinder to conclude that the unavailability of two important documents, both of which bore upon the timing of the call for emergency assistance, was something more than a coincidence. The veteran district judge, after hearing all the evidence limning these mysterious disappearances, put it bluntly in the course of ruling on post-trial motions: I will tell you now that the Xeta Report raises a compelling inference in my mind that personnel at the Marriott Hotel did destroy that record willfully, along with the security officer’s daily log of that date. The inference is compelling that the Marriott Hotel was hiding the delay of the telephone operator in making this telephone call. This is a harsh assessment — but it is based on a firsthand appraisal of the testimony and it is one that a rational jury easily could draw on the record. B. The Motion to Reopen. After the plaintiff rested, the defendant moved for a directed verdict under Fed.R.Civ.P. 50(a). After hearing arguments, the district court permitted the plaintiff to reopen her ease in order to offer certain addi *1160 tional evidence on the issue of causation. 6 The defendant assigns error to this ruling. There is none. The Federal Rules of Evidence give the district court broad discretion in ordering the proof. See Fed.R.Evid. 611. This discretion extends to granting or denying motions to reopen, see Zenith Radio Corp. v. Hazeltine Research, Inc., 401 U.S. 321, 331-32, 91 S.Ct. 795, 802-03, 28 L.Ed.2d 77 (1971); Riveras-Flores v. Puerto Rico Tel. Co., 64 F.3d 742, 746 (1st Cir.1995); Lussier v. Runyon, 50 F.3d 1103, 1113 (1st Cir.), cert. denied, — U.S. -, 116 S.Ct. 69, 133 L.Ed.2d 30 (1995), and such rulings are reviewed principally for abuse of that discretion. A trial court’s decision to reopen is premised upon criteria that are flexible and fact-specific, but fairness is the key criterion. See Rivera-Flores, 64 F.3d at 746; Capital Marine Supply, Inc. v. Thomas, 719 F.2d 104, 107 (5th Cir.1983). The specific factors to be assessed include the probative value of the evidence sought to be introduced, the proponent’s explanation for failing to offer the evidence earlier, and the likelihood of undue prejudice. See Rivera-Flores, 64 F.3d at 746; Joseph v. Terminix Int’l Co., 17 F.3d 1282, 1285 (10th Cir.1994); see also 6A James W. Moore, Moore’s Federal Practice ¶ 59.04[13], at 59-33 (2d ed.1993). The prospect of prolonging the trial is also material. If the additional evidence is immediately available or nearly so, the trial court will have a greater incentive to permit the case to be reopened. Conversely, if gathering the additional evidence portends a significant delay in the trial, the court ordinarily will have a greater reluctance to grant the motion. See Moore, supra, ¶ 59.04[13], at 59-33. Here, the additional evidence that the plaintiff sought to introduce was noncumulative. It had significant probative value on an essential element in the plaintiffs case, helping to connect the defendant’s negligence to the injuries claimed. See supra note 6. There is no sign that the plaintiff withheld the proof as a strategic matter. To the contrary, the record shows quite clearly that she attempted to streamline her case in chief and offered the incremental evidence only after the judge expressed reservations about the state of the proof on the issue of causation. 7 Notwithstanding these circumstances, the defendant insists that permitting the plaintiff to reopen worked substantial prejudice because the defense hoped all along that the plaintiff would fail to prove causation. This is disappointment rather than cognizable prejudice. The evidence taken after reopening consisted of only two witnesses and created no unfair surprise. The added testimony simply fleshed out the plaintiffs basic theory of liability — that the time saved by a prompt call might well have led to James Blinzler’s survival. Moreover, allowing the plaintiff to reopen did not perceptibly delay the trial and did not occasion any interruption of the defense case. In any event, the district court prudently offered the defendant a continuance so that it might regroup and better rebut the additional evidence. By declining the court’s offer, the defendant confirmed the absence of unfair prejudice. See United States v. Diaz-Villafane, 874 F.2d 43, 47 (1st Cir.), cert. denied, 493 U.S. 862, 110 S.Ct. 177, 107 L.Ed.2d 133 (1989). Under these circumstances, the granting of the plaintiffs motion to reopen comes well within *1161 the heartland of the trial court’s discretion. See Rivera-Flores, 64 F.3d at 749. C. The Emotional Distress Award. Where, as here, a federal court sets aside a jury’s verdict and directs the entry of judgment as a matter of law, the court must also rule conditionally on any concomitant motion for a new trial. See Fed.R.Civ.P. 50(c). In this instance the district court held that, if it had erred in granting judgment as a matter of law on count 3, then the jury’s award of damages for emotional distress should stand. The defendant assails this contingent ruling and argues for either a new trial or a remitti-tur on count 3. In its most cogent aspect, the argument is based on the premise that the scanty physical symptoms exhibited by the plaintiff simply do not justify an award of $200,000 in damages. Federal law governs the question of whether the trial court should order a remittitur in a diversity case. See Donovan v. Penn Shipping Co., 429 U.S. 648, 649, 97 S.Ct. 835, 836-37, 51 L.Ed.2d 112 (1977). Under applicable federal standards, appellate review is limited to whether the district court abused its discretion in deciding to endorse the jury award rather than trim it or set it aside as excessive. See, e.g., Ruiz v. Gonzalez Caraballo, 929 F.2d 31, 34 (1st Cir.1991); Wagenmann v. Adams, 829 F.2d 196, 215 (1st Cir.1987). An award of damages will not be deemed unreasonably high or low as long as it comports with some “rational appraisal or estimate of the damages that could be based on the evidence before the jury.” Milone, 847 F.2d at 37 (citation omitted). On the high side, a damage determination will withstand scrutiny unless it is “grossly excessive, inordinate, shocking to the conscience of the court, or so high that it would be a denial of justice to permit it to stand.” Correa v. Hospital San Francisco, 69 F.3d 1184, 1197 (1st Cir.1995) (quoting Grunenthal v. Long Island R.R. Co., 393 U.S. 156, 159 & n. 4, 89 S.Ct. 331, 333 & n. 4, 21 L.Ed.2d 309 (1968)), cert. denied, — U.S.-, 116 S.Ct. 1423, 134 L.Ed.2d 547 (1996). Moreover, “an appellate court’s normal disinclination to second-guess a jury’s evaluation of the proper amount of damages is magnified where ... the damages entail a monetary valuation of intangible losses, and the trial judge, having seen and heard the witnesses at first hand, accepts the jury’s appraisal.” Id. Here, viewing the evidence of damages in the light most amiable to the plaintiff, see Toucet v. Maritime Overseas Corp., 991 F.2d 5, 11 (1st Cir.1993); Ruiz, 929 F.2d at 34, we think that the award, though perhaps generous, passes muster. Under New Jersey law, no particular level of physical symptomatology is necessary to support damages for emotional distress. See Strachan v. John F. Kennedy Mem. Hosp., 109 N.J. 523, 538 A2d 346, 353 (1988). 8 The testimony in this record indicates that the plaintiff watched helplessly as her husband collapsed, vomited, passed out, and became cyanotic. She was still in the room nearly fifteen minutes later when an oxygeii mask was being placed over her unconscious husband’s mouth and nose. In the aftermath of her husband’s death, she experienced daily flashbacks to that time of torment. She still suffers from insomnia, cardiac palpitations, and shortness of breath. Coupled with proof of negligent infliction of emotional distress, this evidence justifies substantial compensation under New Jersey law. Of course, the task of valuing noneconomic losses in tort eases is an imprecise exercise. There is no one “correct” sum, but, rather, a range of acceptable awards. In many instances the spread between the high and low *1162 ends of the range will be great. The choice within the range — which by its nature requires the decisionmaker to translate intangibles (such as pain and suffering) into quantifiable dollars and cents — is a choice largely within the jury’s ken. See Correa, 69 F.3d at 1197. Since we are unable to conclude on this record that $200,000 is a figure beyond the wide universe of acceptable awards, we must uphold the district court’s finding that the figure is not excessive. See Ruiz, 929 F.2d at 34 (explaining that the court of appeals “cannot, and will not, without substantial cause, overrule a trial judge’s considered refusal to tamper with the damages assessed by a jury”). VI. CONCLUSION We need go no further. The record adequately supports the jury’s conclusion that the defendant’s inexplicable delay in calling an ambulance constituted a proximate cause of James Blinzler’s death and negligently inflicted both emotional distress and a loss of consortium on his wife (now his widow). Finding, as we do, that the law of New Jersey permits this multifaceted conclusion to remain fully intact, that the defendant’s several challenges to evidentiary and case-management rulings are meriüess, and that the damages awarded are not grossly excessive, we reinstate the jury verdict in its entirety. As a necessary corollary, we vacate the district court’s entry of judgment for the defendant on count 3. Affirmed in part and reversed in part. Costs in favor of the plaintiff. Footnotes [1] Indeed, this kind of predictive approach is among our conceptions of law itself. See Oliver Wendell Holmes, The Path of the Law, 10 Harv. L.Rev. 457, 461 (1897) ("The prophecies of what the courts will do in fact, and nothing more pretentious, are what I mean by law."). [2] It is commonly thought that the "substantial possibility” standard is more lenient than a standard that requires a plaintiff to prove it is more likely than not that a defendant’s failure to act constituted a substantial factor in bringing about the victim’s injury or death. See. W. Page Keeton et al., Prosser & Keeton on Torts § 41, at 44 (Supp.1988). [3] Like Oklahoma, New Jersey has explicitly adopted section 323 of the Restatement for use in loss of chance cases involving medical malpractice. See Scafidi v. Seiler, 119 N.J. 93, 574 A.2d 398, 405 (1990); Evers v. Dollinger, 95 N.J. 399, 471 A.2d 405, 415 (1984). [4] New Jersey is not alone in its reluctance blindly to follow Dillon's lead. See, e.g., D’Ambra v. United States, 114 R.I. 643, 338 A.2d 524, 528 (1975) (rejecting rigid foreseeability focus). Indeed, even the progenitor of the doctrine has had second thoughts. See Thing v. La Chusa, 48 Cal.3d 644, 257 Cal.Rptr. 865, 877-78, 771 P.2d 814, 826 (1989) (retreating from Dillon on this point). [5] In Portee, the question was not raised squarely. There the plaintiff (the victim’s mother) arrived at the scene after her son became trapped in an elevator. She did not witness either the initial entrapment or the act of negligence (faulty maintenance) that caused the accident. It was quite clear, however, that the mother knew immediately that her child’s injuries had an unnatural cause and stemmed from the elevator's accidental collapse. [6] The supplemental evidence consisted of testimony from two witnesses. The first, plaintiff's medical expert, simply clarified and confirmed his earlier testimony that James Blinzler would have survived had the ambulance arrived ten minutes earlier. The second witness (an employee of the ambulance service) testified that the ambulance service had a unit ready, available, and on call at 8:35 p.m. on November 13, 1992. [7] This is consistent with the method of the Civil Rules. Rule 50(a) exists in part to afford the responding party "an opportunity to cure any deficiency in that party’s proof that may have been overlooked until called to the party’s attention by a late motion for judgment.” Fed.R.Civ.P. 50, advisory committee’s note (1991 amendment). In other words, Rule 50(a) should be construed “to avoid tactical victories at the expense of substantive interests.” Moore, supra, ¶ 50.08, at 50-89 The district court echoed this sentiment when it granted the motion to reopen, stating: "I allow the plaintiff to reopen because I want the truth. I want the facts. I want to achieve a just result in this case....” [8] At one time New Jersey courts did require proof of "substantial bodily injury or sickness” in all emotional distress cases. See, e.g., Caputzal v. The Lindsay Co., 48 N.J. 69, 222 A.2d 513, 515 (1966); Falzone v. Busch, 45 N.J. 559, 214 A.2d 12, 17 (1965). Portee changed this rule in respect to bystander liability, permitting recovery in the absence of physical symptoms if the circumstances are such that severe emotional distress can easily be inferred. See Portee, 417 A.2d at 527-28.