*358OPINION OF THE COURT
NYGAARD, Circuit Judge.James E. Gottshall appeals from a summary judgment in favor of Consolidated Rail Corporation in this action under the Federal Employers’ Liability Act (FELA), 45 U.S.C. § 51 et seq. In the context of the facts that follow, we will examine whether FELA recognizes negligent infliction of emotional distress as a basis for liability, and if so, define the contours of this liability. We will construe the facts in the light most favorable to Gottshall. Metzger v. Osbeck, 841 F.2d 518, 519 (3d Cir.1988). We hold that Gottshall has shown a sufficient basis to sustain his suit under FELA. Conrail is not entitled to judgment as a matter of law, and we will reverse and remand for further proceedings.
I.
On an exceedingly hot and humid August day during the summer of 1988, Conrail dispatched a work gang of nine men and among them were James Gottshall and his friend Richard Johns. The crew was supervised by Michael Norvick. Conrail sent its men to replace a stretch of defective tracks on the Watertown Secondary near Turbotville, Pennsylvania. Despite many technological advances in the railway industry, replacing steel rails still requires raw human labor. It remains heavy, strenuous work, requiring workers to unload, carry, saw, and drill holes in steel rails, and to extract spikes. Because trains were operating on the track, work did not begin until high noon when the temperature had exceeded 95 degrees. The temperature of the rail itself was 118 degrees. The track was out in the flats — in the open.
Most of the men were in their fifties and many were overweight. Conrail knew that one worker had suffered a serious heart attack. It also knew Johns was overweight, had high blood pressure and athero or arteriosclerotic cardiovascular disease, and was taking medication.
Conrail was on a strict time schedule because, as Gottshall explained, it had violated a railway safety regulation and knew it was scheduled for a safety inspection. On this particular day Conrail drove its men hard; the pace of the work was unusually fast. Although conditions became increasingly inhospitable, the men were discouraged repeatedly from taking rest breaks except to get water on a need basis. As a practical matter it was difficult for workers to take unscheduled breaks because they often worked in teams. As Norvick stated, “We aren’t going to stop our maintenance work because of the heat.”
Under these stark conditions the men worked continuously for about two and a half hours until the work routine was unexpectedly disrupted. While Johns was cutting a rail, Gottshall saw him collapse. Gottshall and several men rushed to help their coworker. Norvick saw that Johns was pale and sweating profusely and realized he was having trouble with the weather conditions. They administered a cold compress and soon Johns regained consciousness. Ever aware of the time constraints on the work to be done, Norvick ordered the men to stop assisting Johns and to get back to work. They did so, leaving Johns with Norvick who neither took Johns from the worksite nor sought medical assistance. Five minutes later, Gottshall saw Johns stand up and collapse again.
Gottshall rushed to help his friend and saw that Johns was in trouble. Johns had turned white; his teeth had been knocked out by the fall; his eyes were rolled back; he was gasping for breath; his heart was fluttering; and saliva was drooling from his mouth. Gottshall realized Johns was suffering from a heart attack and began cardiopulmonary resuscitation. At one point Gottshall managed to restart Johns’ heart, but only briefly. Although Gottshall was emotionally perturbed, and at times crying, he continued the cardiac procedure for about forty minutes while waiting for medical help.
This time Norvick realized Johns needed medical help immediately. The men had communication equipment, but that equip*359ment was useless because on this particular day Conrail had taken the Turbotville base station off the air for repairs without notifying the men. Norvick was forced to physically seek help.
Rather than take the dying Johns to the nearest town, Norvick sped along an old country road to a firehall, only to find it closed. Sometime after the initial attempted distress call, the radio equipment apparently became operational and Norvick managed to reach a Conrail dispatcher. Nor-vick then raced to Agway Feed to place a 911 emergency call just in case the dispatcher’s lines would not work. The work-site was isolated, so Norvick arranged to meet the medical help at a nearby road. When the paramedics arrived some thirty minutes to an hour later, he led them through a path to the worksite.
By this time, it was too late. Johns had died. His corpse was covered and laid on the gravel beside the track where it remained in the open, under the hot sun and in full view of the men until the coroner arrived, some three hours later. Meanwhile, Norvick ordered the men back to work because an empty coal train needed to use the railway line. When the men finished the work, Norvick ordered Gotts-hall and the others to stay at the worksite.
After the coroner finished, Gottshall and several men carried Johns’ body to the ambulance parked in the distance. The coroner reported that Johns had died from a heart attack, precipitated by the excessive heat and humidity, combined with the heavy physical exertion. He also found that Johns did not receive prompt medical attention and that, had he, his chances of survival would have been significantly enhanced.
From the outset, the incident hit Gotts-hall hard. Other workers noticed he was emotional and upset during the ordeal. While he was giving CPR to Johns, he kept repeating, “Come on Dick, breathe, breathe.” Even as the work crew returned from the worksite hours after Johns had died, a worker noticed Gottshall was still crying.
Johns was Gottshall’s friend, and their friendship went beyond work. Johns often visited Gottshall to socialize. They spent time on the weekends, socializing and discussing the railway industry. Johns planned to join the National Historical Society of which Gottshall is a member. They attended union meetings, worked and took their meals together, and sometimes went drinking together. They had enjoyed this relationship for fifteen years. Conrail knew they were “personal friends.”
When Gottshall returned to work the next morning, a Conrail supervisor reprimanded him for administering CPR to Johns. Otherwise, it was work as usual. The extreme weather conditions persisted. The men went back to the track and continued to work long, hard hours under the sun. Conrail did not implement scheduled breaks but water, as usual, was provided.
Gottshall began to feel sick and lost his appetite. He became preoccupied with the events surrounding Johns’ death, and he became increasingly afraid that he would die under the same extreme conditions.
Johns’ funeral was held the following weekend. The following Tuesday, Gotts-hall told his supervisor he was sick and was going home. The supervisor told him that he had to explain to the division engineer. As Gottshall recalled, “I had to explain to him after all of this heat and exhaustion, I’m getting sick and I can’t take it, I’m going home and taking a couple of days off, and I just got sick and just couldn’t go back anymore.”
He returned home, retreated to his basement, and stayed there until his father found him several days later. Gottshall, along with the others who witnessed the accident, underwent a stress test given by their local union. The results indicated that the general consensus among the men was that the death was avoidable and that the men exhibited considerable depression and rage towards Conrail. The test identified Gottshall as suffering the most and recommended he seek medical treatment.
Gottshall was admitted to the Northwestern Institute of Psychiatry and remained there under the care of Dr. Byron Braid for *360about three weeks. Doctor Braid diagnosed that Gottshall was suffering from major depression and post-traumatic stress disorder. Gottshall was having suicidal preoccupations, anxiety, sleep onset insomnia, cold sweats, loss of appetite, nausea, physical weakness, repetitive nightmares of the death scene and a fear of leaving home. During this time Gottshall lost forty pounds. Another psychiatrist, Dr. Gary Glass, and a clinical psychologist, Dr. Sharon Silberman, confirmed Dr. Braid’s diagnosis. Since his discharge from the hospital, Gottshall has continued to receive outpatient psychological treatment.
Gottshall brought an action against Conrail under FELA, 45 U.S.C. § 51 et seq., seeking damages for emotional and physical injuries suffered and alleging that Conrail’s negligence created the circumstances under which he was forced to watch and participate in the events surrounding the death of his friend. The district court reasoned that Gottshall’s allegations failed to satisfy the elements of any of the recognized common law theories of liability, including the bystander and the zone of danger tests. Gottshall v. Consolidated Rail Corp., 773 F.Supp. 778, 781 (E.D.Pa.1991). It further reasoned that Conrail did not breach a general duty of care because it considered the failure to provide means of communication the only possible conduct breaching that duty and that Conrail could not have reasonably foreseen that such conduct would lead to Gottshall’s injuries. Last, it applied the causation standard under Pennsylvania law and concluded that “Gottshall’s injury was just too far down the causal chain, and as a result linking defendant’s negligence to the harm treats the defendant like an insurer.” Id. at 784. For these reasons the district court granted Conrail’s motion for summary judgment, and Gottshall appeals, contending he is entitled to a trial on the merits under FELA.
II.
FELA creates a cause of action for railway employees who have been killed or injured by the railway’s negligence. It provides:
Every common carrier ... shall be liable in damages to any person suffering injury while he is employed by such carrier ... for such injury or death resulting in whole or in part from the negligence of any of the officers, agents, or employees of such carrier, or by reason of any defect or insufficiency, due to its negligence ....
45 U.S.C. § 51 (emphasis added). The statute nowhere draws a distinction between claims for physical or emotional injury.
In Atchison, Topeka & Santa Fe Ry. Co. v. Buell, 480 U.S. 557, 107 S.Ct. 1410, 94 L.Ed.2d 563 (1987), the Supreme Court addressed whether negligent infliction of an emotional injury is actionable under FELA. The employee there brought a FELA action alleging that the railroad harassed, threatened and intimidated him and that as a result he suffered “a mental breakdown, and certain associated physical disorders.” The Court first held that an employee retains his right to bring a FELA action for damages even though the injury was caused by conduct possibly subject to arbitration under the Railway Labor Act. 480 U.S. at 565-69, 107 S.Ct. at 1416-17. The Court then turned to the railway's contention that an employee’s “wholly mental injury” was not compensable under FELA. It declined to resolve this issue, noting that the record was incomplete as to the exact nature of the allegedly tortious activity and the extent of the injuries allegedly suffered. It opined that “broad pronouncements in this area may have to bow to the precise application of developing legal principles to the particular facts at hand.” Id. 480 U.S. at 570, 107 S.Ct. at 1418. Thus federal courts are left to develop the degree to which negligent infliction of emotional distress is actionable under FELA.
The Buell Court said, however, that federal courts should “glean[] guidance from common-law developments.” Id. 480 U.S. at 568, 107 S.Ct. at 1417. So, we first look to the common law. In seeking guidance we are not limited to the law of the forum state, but instead must gauge the development of the common law from a broader *361base. Outten v. National R.R. Passenger Corp., 928 F.2d 74, 77 (3d Cir.1991).
A.
Among states, a majority recognizes a cause of action in tort for negligent infliction of emotional distress. See Buell, 480 U.S. at 569-71 & n. 20, 107 S.Ct. at 1418 & n. 20. But from recognition to application, the doctrinal divergence is great.
A minority employs the “physical impact” rule.1 The rule requires a contemporaneous physical injury or impact to recover for negligent infliction of emotional distress. “[T]he purpose of the rule requiring physical impact is to prevent ‘illusory or imaginative or faked’ claims.” Zelinsky v. Chimics, 196 Pa.Super. 312, 318, 175 A.2d 351, 354 (1961). See Restatement (Second) of Torts, § 436A, comment b (1965).
A large majority of jurisdictions employs either the “zone of danger” or “physical manifestation” rules. See supra n. 1, Comment, 33 Vill.L.Rev. at 796-98 & n. 91; Restatement §§ 313, 436, 436A. The zone of danger rule permits recovery for emotional injuries resulting from witnessing physical harm to another or from fearing physical harm to oneself, provided that plaintiff was actually threatened by physical harm. The physical manifestation rule requires that plaintiff exhibit a physical injury or symptom as the direct and natural result of the initial emotional distress suffered.
Many jurisdictions also permit a bystander to recover for emotional injuries. The quintessential common law case involves the mother who witnesses the death of her child. See Dillon v. Legg, 68 Cal.2d 728, 69 Cal.Rptr. 72, 441 P.2d 912 (1968) (en banc). See generally supra n. 1, Comment, 33 Vill. L.Rev. at 806-07 n. 139 (as of 1988 at least ten states have adopted the Dillon approach). The Supreme Court of California restated the Dillon bystander test as this: Plaintiff can recover if she (1) is present at the accident scene at the time it occurs and contemporaneously perceives the injury to the victim, (2) suffers serious emotional distress as a result of watching the injury, and (3) is closely related to the injured victim. Thing v. La Chusa, 48 Cal.3d 644, 668-69, 257 Cal.Rptr. 865, 771 P.2d 814 (1989). As for the last factor, the court limited recovery to relatives residing in the same household, parents, siblings, children, and grandparents of the victim. 257 Cal.Rptr. at 880 n. 10, 771 P.2d at 829 n. 10.
While some jurisdictions require that the plaintiff and the victim stand in some familial relationship, others seem to suggest a showing of strong emotional ties will suffice. See Champion v. Gray, 420 So.2d 348, 353 (Fla.App.1982), “decision quashed” 478 So.2d 17, 20 (Fla.1985); Versland v. Caron Transp., 206 Mont. 313, 671 P.2d 583, 587 (1983); Paugh v. Hanks, 6 Ohio St.3d 72, 79-80, 451 N.E.2d 759, 766-67 (1983); Leong v. Takasaki, 55 Haw. 398, 520 P.2d 758 (1974); Hunsley v. Giard, 87 Wash.2d 424, 553 P.2d 1096, 1103 (1976); Toms v. McConnell, 45 Mich.App. 647, 207 N.W.2d 140, 144-45 (1973); Keck v. Jackson, 122 Ariz. 114, 593 P.2d 668, 670 (1979). Thus a plaintiff must show physical, temporal, and relational proximity.
Although these common law tests seem as different as they are arbitrary, and have been so criticized, they share common roots. While severe emotional injuries can be just as debilitating as physical injuries, they are not as manifest. One can see frayed skin, but one cannot see frayed nerves; hence, an emotional injury is easier to fake. These tests therefore have been judicially developed to screen causes of action and send only the meritorious ones to juries.
As an unfortunate product of this thinking, the focus shifted from traditional principles of liability to mechanical adherence to these tests, which are based upon a theory that liability is the extraordinary exception. Thus, even in eases of gross negligence resulting in reasonably foreseeable injuries, common law courts preserve *362the limits of the established rules, thereby-precluding recovery for many meritorious claims.2
B.
These common law tests provide instruments that are intended to separate the meritorious claims from the others, but they do not necessarily etch the contours of the federal right. While we assume that “FELA jurisprudence gleans guidance from common-law developments,” Buell, 480 U.S. at 568, 107 S.Ct. at 1417 (citing Urie v. Thompson, 337 U.S. 163, 172-74, 69 S.Ct. 1018, 1026, 93 L.Ed. 1282 (1949)), we are not called upon to resolve matters of state law, nor predict how state courts would resolve them. Cf. Erie R.R. Co. v. Tompkins, 304 U.S. 64, 78, 58 S.Ct. 817, 822, 82 L.Ed. 1188 (1938); Robertson v. Allied Signal, Inc., 914 F.2d 360, 364 (3d Cir.1990). Determining FELA liability is distinctly a federal question. Urie, 337 U.S. at 174-78, 69 S.Ct. at 1027-28.
While federal courts uniformly agree that claims for negligent infliction of emotional distress are cognizable under FELA, they are badly splintered as to the elements necessary to bring such a claim.
The Court of Appeals for the Seventh Circuit has held that FELA does not create a cause of action for emotional injuries absent actual or threatened physical impact. Ray v. Consolidated Rail Corp., 938 F.2d 704, 705 (7th Cir.1991) (quoting Lancaster v. Norfolk & Western Ry. Co., 773 F.2d 807, 813 (7th Cir.1985) (pre-Buell case) (“the critical question in this case is simply whether [plaintiff] has stated a claim under the FELA, which means a claim for a violation of one of the traditional ‘physical’ torts, such as assault, battery, and negligent infliction of personal injury, all of which are torts actionable under the [FELA]”)). Accord Hammond v. Terminal R.R. Ass’n, 848 F.2d 95, 98 (7th Cir.1988); Gillman v. Burlington N. R.R. Co., 878 F.2d 1020, 1025 (7th Cir.1989).
The Court of Appeals for the Ninth Circuit has taken the position that purely emotional injuries are compensable under FELA, though it has yet to delineate the circumstances in which recovery is permitted. Taylor v. Burlington N. R.R. Co., 787 F.2d 1309, 1313 (9th Cir.1986) (pre-Buell case). See Pierce v. Southern Pacific Transp. Co., 823 F.2d 1366, 1372 n. 2 (9th Cir.1987) (applying the “egg-shell plaintiff” doctrine to emotional claims if injury results in physical manifestation, in this case a heart attack).
The Court of Appeals for the Sixth Circuit has decided that FELA does not recognize a claim for intentional infliction of emotional distress resulting in purely emotional injuries. Adkins v. Seaboard Sys. R.R., 821 F.2d 340, 342 (6th Cir.1987). It has yet to decide whether a claim for negligent infliction of emotional distress without verifiable physical manifestation is actionable. See Stoklosa v. Consolidated Rail Corp., 864 F.2d 425, 426 (6th Cir.1988); Adams v. CSX Transp., Inc., 899 F.2d 536, 539 (6th Cir.1990). But in Stoklosa and Adams, the court took two potentially contradictory views of an employer’s duty to its employees to provide an “emotionally safe” workplace.
In Stoklosa the court concluded that an employer’s duty is defined by whether its conduct could be reasonably foreseen to cause an ’ employee’s emotional distress. 864 F.2d at 426. In Adams, however, the court concluded that “an employer has not breached its duty to provide an emotionally safe workplace unless the employer (acting through its agents) engages in ‘unconscionable abuse’ of an employee.” 899 F.2d at 540.3
*363The Court of Appeals for the Fourth Circuit requires plaintiff to show outrageous conduct as a predicate to a claim for negligent infliction of emotional distress under FELA, thus taking the same position as the Sixth Circuit Adams court. Elliott v. Norfolk & W. Ry. Co., 910 F.2d 1224, 1229 (4th Cir.1990) (citing Buell, 480 U.S. at 565-67 n. 13, 107 S.Ct. at 1416 n. 13; Netto, 863 F.2d at 1214-15). It declined to decide whether one may recover under FELA for purely emotional injuries without additional physical symptoms. Id.
The Court of Appeals for the First Circuit understands Buell as “an attempt to leave the door to recovery for wholly emotional injury somewhat ajar but not by any means wide open.” Moody v. Maine Cent. R.R. Co., 823 F.2d 693, 694 (1st Cir.1987). The court declined “to make this a pioneer case exploring the frontier possibility opened up by Buell” and rejected plaintiffs claim because he failed to show causation between the alleged injury and the tortious conduct. Id. at 694, 696. Likewise, in Robert v. Consolidated Rail Corp., 832 F.2d 3 (1st Cir.1987), the court avoided the issue and held that plaintiff failed to establish that Conrail could have reasonably foreseen that he would suffer a heart attack from stress of which Conrail was never informed. It declined to resolve whether a heart attack induced solely by stress and not accompanied by any precipitating physical injury is a compensable injury. Id. at 7. But see Pierce, 823 F.2d at 1372 n. 2 (emotional distress leading to heart attack is actionable under FELA).
The Court of Appeals for the Fifth Circuit has decided a series of cases on this issue. In Hagerty v. L & L Marine Serv., Inc., 788 F.2d 315 (5th Cir.1986) (pre-Buell *364case), plaintiff was drenched with toxic, carcinogenic chemicals and suffered emotional injuries arising from cancerphobia. The court concluded that FELA allows plaintiffs to recover for psychic and emotional harm because it was intended to provide broad coverage for all work-related “injuries” whether characterized as physical or mental. Id. at 318. On reconsideration, the court modified Hagerty, stating that a plaintiff may recover for serious mental distress assuming an “actionable injury.” 797 F.2d 256 (1986).
In Netto v. Amtrak, 863 F.2d 1210 (5th Cir.1989) the court declined to decide whether a plaintiff may recover for purely emotional injuries under FELA because he had failed to show any evidence of unconscionable or outrageous conduct. Id. at 1214. While the Sixth Circuit position in Adams and the Fourth Circuit position in Elliott purportedly adopted the Netto holding, it is clear Netto is of a different genre because the court emphasized that the claim was only for intentional infliction of emotional distress. Id. at 1214 n. 4. The Netto court did not require unconscionable conduct for all claims for emotional distress. Rather, since it was presented with a claim based on intentional infliction of emotional distress only, it had no occasion to discuss negligent infliction of emotional distress. See supra n. 3.
In Gaston v. Flowers Transp., 866 F.2d 816 (5th Cir.1989), a seaman saw his half-brother crushed to death and sued under the Jones Act.4 The facts indicate that plaintiff, in some capacity, was not merely a witness, but was an active participant in the events: “Gaston [plaintiff] fell to the deck; but James [victim] slipped between the barge and the vessel and was crushed to death, despite Gaston’s efforts to pull him to safety.” Id. at 817. The court refused to provide recovery for an emotional injury stemming from witnessing the violent death of a relative, reasoning that the plaintiff’s injury was the result of watching another person die and was not a result of fear for his own safety. It thus rejected the bystander rule of liability and required that plaintiff be in the zone of danger to recover for emotional distress. Id. at 819 (restricting Hagerty as a “physical impact” case). The court left open the question whether a purely emotional injury is cognizable. Id. at 821.
In Wilson v. Zapata Off-Shore Co., 939 F.2d 260 (5th Cir.1991), the plaintiff sued under the Jones Act for emotional harm suffered as a result of sexual harassment. The evidence showed that the plaintiff was subjected to tortious physical contact and that she suffered physical manifestations of harm, including weight loss, vomiting, and diarrhea. Id. at 265. The court held that in these circumstances the plaintiff’s claim was cognizable under the Jones Act. The court did not reach the issue of whether a purely emotional injury would be cognizable. Id. at 266 n. 8.
In Plaisance v. Texaco, Inc., 937 F.2d 1004 (5th Cir.1991), the plaintiff, a tug boat captain with over 30 years experience, witnessed an explosion when a barge ruptured a pipeline. He helped to ease the emergency by moving several vessels to safety. No one was injured in the accident. Soon thereafter, he developed post-traumatic stress disorder and depression resulting from his belief that he could have been injured or killed. He sought to recover under the Jones Act for a purely emotional injury.
After thoroughly reviewing the case law and the different theories of recovery under common law, the court concluded that a plaintiff may recover if the injury resulted from a situation in which a reasonable person, normally constituted, would not be able to cope adequately with the mental distress arising from the circumstances. Id. at 1010. It reasoned “one who has suffered a significant emotional injury should not be denied recovery simply because the factual scenario does not satisfy some earlier-selected, fictitious method of authenticating the injury.” Id. at 1011. The court stated in a broad stroke:
*365Excepting our colleagues in the Ninth Circuit, our circuit brethren have indicated a reluctance to field the Buell pitch. We end that reluctance today, persuaded that an emotional injury can be every bit as harmful, debilitating, and destructive of the quality of one’s life as a physical injury. We therefore hold that a claim for an emotional injury caused by emotional distress negligently inflicted, even without an accompanying physical injury or physical contact, is cognizable under the FELA.
Id. at 1009. Because one would not expect an experienced seaman to have such emotional reactions to fires in the oilfield, the court concluded that the injury was not a reasonably foreseeable consequence and therefore the claim was not cognizable. Indeed, the treating psychiatrist testified that he was baffled by the plaintiffs reaction to the accident.
The Court of Appeals for the Fifth Circuit granted en banc rehearing. 954 F.2d 266 (1992). The en banc court agreed that the plaintiff’s claim was not cognizable even under the most liberal theory of recovery. 966 F.2d 166 (1992). The court, however, concluded that the facts did not permit the court to decide whether a claim for purely emotional injuries is cognizable. It emphasized that Gaston continued to represent the law of the circuit. “The [Ga-ston\ court recognized that a plaintiff suing for purely emotional injuries could recover if he also suffered physical injury. The court also determined that a bystander could not recover for merely witnessing harm to another where the plaintiff suffered no harm or threat of harm.” Id. at 169. But the en banc court believed that Gaston left open the question whether a claim for purely emotional injury under a zone of danger theory is cognizable. Id.
It is clear that there are open conflicts among the circuits.5 A review of the decisions in other circuits in the wake of Buell indicates that there is no common discerna-ble principle, test, view or attitude.
C.
We have refused to designate a particular common law test as the test in this circuit, choosing instead to take an ad hoc approach that focuses more on the facts rather than rules of law. In Holliday v. Consolidated Rail Corp., 914 F.2d 421 (3d Cir.1990), Holliday brought a FELA action against the railroad alleging that he was injured from the stress created when Conrail forced him to work on a job for which he claimed to have been unqualified. We held that the facts in the light most favorable to Holliday, considering the character of the allegedly tortious activity (putting *366him in a position for which he was unqualified) and the lack of physical impact on him, could not support a FELA claim. Id. at 427.
Holliday’s claim was essentially for emotional injuries suffered from job stress. If an employee showing some possible negligence with psychological consequences will always present a jury question, even the most attenuated claims could be advanced. Because claims for negligent infliction of emotional distress are so fact-dependent, we declined to make broad pronouncements that could restrict future meritorious claims. We did not hold that there could not be a recovery for emotional injuries unless the plaintiff suffers a physical injury or unless there is an accident, as we did not need to reach these issues. We stressed that the allegedly tortious activity did not result in a physical injury to another party, which Holliday witnessed, or indeed injury to another party at all. In fact, there was no accident. Thus “our opinion is narrow.” Id. at 426.
In Outten v. National R.R. Passenger Corp., 928 F.2d 74 (3rd Cir.1991), Outten brought a FELA action for emotional injuries he sustained when he saw two railway cars were about to collide. There was a collision, but it took place a full mile away from Outten. Outten did not see the accident because he was running in the other direction and, like Holliday, no one was hurt. Outten failed to show how the railway breached a duty: He was neither a bystander who witnessed a closely-related person injured, see Dillon v. Legg, 441 P.2d 912 (1968), nor an active participant in the injury-causing event, see Althoff v. Consolidated Rail Corp., 1988 WL 61734, 1988 U.S.Dist. Lexis 5343 (E.D.Pa.1988) (when a worker operated the instrumentality that killed a coworker, he became an “active participant” in the accident through the employer’s negligence). There was thus no “solid basis in the present state of common law” to permit recovery. 928 F.2d at 79. Because the Supreme Court has cautioned us to finely scrutinize the facts of the case, we declined to devise a generally applicable litmus test for FELA claims based on negligent infliction of emotional distress.
The core of the problem lies in the fundamental tension between two conflicting policies. On the one hand, common law doctrines are built on the suspicion that emotional distress claims are subject to fraud and frivolity. In Outten we rejected the notion that FELA liability extends to emotional injuries suffered when an employee “reasonably feared personal physical injury,” believing such a rule would remove reasonable limits on these types of claims. Id. at 79. And in Holliday we adopted the policy considerations set forth in Kraus v. Consolidated Rail Corp., 723 F.Supp. 1073, 1090 (E.D.Pa.1989): a need to prevent incalculable and potentially unlimited damages, a flood of litigation, and fraud. 914 F.2d at 424.
On the other hand, contrasting the often harsh and arbitrary doctrines of the common law is the policy underlying FELA itself. FELA was enacted in broad, open-ended terms so that recovery will be liberally granted and that an employer’s standard of conduct will be malleable, and adapt to the changing realities of industrial conditions and standards. While courts must be neither an instrument for the opportunistic con-artist nor an understanding shoulder for the emotionally frail, we must follow Congress’ mandate. So we must reconcile the two policies, discerning the trends and developments in common law and then gauging them according to the policy underlying FELA, for FELA is at issue here and creates the cause of action.
D.
Congress promulgated FELA to provide recovery in meritorious cases and to remedy deficiencies in the common law. At common law the liability of the master to his servant was predicated on generally applicable tort laws. The road to liability was treacherous, requiring the servant to negotiate through various defenses designed to limit liability. This limited liability insulated fledgling industries, too weak to “bear[] the ‘human overhead’ which is an inevitable part of the cost — to someone — of the doing of industrialized busi*367ness.” Tiller v. Atlantic Coast Line R.R. Co., 318 U.S. 54, 59, 63 S.Ct. 444, 447, 87 L.Ed. 610 (1943). As railways prospered and grew strong and because workers lacked the economic and bargaining power to provide for their safety, the policy shifted to place the burden of the “human overhead” on the railway industry. Kernan v. American Dredging Co., 355 U.S. 426, 432, 78 S.Ct. 394, 397-98, 2 L.Ed.2d 382 (1958). The statute considered the railway industry as a “unitary enterprise, its economic resources obligated to bear the burden of all injuries befalling those engaged in the enterprise arising out of the fault of any other member engaged in the common endeavor.” Sinkler v. Missouri Pacific R.R. Co., 356 U.S. 326, 330, 78 S.Ct. 758, 762, 2 L.Ed.2d 799 (1958). FELA therefore provides compensation for the fatalities and injuries the railway industry inevitably imposes. Wilkerson v. McCarthy, 336 U.S. 53, 68, 69 S.Ct. 413, 420, 93 L.Ed. 497 (1949) (Douglas concurring).
Although FELA is predicated on common law negligence principles, Chesapeake & Ohio Ry. Co. v. Stapleton, 279 U.S. 587, 589-90, 49 S.Ct. 442, 443, 73 L.Ed. 861 (1929), instead of a detailed statute codifying them Congress enacted the statute in general terms. See 45 U.S.C. § 51. Congress left the onus of fashioning remedies to the courts in a manner analogous to the development of common law tort remedies. Kernan, 355 U.S. at 431-33, 78 S.Ct. at 398. But it is clear that courts should liberally grant relief. And it is equally clear that “Congress did not intend to create a static remedy, but one which would be developed and enlarged to meet the changing conditions and changing concepts of industry’s duty toward its workers.” Id. (emphasis added). See Coray v. Southern Pacific Co., 335 U.S. 520, 524, 69 S.Ct. 275, 277, 93 L.Ed. 208 (1949) (“Congress has thus for its own reasons imposed extraordinary safety obligations upon railroads and has commanded that if a breach of these obligations contributes in part to an employee’s death, the railroad must pay damages.”); Ries v. National R.R. Passenger Corp., 960 F.2d 1156, 1165 (3d Cir.1992) (concurrence) (“[FELA] ensures that a common carrier’s duties include, and extend, the spectrum of common law standard of recovery.”) (emphasis in original); Ackley v. Chicago & N.W. Transp. Co., 820 F.2d 263, 267 (8th Cir.1987) (“duty to provide a reasonably safe place of work ... is broader under the statute than a general duty of due care”). See also Jamison, 281 U.S. at 638-41, 50 S.Ct. at 442 (extending the scope of FELA liability to cover some intentional torts even though the language of the statute refers only to negligent conduct).
Thus many traditional impediments to recovery found in common law have been eradicated by FELA itself or by the courts following the intent and dictates of the Act. “A primary purpose of the Act was to eliminate a number of traditional defenses to tort liability and to facilitate recovery in meritorious cases.” Buell, 480 U.S. at 561, 107 S.Ct. at 1413. We see evidence of this in the statute. At common law the doctrines of contributory negligence and assumption of the risk produced harsh and unfair results, often barring recovery for the victim. But sections 53 and 54 of FELA respectively limit the application of these defenses. At common law an at-will employee may be fired for any or no reason. See for example, Geary v. United States Steel Corp., 456 Pa. 171, 319 A.2d 174, 176 n. 4 (1974). But section 60 prohibits an employer from blocking or otherwise regulating the free flow of information concerning a FELA matter, including prohibiting retaliatory discharge of an employee who furnishes such information. And section 55 prohibits an employer from contracting to limit or otherwise regulating its FELA liability.
Some of the inequities the statute did not eradicate, the courts have. In Rogers v. Missouri Pacific R.R. Co., 352 U.S. 500, 77 S.Ct. 443, 1 L.Ed.2d 493 (1957), the Court expounded upon the relationship between FELA and the common law:
[FELA] was enacted because the Congress was dissatisfied with the common-law duty of the master to his servant. The statute supplants that duty with the far more drastic duty of paying damages for injury or death at work due in whole *368or in part to the employer’s negligence. The employer is stripped of his common-law defenses and for practical purposes the inquiry in these cases today rarely presents more than the single question whether negligence of the employer played any part, however small, in the injury or death which is the subject of the suit.
352 U.S. at 507-08, 77 S.Ct. at 449. Thus the Court did away with the strict requirements of causation in common law and replaced it with the inquiry of “simply whether the proofs justify with reason the conclusion that employer negligence played any part, even the slightest, in producing the injury or death for which damages are sought.” Id. 352 U.S. at 506, 77 S.Ct. at 448.
In Kernan v. American Dredging Co., 355 U.S. 426, 78 S.Ct. 394, 2 L.Ed.2d 382 (1958) an employee was killed because the employer violated a regulation that required a navigation light to be at a certain height above the water to provide visibility for passing ships. Because the light was too low, it ignited fumes and caused an explosion. The Supreme Court held that the violation constituted negligence per se under section 51.
Traditionally,' common law limited the negligence per se doctrine to injuries the statute was designed to prevent. See Restatement § 286, Comment on Clause (c). The Court, however, rejected this common law limitation and concluded that FELA liability springs from the violation of statutory duty regardless of the common law limitation. 355 U.S. at 437-39, 78 S.Ct. at 401. The Court succinctly articulated the theory and purpose of FELA liability:
The courts, in developing the FELA with a view to adjusting equitably between the worker and his corporate employer the risks inherent in the railroad industry, have plainly rejected many of the refined distinctions necessary in common-law tort doctrine for the purpose of allocating risks between persons who are more nearly on an equal footing as to financial capacity and ability to avoid the hazards involved.... [T]he theory of the FELA is that where the employer’s conduct falls short of the high standard required of him by this Act, and his fault in whole or in part, causes injury, liability ensues.
Id. Because the worker and the railway employer do not stand on equal footing, and because railway employment remains dangerous, FELA was intended to adjust the equities. See Sinkler, 356 U.S. at 329-31, 78 S.Ct. at 762.
In Urie v. Thompson, 337 U.S. 163, 69 S.Ct. 1018, 93 L.Ed. 1282 (1949), the Court allowed an employee to recover for silicosis, a lung disease, even though the statute violated was designed to ensure an adequate auxiliary braking system and not to protect employees against silicosis. Again the Court disregarded common law limitations:
To read into this all-inclusive wording [of FELA] a restriction as to the kinds of employees covered, the degree of negligence required, or the particular sorts of harms inflicted, would be contradictory to the wording, the remedial and humanitarian purpose, and the constant and established course of liberal construction of the Act followed by this court.
337 U.S. at 181-82, 69 S.Ct. at 1030. The Court again stressed that FELA creates the cause of action and that FELA grants relief for meritorious claims for injuries as a result of an employer’s negligent conduct.
The Buell Court cited Rogers and Urie for the proposition that causation in a FELA action is a lenient inquiry and that FELA is to be given a liberal construction. 480 U.S. at 560-62 n. 8, 562-64, 107 S.Ct. at 1413 n. 8, 1414. The policies underscoring Rogers, Kernan and Urie — the humanitarian purpose of shifting the costs of “human overhead” to the railway industry — remain as vital today as then and do not change with the nature of the injury. See Urie, 337 U.S. at 181, 69 S.Ct. at 1030 (“The wording [of FELA] was not restrictive as to ... the particular kind of injury resulting.”).
While the long-standing fundamental elements of bringing a FELA action are *369traditional common law elements, Robert, 832 F.2d at 6, in many ways FELA is the antithesis of the common law. FELA imposes upon carriers a higher standard of conduct and has eliminated many of the refined distinctions and restrictions that common law imposed to bar recovery (even on meritorious claims). FELA liability and common law liability are thus different. The inexorable conclusion is: State law tests are not the sine qua non for determining whether plaintiff has a cognizable FELA claim and doctrinal common law distinctions are to be discarded when they bar recovery on meritorious FELA claims.6
Although several courts of appeals have adopted to the letter or rejected one or more of the common law tests, we find it unnecessary, and indeed undesirable, to do so for as one district court astutely observed, such a solution leads to chaos: “the federal courts in the fifty states might produce as many as fifty versions of what should be unified federal law.” Teague, 708 F.Supp. at 1348 n. 5. Cf. Lancaster, 773 F.2d at 813 (7th Cir.) (adopting the physical impact and zone of danger tests); Gaston, 866 F.2d at 819 (5th Cir.) (adopting the physical impact and zone of danger tests and rejecting the bystander test); Netto, 863 F.2d at 1214 (5th Cir.) (requiring unconscionable conduct for claims based on intentional infliction of emotional distress); Adams, 899 F.2d at 540 (6th Cir.) (requiring unconscionable conduct for claims based on negligent infliction of emotional distress); Elliott, 910 F.2d at 1229 (4th Cir.) (same). By focusing on these exceptions, courts have gone their separate ways while ignoring the one thing that should unite them — the broad remedial scope of FELA. And by strictly adhering to them, we would, as some common law courts have chosen to do, preclude a great number of meritorious claims.
As we have seen, the common law has developed many doctrines and elaborate rules when dealing with emotional harm claims. At the heart of all these doctrines and rules lies but one function, to glean the meritorious from the feigned and frivolous. See Restatement § 436 A, comment b. For example, the physical impact and zone of danger rules are based on the notion that a plaintiff who has been subjected to or threatened with the negligent force would be less likely to fake a claim and would likely suffer genuine and severe emotional injuries. See Neff v. Lasso, 382 Pa.Super. 487, 555 A.2d 1304, 1307 (1989); Zelinsky, 175 A.2d at 354. And the physical manifestation rule is based on the notion that physical symptoms of emotional distress show that the plaintiff “really” suffered an injury. Payton v. Abbott Lab., 386 Mass. 540, 437 N.E.2d 171, 175 & n. 5 (1982). See Molien v. Kaiser Foundation Hosp., 27 Cal.3d 916, 167 Cal.Rptr. 831, 838, 616 P.2d 813, 820 (1980) (“It supposedly serves to satisfy the cynic that the claim of emotional distress is genuine.”). The Supreme Court has long recognized that emotional distress and suffering are compensable under FELA, so long as that emotional distress can be objectively validated to a sufficient degree by a precipitating physical injury. Erie R.R. Co. v. Collins, 253 U.S. 77, 85, 40 S.Ct. 450, 453, 64 L.Ed. 790 (1920) (plaintiffs emotional distress, including “shame and humiliation” and “mental pain,” is cognizable under FELA when validated by a physical mutilation and disfigurement), overruled on other grounds, Chi*370cago & E. Ill. R.R. Co. v. Industrial Comm’n, 284 U.S. 296, 299, 52 S.Ct. 151, 152, 76 L.Ed. 304 (1932). See East Ala. Express Co. v. Dupes, 271 Ala. 504, 124 So.2d 809 (1960) (plaintiff who sustained whiplash and fracture allowed to recover for her “worry” about the future result of her injury). “The problem from this perspective is one of adequate proof, and it is not necessary to deny a remedy in all cases because some claims may be false.” The William Prosser, Law of Torts, § 54, at 361 (5th ed. 1984).
That which distinguishes merit from fraud and frivolity most often depends on one issue — whether there was injury in fact reasonably attributable to the negligent conduct — for the law does not reward those who did not suffer injuries or whose injuries are trivial, or those who are faint of heart. See Restatement § 46 comment j. Dean Prosser suggests:
It is entirely possible to allow recovery only upon satisfactory evidence and deny it when there is nothing to corroborate the claim, or to look for some guarantee of genuineness in the circumstances of the case.
Sfc * 5}: ij! * ¡fc
What all of these cases appear to have in common is an especial'likelihood of genuine and serious mental distress, arising from the special circumstances, which serves as a guarantee that the claim is not spurious. There may perhaps be other such cases. Where the guarantee can be found, and the mental distress is undoubtedly real and serious, there may be no good reason to deny recovery.
The Law of Torts, § 54, at 361, 362. See Johnson v. New York, 37 N.Y.2d 378, 383-84, 372 N.Y.S.2d 638, 643, 334 N.E.2d 590, 593 (1975) (“recovery for emotional harm to one subjected directly to the tortious act may not be disallowed so long as the evidence is sufficient to show causation and substantiality of the harm suffered, together with a ‘guarantee of genuineness’ ”); Rodrigues v. State, 52 Haw. 283, 52 Haw. 156, 472 P.2d 509, 520 (“the general standard of proof required to support a claim of mental distress is some guarantee of genuineness in the circumstances of the case”); Restatement § 436 A, comment b (“in the absence of the guarantee of genuineness provided by resulting bodily harm, such emotional disturbance may be too easily feigned”). Thus the contours of the common law rules of liability (and preclusion) can best be understood as encompassing, albeit crudely, only those special situations most likely to present a claim with sufficient indicia of genuineness. See The Law of Torts, § 54, at 360-66; Hagerty, 788 F.2d at 318.
But the guarantee is not fixed by the limited set of circumstances the common law has ordained. The common law tests provide only a dubious method for assuring that the injury is genuine. We find guidance in Hagerty, where the Court of Appeals for the Fifth Circuit rejected the common law’s doctrinal and definitional prerequisites for bringing an emotional distress claim:
The physical injury requirement, like its counterpart, the physical impact requirement, was developed to provide courts with an objective means of ensuring that the alleged mental injury is not feigned. We believe that notion to be unrealistic. It is doubtful that the trier of fact is any less able to decide the fact or extent of mental suffering in the event of physical injury or impact.... The circumstances surrounding the fear-inducing occurrence may themselves supply sufficient indicia of genuineness. It is for the jury to decide questions such as the existence, severity and reasonableness of the fear.
788 F.2d at 318 (citation omitted). See Ferrara v. Galluchio, 5 N.Y.2d 16, 176 N.Y.S.2d 996, 999-1000, 152 N.E.2d 249, 252 (1958) (cancerphobia as a result of negligent X-ray treatment is actionable as the court found a “guarantee of genuineness [of the cancerphobia] in the circumstances of the case”); Molien, 616 P.2d at 821 (a false diagnosis of spouse having syphilis is sufficient to provide a guarantee of genuineness and allow recovery for a purely emotional injury). See also Elden v. Sheldon, 46 Cal.3d 267, 284, 250 Cal.Rptr. 254, 758 P.2d 582 (1988) (Broussard dissenting) *371(“In my view, recovery should not be cut off on arbitrary, definitional grounds but on functional grounds that correspond with real loss. As the commentators have suggested, the problem should be solved by the application of the principles of tort, not by the creation of exceptions to them.”).
The solution is not found in rules of law so much as in an active scrutiny of the facts. The search for the guarantee is a fluid analysis, and it “is not necessarily an abstract point of law ... that might be answerable without exacting scrutiny of the facts of the case.” Buell, 480 U.S. at 568, 107 S.Ct. at 1417. Although a fact-specific analysis may mean losing the comfort and security that bright-line, arbitrary rules afford, FELA jurisprudence cannot be reduced to a set of rules that mechanistically drive results without exacting scrutiny of the facts as they bear upon such considerations as the employer’s duty to provide a safe workplace, the nature of the alleged tortious conduct, the injury, and the underlying equities. “[A]s Judge Becker noted in a very different context [ ] ‘developing the law on a case-by-case basis and drawing lines depending on the facts is the stuff of judging.’ ” Holliday, 914 F.2d at 427 (citation omitted) (quoting United States v. Balascsak, 873 F.2d 673, 685 (3d Cir.1989) (concurrence)). See Ries, 960 F.2d at 1167 (concurrence) (“The development of FELA, like common law principles, should proceed on a case by case [basis].”). We implicitly recognized that we would not limit our inquiry to the common law rules of law when we adopted the “active participation” rule of liability explained in Althoff, for that rule does not fall within the parameters of any common law theory of liability. See Outten, 928 F.2d at 79.
III.
The issue is whether the factual circumstances here provide a threshold assurance that there is a likelihood of genuine and serious emotional injury. We look not to any specific fact or standard, but to the totality of the factors. Thus, one consideration is whether plaintiff has a “solid basis in the present state of common law to permit him to recover.” Outten, 928 F.2d at 79.
We first consider the common law bystander theory of recovery. Had Gottshall and Johns been father and son this would represent the classic common law claim. In some jurisdictions, courts have discarded the familial proximity element, recognizing its arbitrary character. In Hunsley, the Supreme Court of Washington commented, “We decline to draw an absolute boundary around the class of persons whose peril may stimulate the mental distress. This usually will be a jury question bearing on the reasonable reaction to the event unless the court can conclude as a matter of law that the reaction was unreasonable.” 553 P.2d at 1103. Other courts have concluded similarly. See Leong, 520 P.2d 758 (“Neither should the absence of a blood relationship between victim and plaintiff-witness foreclose recovery.”); Paugh, 451 N.E.2d at 766-67 (“With respect to the third factor, we believe that a strict blood relationship between the accident victim and the plaintiff-bystander is not necessarily required.”). Indeed, in the seminal case of Dillon v. Legg, the Supreme Court of California warned that the three proximity elements should serve as flexible guidelines, each shading into each other, and that liability would need to be determined on a case-by-case basis. 69 Cal.Rptr. at 80-81, 441 P.2d at 920-21. Considering that Gottshall and Johns had been friends for fifteen years, in these jurisdictions Gotts-hall’s claim would be cognizable and would be presented to the jury.
Even if Gottshall does not satisfy the traditional bystander test, the principle underlying the rule applies in this FELA case. In light of the relationship between FELA and the common law, and the policies governing FELA, the lack of a familial relationship between the plaintiff and the victim does not necessarily remove the indicia of genuineness.
We must provide relief for injuries consistent with the realities of working in the railway industry. The conditions and concepts of industry’s duties toward its workers often mirror the development of com*372mon law’s concept of tort law, which has largely embraced the theory of a claim for emotional harm. We can no more ignore this development than go back to the traditional rules of master and servant. But the reality in the railway industry is that there will hardly be a situation where one sees another family member injured while working in the railroad yard. The common law rule requiring a familial relationship is based on the same principles that bind all rules governing negligent infliction of emotional distress, that is, to separate the wheat from the chaff. The theory, of course, is that when a family member is injured or killed there is real emotional trauma and grief.
Aside from the obvious certainty in the law it produces, the familial proximity element is of little value as a test to determine valid FELA claims. That element is tied to two policy considerations. First, common law courts must set forth generally applicable rules of law that must apply to society as a whole. In Corso v. Merrill, 119 N.H. 647, 406 A.2d 300 (1979), the New Hampshire Supreme Court recognized the bystander cause of action when a mother witnessed her child injured. The dissent articulated the concerns all common law courts share:
Past experience teaches me that this is not the end. This case may be limited to parent and child, but what about the child-parent case, or that of the husband and wife, brother and sister, grandparent and grandchild, the fiancee or close friend? There is just no way a logical distinction can be drawn and liability to these people will be thrust upon us in years to come. Love and affection is not necessarily confined to parenthood or even blood relationship.
406 A.2d at 308. Thus courts are wary of judicially expanding the bystander cause of action into the “fantastic realm of infinite liability.” Amaya v. Home Ice, Fuel & Supply Co., 59 Cal.2d 295, 315, 29 Cal.Rptr. 33, 379 P.2d 513 (1963), overruled, Dillon v. Legg, 69 Cal.Rptr. 72, 441 P.2d 912 (1968).
Second, state courts have been wary of the public policy ramifications from recognizing certain relationships for purposes of tort law. For example, in Elden v. Sheldon, 46 Cal.3d 267, 250 Cal.Rptr. 254, 758 P.2d 582 (1988) the California Supreme Court refused to apply bystander liability for unmarried cohabitants because “the state has a strong interest in the marriage relationship; to the extent unmarried cohabitants are granted the same rights as married persons, the state’s interest in promoting marriage is inhibited.” 250 Cal.Rptr. at 258, 758 P.2d at 586. One can easily see the public policy quagmire states would be bogged in if they recognize other less traditional relationships for the purposes of common law torts.
The insistence that the victim and plaintiff stand in a familial relationship affixes the cause of action at an arbitrary point on a continuum of relationships and is inconsistent with FELA policies. Common experience tells us that we do not grieve for only those in our immediate family. It is true we are likely traumatized whenever a family member is killed or severely injured; but the “same would often be true of very close friends.” Id. By trying to exclude the frauds and opportunists, this common law theory excludes the truly injured as well.
Moreover, the policy concerns that drive the rule do not apply to FELA. There is little chance that federal courts will be enmeshed in state public policy implications by recognizing certain relationships when considering whether the plaintiff and victim railroad coworkers were closely related. In FELA, Congress, rather than the courts, created the cause of action and allocated the economic and social burdens. See supra Opinion 366-68. In discerning the congressional intent, the Supreme Court has repeatedly said that recovery should be liberally granted under the FELA policies. Recovery will not exponentially expand to the “fantastic realm of infinite liability” and place an intolerable burden upon society because the cause of action is limited to FELA cases.
The relational proximity element is “indisputably arbitrary since it is foreseeable *373that in some cases unrelated persons have a relationship to the victim or are so affected by the traumatic event that they suffer equivalent emotional distress.” Thing, 257 Cal.Rptr. at 879, 771 P.2d at 828. In FELA cases, more so than common law negligence actions, we must curtail arbitrary line drawing so that the remedial and humanitarian goals of the statute can be fully implemented. To the extent doctrinal requirements of the common law encroach upon FELA policies and preclude remedies in meritorious cases, the common law must bow. We have full confidence that, with rigorous judicial attention to the facts to assure that the circumstances are sufficiently clear to confirm the genuineness of the claim, the court and the jury, drawing upon their combined human experiences in the course of a trial, will distinguish between victims and fakes and between close relationships and mere acquaintances.
Gottshall was not only a bystander closely related to the victim, but was also an active participant. The factual scenario akin to the one in Althoff is sufficient to bring a FELA action. Outten, 928 F.2d at 78-79. There, Althoff was operating a crane when the crane released an object that fell and killed the victim. He did not witness the injury nor was he within the zone of danger, but he did see the victim immediately after the accident. The district court concluded “plaintiff was not a passive bystander, but an active 'participant in the events giving rise to the accident: plaintiff was the operator of the device which turned out to be the prime mover of the events which culminated in Mr. Lauver’s destruction.” (Emphasis added.) Because an emotional injury is reasonably foreseeable in these kinds of circumstances, they provides the necessary assurance to present a jury question. Id. at 79.
Here Gottshall was more than a mere witness at the scene. While not a cause, he was an active participant in the events leading to Johns’ death. He was among the first to rush to Johns’ assistance. He gave mouth-to-mouth resuscitation and CPR to Johns for a full forty minutes. He carried the body of his friend to the ambulance parked two thousand feet away. The sense of loss and helplessness as one’s friend dies beneath one’s breath and the resulting emotional injury that he would suffer is easily foreseeable. Gottshall recalled, “Watching my close friend die, having him die in my own hands during CPR as a result of the extreme working conditions that Conrail imposed upon us, was unlike anything I had ever experienced.” One worker observed, “it really hit him hard. With him knowing CPR and he couldn’t bring him back, and I guess Dick was a good friend of his.” The next day the same worker observed Gottshall was “[l]ike more or less upset with himself about it.” Even Norvick, the Conrail supervisor, acknowledged, “Well, he took it hard like all of us. I mean, Mr. Gottshall was performing CPR on him and giving his breath into Johns. And when you don’t revive somebody, you take it personal. I mean, I know I would.”
Finally, the Buell Court hinted that physical symptoms of severe psychological illness are a factor. 480 U.S. at 569-71 & n. 22, 107 S.Ct. at 1418 & n. 22. The Court there confronted the issue of whether a claim for purely emotional injuries is cognizable. In declining to decide this issue, the ’ Court noted that once the facts are fleshed out the case might not present this issue of pure emotional injury as the plaintiff “claimed to have suffered physical symptoms in addition to his severe psychological illness.” Id. Indeed it is this issue that has confounded, divided and silenced federal courts. See Stoklosa, 864 F.2d at 426 (declining to address the issue); Adams, 899 F.2d at 539 (same); Elliott, 910 F.2d at 1229 (same); Robert, 832 F.2d at 7 (same); Gaston, 866 F.2d at 821 (same); Plaisance, 966 F.2d at 169 (same). But see Taylor, 787 F.2d at 1313 (purely emotional injuries cognizable); Plaisance, 937 F.2d at 1009 (same).
This case likewise does not involve a claim for purely emotional injuries. At oral argument Gottshall’s counsel challenged Conrail to argue that Gottshall’s claim was fraudulent. Conrail wisely declined the challenge. Its reticence is understandable considering Gottshall has satis*374fied the common law’s “physical manifestation” test. He lost forty pounds. See Wilson, 939 F.2d at 265 (weight loss is physical manifestation of emotional injury). He was institutionalized for three weeks and thereafter received continuing out-patient treatment. See Restatement § 436 A & Comment c (emotional distress may be com-pensable if accompanied by “bodily harm or other compensable damage”; long continued mental disturbance may be classified as a “physical illness, which is bodily harm”). He was diagnosed by three doctors as suffering from major depression and post-traumatic stress disorder. He had suicidal preoccupations, anxiety, sleep onset insomnia, cold sweats, loss of appetite, nausea, physical weakness, repetitive nightmares and a fear of leaving home. See Towns v. Anderson, 195 Colo. 517, 579 P.2d 1163, 1164 (1978) (en banc) (nightmares, sleepwalking, nervousness and irritability showed sufficient physical manifestation resulting from emotional distress); Savard v. Cody Chevrolet, Inc., 126 Vt. 405, 234 A.2d 656, 657, 660 (1967) (nervous shock, sleeplessness, loss of appetite resulting in weight loss, faintness and trembling showed sufficient physical manifestation resulting from emotional distress); Daley v. LaCroix, 384 Mich. 4, 12-13, 179 N.W.2d 390, 395-96 (1970) (weight loss, inability to perform household duties, extreme nervousness and irritability are facts from which a jury could find a compensable physical injury); Dobelle v. National R.R. Passenger Corp., 628 F.Supp. 1518, 1522, 1526-27 (S.D.N.Y.1986) (hospitalization, subsequent out-patient treatment, major depression, crying spells, weight gain, and sleep disturbances showed sufficient physical symptoms to constitute “bodily harm”); Johnson, 372 N.Y.S.2d at 639, 334 N.E.2d at 591 (recurrent nightmare, terrifying dreams of death, difficulty in concentrating, irritability, inability to function at work properly, tenseness and anxiety, as well as psychiatric confirmation of the emotional distress showed objective manifestations of the injury). Gottshall's physical symptoms objectively confirm what he has been alleging and what three doctors diagnosed.7 These symptoms are then evidence of the degree of mental distress suffered. Leong, 55 Haw. at 403, 520 P.2d 758; Paugh, 451 N.E.2d at 765; Versland, 671 P.2d at 588.
After considering Gottshall’s claim under the common law physical manifestation rule, the principles underlying the Dillon bystander and Althoff active participant rules as adopted in Outten, as well as the medical diagnosis of three doctors, we believe that Gottshall has sufficiently shown that his injuries are genuine and severe to create a genuine issue for the jury to decide.
IV.
Satisfied that this claim has passed the threshold burden, we must next consider it in the context of what actually is required for FELA: breach of a duty, injury, and causation.
The concept of negligence in the context of emotional harm claims remains the same as for physical harm. Negligence is conduct that falls below the *375standard established by law for the protection of others against unreasonable risk. Restatement § 282. It is a failure to do what the reasonable person would do under like circumstances. Id. § 283. One’s conduct must be commensurate with the known danger and the consequences that might reasonably be anticipated from the negligent conduct. Urie, 337 U.S. at 178-80, 69 S.Ct. at 1029. Negligence therefore necessarily involves a foreseeable risk of injury and conduct unreasonable in proportion to the danger. Inman v. Baltimore & Ohio R.R. Co., 361 U.S. 138, 140, 80 S.Ct. 242, 243, 4 L.Ed.2d 198 (1959). See The Law of Torts, § 43, at 280.
Foreseeability was the focal point in Out-ten. We explained that the Dillon bystander rule “reduces the likelihood that fraudulent or trivial claims will be brought and ensures that the resulting injury is sufficiently foreseeable so that it is not unreasonable to hold the defendant responsible.” 928 F.2d at 78. Likewise we said the rule in Althoff is premised on the same notion, that “it is foreseeable that one who operates an instrumentality that causes another’s death would be subject to considerable emotional distress from any guilt that operator might feel.” Id. at 79. Thus “it is hardly foreseeable,” we remarked, “to the railroad that one of its employees might suffer serious psychological injuries as a result of the fear of injury from a train collision over a mile away. There is a ‘perceived unfairness of imposing heavy and disproportionate financial burdens upon a defendant, whose conduct was only negligent, for consequences which appear remote from the wrongful act.’ ” Id. (quoting The Law of Torts, § 54, at 361).
In Holliday we concluded that Holliday’s injuries, “though obviously germane to the assessment of Holliday’s condition, have no bearing on the character of Conrail’s allegedly tortious activity.” 914 F.2d at 425. Our focus on the relationship between the injury and the alleged tortious act suggests that our conclusion there rested on the unforeseeability of emotional injuries attributable to “an ordinary management decision.” Thus if an employer is liable for consequences that cannot reasonably be foreseen, even “the most attenuated claims could be advanced.” Id. In Holliday and Outten the railways did not breach a duty because the consequences could not be reasonably anticipated from their conduct, which, as we emphasized, did not subject the plaintiffs to physical danger or to witness an injury-causing accident.
This ease is not like Holliday and Out-ten. Gottshall was subjected to the same physical danger that killed Johns, and he was aware that that force could kill him too. His emotional distress was not attributable to merely the ordinary stress of the job, but was attributable to the extreme stress of witnessing and participating in a fatal accident involving one of his close friends as well as the continuing fear that he would die in the same manner.
The issue is whether Conrail’s conduct breached a duty of care. Had Conrail simply worked its men in hot weather, weather falling within the broad spectrum of conditions that constitutes ordinary working conditions, Holliday would preclude this claim. Holliday stands for the general proposition that employment in the railway industry, like employment elsewhere, entails some risks that all employees must bear and employers should not be the insurers for injuries arising out of working in ordinary conditions of the job. We there refused to extend liability for job stress because “employees in all walks of life are placed in difficult, tense job situations, sometimes for extended periods, so there was nothing unusual here.” Id. But the conditions Gottshall and his crew experienced were unusual.
Adverse weather conditions may constitute dangerous working conditions and impose upon the employer a duty to exercise a degree of care commensurate with the risks. Fort Worth & Denver City Ry. Co. v. Smith, 206 F.2d 667, 669 (5th Cir.1953); Anderson v. Elgin, Joliet & E. Ry. Co., 227 F.2d 91, 95-96 (7th Cir.1955); Britt v. Terminal R.R. Ass’n, 311 S.W.2d 130, 134 (Mo.App.1958). Indeed had Conrail sent its men to work on steel rails in the middle of an electrical storm and had an employee been killed by lightning, there *376would certainly be a jury question as to negligence. See Childs v. Rayburn, 169 Ind.App. 147, 346 N.E.2d 655, 660 (1976); Johnson v. Kosmos Portland Cement Co., 64 F.2d 193, 196 (6th Cir.1933). See also Crutchfield v. Bogle, 270 P.2d 640, 642 (Okla.1954). Here Conrail’s conduct goes far beyond simply working men in hot weather. There is a genuine issue of whether Conrail subjected its men to unreasonably dangerous working conditions and whether that conduct was unreasonable .in proportion to the risk.
The accident happened during one of the most severe heat waves in recent memory. A Conrail supervisor recalled that he had not experienced such a stretch of hot weather since 1946 or 1948. The extreme heat combined with the humidity made several workers apprehensive. They complained to Conrail supervisors that conditions were too extreme. One worker said, “They guys [sic] were saying it’s too hot out here to work. You shouldn’t be out here working.”
When asked to define “extreme weather conditions,” Dr. David C. Deubner, Chief Medical Officer for Conrail at the time of the accident and currently Assistant Vice President of Conrad’s Medical Resources Management Department, opined “it is when it’s hot, it’s humid and the sun is shining.” In such extreme weather conditions, he said, Conrail had a widely spread oral practice to vary the work rate with the severity of the weather conditions. Such precautions are necessary because “If people are doing very heavy cardiovascular labor, then it should be slowed down and have longer work breaks, more frequently longer breaks when it gets very hot and humid.... that’s why when the sun gets too high, you recommend slowing down and having more rest. So your goal in protecting people is to keep the total degree of effort within bounds.” He believed these precautions make simple “common sense.”
Dr. Deubner also said Conrail had a policy of evaluating an employee’s “safe function capacity.” Upon such evaluation, Conrail matched an employee’s physical health with a physically compatible job function. On March' 29, 1988, about four months before the accident, Johns was examined by a Conrail doctor. The medical report indicated Johns was 56 years old, was 5' Ph" tall and weighed 187 pounds. It also indicated that Johns suffered from high blood pressure, athero or arteriosclerotic cardiovascular disease, and was taking medication. Additionally, Conrail knew most of the men, many overweight, were over 50 years old and some approaching 60 years. It knew that one other worker had previously suffered a serious heart attack. Dr. Deub-ner suggested that heavy cardiovascular work can interact with a worker’s physical conditions so that, depending on those conditions, the overall risk can be enhanced.
He also believed that working in an isolated area poses a greater medical risk because an injured worker “can benefit by timely medical care.” The coroner confirmed that had Johns received prompt medical attention his chances of survival “would have significantly increased.” Access to communication was therefore crucial.
On the day of the accident, the temperature was hovering at 97 degrees; it was very humid; the sun was shining; the field was in the open; and work started at high noon. Conrail sent middle-aged, overweight men, some with serious cardiovascular health problems, to do very heavy cardiovascular labor at an extremely rapid pace of work. Although Conrail’s own medical director recommends work “be slowed down and have longer work breaks, more frequently when it gets very hot and humid,” Conrail failed to give the men any scheduled breaks. It concedes on appeal, “They were not entitled to formal breaks, but could drink water when they wished.” Conrail Br. 8. A Conrail supervisor confirmed, “That’s one thing they were allowed to do, get water.” The tracks were isolated, but Conrail failed to provide means of communication in case of an emergency.
These circumstances alone would be sufficient to conclude that a reasonable employer should have foreseen the risks and, *377at minimum, taken one or more of the “common sense” precautions suggested by Conrail’s Chief Medical Officer, such as providing communication equipment in case of an emergency, varying hours and pace of the work, incorporating periodic rest breaks and removing those employees who might be particularly susceptible to the conditions.
There is also direct evidence that Conrail actually knew the risk of danger and foresaw the tragedy that was to come. On the day of the accident, several workers complained of the weather conditions and told Norvick it was too hot to be working. Before Johns died, Conrail had frequently discussed implementing a formal policy to address extreme weather conditions such as the conditions experienced by Gottshall’s work gang. Conrail had also considered giving its men CPR training. Norvick thought CPR training was a good idea because “all of the fellows were up in age.” A training session with the American Red Cross had been set up, but Conrail ultimately canceled it “because of the liability factor.”
In Burns v. Penn Central Co., 519 F.2d 512 (2d Cir.1975), unknown assailants had been stoning the railway’s cars for months. The railway knew of these attacks but did nothing, and one day an employee was killed by a sniper’s rifle shots. The Court of Appeals for the Second Circuit reversed the district court’s grant of directed verdict for the railway based on an alleged lack of foreseeability. It concluded, “Based on the railway’s actual knowledge of stonings in the vicinity in recent months and its constructive (and indubitably actual) knowledge of the generally dangerous conditions prevailing in the neighborhood in which the fatality transpired, the jury would have acted well within its authority under the FELA by returning a verdict for Mrs. Burns.” Id. at 514-15. The factual circumstances in Burns are similar to this ease, and that court’s conclusion is correct.
Furthermore, foreseeability of harm and foreseeability of the particular injury resulting are different. In Palsgraf v. Long Island R.R. Co., 248 N.Y. 339, 162 N.E. 99 (1928), the Court held that a railroad guard could not have reasonably foreseen that his benign action, helping to board a man carrying a small package, could have resulted in an explosion injuring Helen Palsgraf who was standing far away. In another words, the railroad guard could not possibly have foreseen any harm to Palsgraf. Importantly, Palsgraf was not a FELA case. Moreover, we simply have a different case.
In Gallick v. Baltimore & Ohio R.R. Co., 372 U.S. 108, 83 S.Ct. 659, 9 L.Ed.2d 618 (1963), the worker was bitten on the leg by an insect, which was apparently attracted to a pool of stagnant water the railway left on the worksite. An infection grew and the leg was finally amputated. The railway specifically argued that the injury was not reasonably foreseeable. 372 U.S. at 116-18, 83 S.Ct. at 665. The Court disagreed and held that the precipitating event triggering the injury, the insect bite, was reasonably foreseeable. Id'. 372 U.S. at 118-20, 83 S.Ct. at 666. It then reasoned “that for a defendant to be liable for consequential damages he need not foresee the particular consequences of his negligent acts.” Id. 372 U.S. at 120, 83 S.Ct. at 667. See Restatement § 435. Thus, “[t]he test of foreseeability does not require that the negligent person should have been able to foresee the injury in the precise form in which it in fact occurred. Rather it is sufficient if the negligent person might reasonably have foreseen that aw injury might occur_” Miller v. Cincinnati, New Orleans & Texas Pacific Ry. Co., 203 F.Supp. 107, 113 (E.D.Tenn.1962) (emphasis in original), affirmed, 317 F.2d 693 (6th Cir.1963).
Like Gallick, the precipitating event triggering Gottshall’s injury was foreseeable. A genuine issue exists whether Conrail could reasonably have foreseen that its men, including Gottshall, would suffer injuries, emotional or physical, when it subjected middle-aged men, some with health problems, to the stark conditions without scheduled periodic breaks and ready access to communications equipment. If Conrail had a hand in creating the dangerous working conditions that ultimately killed Johns, *378then under the rationale we enunciated in Outten a genuine issue also exists as to whether Conrail could have reasonably foreseen that Gottshall, or any other employee, would then suffer emotional injuries from the tragic events brought about by Conrail’s negligence. 928 F.2d at 78-79.
But assuming, arguendo, that Conrail’s conduct comported with a reasonable person’s leading up to Johns’ death, its conduct after the death raises an issue of whether it breached a legal duty. The general duty of care includes a duty to provide a reasonably safe workplace. Bailey v. Central Vermont R.R., Inc., 319 U.S. 350, 352-53, 63 S.Ct. 1062, 1064, 87 L.Ed. 1444 (1943). This duty includes a duty to guard against conditions in .the workplace that cause emotional harm to employees. Buell v. Atchison, Topeka & Santa Fe Ry. Co., 771 F.2d 1320, 1323 (9th Cir.1985),’affirmed in part and vacated in part, 480 U.S. at 567-71, 107 S.Ct. at 1417-18.
Gottshall alleges that he suffered emotional injuries partly because after the accident he feared he would also die in the same manner. After Johns died, his corpse was laid on the ballast beside the track. With work to be done, Conrail ordered its men back to work to do the very thing that felled Johns. As one Conrail supervisor explained, “After everything was taken care of, they had to go back and finish the railroad, because we had to give the railroad back.” During this time Johns’ corpse was in full view of the men, including Gottshall. Even after they completed the work, Conrail ordered the men to stay at the worksite, though nothing required them to stay. Gottshall was forced to view the corpse for a full three hours. The very next day, Conrail did nothing to remedy the extreme physical conditions. The heat wave persisted. Conrail pushed its men even harder and still discouraged them from taking breaks, with water the only source of relief from the severe conditions. They were told, “Don’t come in until you are done.” Under these conditions the men worked a full shift and three or four hours of overtime, and when they finished, they were told, “Is that all you did today.” But for the Grace of God, thought Gottshall, the physical force Conrail subjected its men could have killed him as well.
Ordinary stress of the workplace does not create an unsafe working environment. It and a variety of working conditions comprise the negotiated circumstances of the employees’ duties. There is, however, evidence that conditions that day were anything but “ordinary,” much less characterized as a “negotiated circumstance.” Indeed, they were extreme. Gottshall swore, “Watching my close friend die, having him die in my own hands during CPR as a result of the extreme working conditions that Conrail imposed upon us, was unlike anything I had ever experienced' either on the railroad, or otherwise, and it deeply, deeply, disturbed me:” Gottshall had worked for Conrail for fourteen years and never had psychological or emotional problems. Another worker swore in an affidavit, “The conditions at the Watson-town Secondary on the date of the accident were substantially more severe than the general working conditions associated with the job. The extreme heat and humidity, and the pace of the work, were unbearable.” Another worker swore, “Working on a division gang can be strenuous. However, the situation on the date of Mr. Johns’ death, involving extreme heat, no breaks, no shade and being pushed very hard, was something different.” Gottshall recalled, “It was 97 degrees. It was hot and humid. They put us out there to change rails, push, push, push_ I’m not an animal.”
We cannot casually assume work under these extreme conditions to be a “negotiated circumstance.” The combination of the conditions created not only physical hazards, but constituted emotional hazards which can equally debilitate and scar an employee, particularly one who had just witnessed a friend die under the same conditions. Additionally, Conrail wholly created the circumstances where Gottshall had to watch his friend’s body “bake like a potato” while doing the very thing that killed him under the same hostile conditions.
*379FELA “is not to be narrowed by refined reasoning or for the sake of giving ‘negligence’ a technically restricted meaning. It is to be construed liberally to fulfill the purposes for which it was enacted, and to that end the word may be read to include all the meanings given to it by the courts, and within the word as ordinarily used.” Jamison, 281 U.S. at 640, 50 S.Ct. at 442 (citations omitted). In the totality of the terrible circumstances surrounding this case, we are unwilling to say that Conrail was not negligent as a matter of law. “[F]oreseeability of harm is no less a matter generally left to the jury’s broad decision than any other part of the requisite proof to recover under the FELA.” Burns, 519 F.2d at 514. We have no trouble concluding that the facts, as presented to us and in light most favorable to Gotts-hall, present a genuine issue of material fact whether Conrail breached a general duty of acting with care and a duty to provide a safe workplace.
We likewise have no trouble concluding that a genuine issue exists whether Gotts-hall sustained cognizable injury as a result of emotional trauma. Gottshall was institutionalized for about three weeks; three doctors diagnosed him as suffering from severe post-traumatic stress syndrome; and he exhibited objective, physical symptoms of his emotional injuries. See supra Opinion 373-74. We also have no trouble concluding that a genuine issue exists as to whether Conrail’s negligence “played any part, however small, in the injury or death which is the subject of the suit.” Rogers, 352 U.S. at 508, 77 S.Ct. at 449. The law of this circuit is that a court may take a case from jury consideration on causation grounds “only in those extremely rare instances where there is zero probability either of employer negligence or that any such negligence contributed to the injury of an employee.” Pehowic v. Erie Lackawanna R.R., 430 F.2d 697, 699-700 (3d Cir.1970).
Because a railway employer’s duty is high and FELA provides for a most lenient causation inquiry, a court must be cautious in granting summary judgment. “It is well established that the role of the jury is significantly greater in Jones Act and FELA cases than in common law negligence actions. The right of the jury to pass upon the question of fault and causation must be most liberally viewed.” Johannessen v. Gulf Trading & Transp. Co., 633 F.2d 653, 656 (2d Cir.1980). See Green v. River Terminal Ry. Co., 763 F.2d 805, 807 (6th Cir.1985); Eggert v. Norfolk & W. Ry. Co., 538 F.2d 509, 511 (2d Cir.1976); Boeing Co. v. Shipman, 411 F.2d 365, 370-73 (5th Cir.1969) (en banc); Pierce, 823 F.2d at 1370 (9th Cir.) (citing Ybarra v. Burlington N., Inc., 689 F.2d 147, 149 (8th Cir.1982)). “By enacting FELA, Congress desired to ‘secure jury determinations in a larger proportion of cases than would be true of ordinary common law actions.’ ” Hines v. Consolidated Rail Corp., 926 F.2d 262, 269 (3d Cir.1991) (quoting Shipman, 411 F.2d at 371). Indeed under FELA “trial by jury is part of the remedy.” Atlantic & Gulf Stevedores, Inc. v. Ellerman Lines, Ltd., 369 U.S. 355, 360, 82 S.Ct. 780, 783, 7 L.Ed.2d 798 (1962). Viewing all the facts in the light most favorable to Gottshall, we are confident that his claim has sufficient indicia of genuineness and that Gottshall has satisfied the requisite elements for pleading a FELA action to entitle him to trial on the merits.
V.
In so concluding, we are aware, and emphasize, that even under the liberal recovery policy under FELA, the employer is not the insurer of the worker’s safety. As in the common law, there must be some finite limit to the railway’s potential liability. See Thing, 257 Cal.Rptr. at 878, 771 P.2d at 827 (to delimit liability “policy considerations justify restrictions on recovery for emotional distress notwithstanding the sometimes arbitrary result”); Elden, 250 Cal.Rptr. at 258, 758 P.2d at 586 (“social policy must at some point intervene to delimit liability”). The Court of Appeals for the Fifth Circuit explained that one reason behind the common law's restrictive tests is to avoid
*380multiplying damages incalculably on the basis of factors which are entirely incidental to the operation of the train or vessel and which bear little or no relationship to the safety of that venture— factors such as how many crew members happen to be standing around to observe an incident, .whether their vision is obscured by darkness or weather, how close they are to the occurrence and the like.
Gaston, 866 F.2d at 820. Even when a worker’s injuries are real and demonstrable, the Gaston court believed that exposing employers to such potentially ruinous liability “involves evaluating economic trade-offs which we (or any other court) are poorly equipped even to envision, let alone to weigh against each other.” Id.
We agree insofar as strong policy reasons preclude finding liability for “every possible distressful happening to which a railroad worker is exposed.” Outten, 928 F.2d at 79. See Holliday, 914 F.2d at 425. This policy simply echoes the belief that frivolous claims for trivial emotional disturbances and bruised feelings that are ordinary and expected as a part of daily life, such as stress from the ordinary conditions of the workplace, are not cognizable. To this end, the proper limits on liability are those fixed by generally applicable principles of tort law. Recognizing the reality that every act or omission can affect the procession of human events, the law through the years has considered the proper balance between remedying the injured individual and exposing every member of society to the myriad of possible consequences of living in a complex world. In the case of physical injuries, the law has dealt with this problem through concepts such as duty, foreseeability, and proximate cause. Given that the employer owes a duty to provide a safe workplace, once the employer breaches that duty through an act or omission creating a foreseeable risk, that fortuity determines the number of persons actually injured or the extent of those injuries does not prevent recovery.
A simple example will serve. Imagine that the employer negligently caused a railroad accident in which fifty workers were killed or injured. That the employer’s liability depends, to some extent, on factors that it had no way of controlling is irrelevant for the purpose of imposing liability. True, the aggregate liability on the negligent employer in this case would be severe. But where the risk was reasonably foreseeable, where the danger was disproportionate to that risk, and “where the concern is to avoid imposing excessive punishment upon a negligent defendant, it must be asked whether fairness will permit leaving the burden of loss instead upon the innocent victim.” The Law of Torts, § 54 at 361. When a claim for emotional injuries shows a sufficient indicia of genuineness so as to create a factual dispute, the same tort principles apply to the .employer’s liability.
The federal courts bear the duty of determining this liability and fashioning remedies. By placing the onus of performing this task on the federal courts, Congress has dictated that we evaluate the economic trade-offs and properly balance the competing policies. Congress has, however, given us some guidance in how to strike this balance. It has provided that recovery be liberally granted and that the railway industry, as a single economically sufficient, enterprise, bear the social and economic burdens of workers who give their labor to further the common endeavor and who have suffered as a result of their employers’ negligence. Congress promulgated FELA to promote “the welfare of both employer and employee, by adjusting the losses and injuries inseparable from industry and commerce to the strength of those who in the nature of the case ought to share the burden.” Sinkler, 356 U.S. at 330, 78 S.Ct. at 762 (quoting Sen.Rep. No. 460, 60th Cong., 1st Sess. 3). See supra Opinion 366-68. Where the burden becomes heavy on a railway, it is only because in seeking to achieve the ends of the enterprise the railway has placed an equally heavy burden on its workers, a burden paid in lost lives, limbs, and livelihoods. When an employer breaches a duty of care to provide a safe workplace, it is liable *381under FELA if an employee suffers an injury, physical or emotional, causally related to its negligence.
We are also aware of the Kraus court’s admonitions about incalculable and potentially unlimited damages, floods of litigation, and fraud. See Holliday, 914 F.2d at 424; The Law of Torts, § 54, at 360-61; Restatement § 436A comment b. But we must also be aware that these concerns cannot become the incantations by which railway attorneys make all plaintiffs claims disappear. Dean Prosser has observed:
The same objections against allowing recovery have been advanced here as well: that mental disturbance cannot be measured in terms of money, and so cannot serve in itself as a basis for the action; ... that there is a lack of precedent, and that a vast increase in litigation would follow. All these objections have been answered many times, and it is threshing old straw to deal with them. Mental suffering is no more difficult to estimate in financial terms, and no less a real injury, than “physical” pain; it is not an independent intervening cause, but a thing brought about by the defendant’s negligence itself, and its consequences may follow in unbroken sequence from that negligence; and while it may be true that its consequences are seldom very serious unless there is some predisposing physical condition, the law is not for the protection of the physically sound alone. It is the business of the courts to make precedent where a wrong calls for redress, even if lawsuits must be multiplied; and there has long been precedent enough, and no great increase in litigation has been observed.
The Law of Torts, § 54, at 360. See Corso, 406 A.2d at 306 (where rule allows for more liberal recovery “the adoption of well-defined foreseeability factors will not lead to unlimited liability, and that the threat of remote and unexpected liability is not a substantial fear”) (cited by Buell, 480 U.S. at 569-71 n. 21, 107 S.Ct. at 1418 n. 21); Ramirez v. Armstrong, 100 N.M. 538, 542, 673 P.2d 822, 826 (1983) (same); Schultz v. Barberton Glass Co., 4 Ohio St.3d 131, 133, 447 N.E.2d 109, 111 (1983) (expressing doubts that problem of increasing litigation is real, and if real, would be unacceptable reason for denying recovery in meritorious cases); Toms, 207 N.W.2d at 145 (rejecting the “opening the flood gates of litigation” argument as meritless).8
*382Our research confirms the observation that litigation has not increased by allowing claims for emotional distress. In the over five years since the Buell Court “opened the door,” the courts of appeals reported seventeen cases involving a claim for emotional distress. See supra Opinion 362-66. This appeal makes eighteen. This number in a span of five years can hardly be called a “flood” of litigation. If there was ever a time when we expected the flood of litigation, it was when the door was first opened. The apocalypse argued by the railway has not come to pass.
In Buell the railway contended that if emotional injuries are cognizable under FELA, virtually no employees will pursue grievances through the RLA since “every employee who believes he has a legitimate grievance will doubtless have some emotional anguish.” 480 U.S. at 567 n. 13, 107 S.Ct. at 1416 n. 13 (quoting from the brief for the railway). One amicus asserted that a large portion of the 183,800 grievances filed in 1985 would be pursued as FELA actions instead, thus creating the “potential for doubling the volume of civil filings in the federal courts.” Id. (quoting from ami-cus brief).
The Supreme Court unanimously and summarily rejected this “parade of horri-bles,” exposing the strained assumption on which the railway relies, that is, that a significant percentage of employees bringing grievances suffer the type of genuine and severe emotional injury that has generally been required to establish liability for purely emotional injury. Id. See Moody, 823 F.2d at 694 (recognizing Buell substantially rejected the “flood of litigation” argument).
The concerns articulated in Kraus must be balanced with the notion that “a flood of litigation” and fraudulent claims may serve judicial prudence, but it can never serve judicial convenience. “Although fraud, extra litigation and a measure of speculation are, of course, possibilities, it is no reason for a court to eschew a measure of its jurisdiction. ‘The argument from mere expediency cannot commend itself to a Court of justice, resulting in the denial of a logical legal right and remedy in all cases because in some a fictitious injury may be urged as a real one.’ ” Battalla, 219 N.Y.S.2d at 37, 176 N.E.2d at 731 (emphasis in original) (quoting Green v. T.A. Shoemaker & Co., 111 Md. 69, 81, 73 A. 688, 692 (1909)).
VI.
Before we conclude, it is important that we spell out what we are not holding. We are not holding that a railroad breaches a duty of care when it works its men in hot weather. We are not holding that an employer must refrain from pushing its men to do hard labor. We are not holding that every employee who suffers emotional injuries as a result of the railway employer’s conduct has a valid FELA claim. We are not holding that every bystander to an accident has a valid FELA claim. We are not even holding that every bystander who sees a close friend injured or killed has a valid FELA claim.
We simply hold that considering the totality of these extreme facts in the light most favorable to Gottshall — that is, seeing and actively participating in the events surrounding the death of a friend of fifteen years as a result of Conrail’s actions, which includes working men under extreme conditions without taking the precautions arguably required, being forced to stay at the worksite by Conrail for several hours, being forced to be subjected to the same physical force after the accident without the benefits of the precautions arguably required, checking into a mental hospital for three weeks, losing forty pounds, exhibiting other physical manifestations of severe emotional injury, being diagnosed by *383three doctors as suffering from severe post-traumatic stress syndrome — Gotts-hall’s claim has sufficient indicia of genuineness and he presents genuine issues of whether there is a breach of a duty, injury and causation. He is entitled under FELA to a trial on the merits. We will reverse and remand.
. See Comment, Negligent Infliction of Mental Distress: A Jurisdictional Survey of Existing Limitations Devices and Proposal Based on an Analysis of Objective Versus Subjective Indices of Distress, 33 Vill.L.Rev. 781, 792-93 n. 59 (1988) (as of 1988 five states plus the District of Columbia have adopted this rule).
. See for example, District of Columbia v. Smith, 436 A.2d 1294 (D.C.1981) (negligent failure to inform next of kin about death of loved one is noncompensable absent physical impact); Woodman v. Dever, 367 So.2d 1061 (Fla.Dist.Ct.App.1979) (a child who witnessed robbery and sexual assault of mother was denied recovery because there was no physical impact on the child); James v. Harris, 729 P.2d 986 (Colo.Ct.App.1986) (parents who witnessed their child being run over by a car could not recover because they were not within the zone of danger).
. We believe that the logic in Adams is flawed. By requiring "unconscionable abuse” as a prerequisite for recovery for purely emotional inju*363ry, the court relied on Netto v. Amtrak, 863 F.2d 1210 (5th Cir.1989) and footnote 13 in Buell. But Netto involved a claim for intentional infliction of emotional distress, 863 F.2d at 1214 n. 4, while the claim in Adams involved negligent infliction of emotional distress. This distinction is critical. Footnote 13 in Buell addressed the railway's contention that if claims for emotional injuries are cognizable under FELA, no employee will pursue grievance through the Railway Labor Act since every aggrieved worker will undoubtedly have suffered some degree of emotional anguish. The Court responded:
This parade of horribles mistakenly assumes that a significant percentage of employees bringing grievances suffer the type of severe emotional injury that has generally been required to establish liability for purely emotional injury, and that a significant percentage of employees are subject to the type of unconscionable abuse which is prerequisite to recovery. In Farmer v. Carpenters, 430 U.S. 290, 97 S.Ct. 1056, 51 L.Ed.2d 338 (1977), we held that a state action for intentional infliction of emotional distress was not pre-empted by the National Labor Relations Act (NLRA), and pointed out that the risk of interference with the NLRA was minimized by the fact that state law permitted "recovery only for emotional distress sustained as a result of ‘outrageous' conduct.”
480 U.S. at 567 n. 13, 107 S.Ct. at 1416 n. 13 (emphasis added). The Court here was discussing a claim for intentional infliction of emotional distress. Such claims require unconscionable or outrageous action. See William Prosser, The Law of Torts, § 12, at 60-65 (5th ed. 1984); Restatement § 46; Womack v. Eldridge, 215 Va. 338, 342, 210 S.E.2d 145, 148 (1974); Fischer v. Maloney, 43 N.Y.2d 553, 557, 402 N.Y.S.2d 991, 992-93, 373 N.E.2d 1215, 1217 (1978). Indeed the Court proceeded to distinguish claims for negligent and intentional infliction of emotional distress, concluding:
[Wjhether one can recover for emotional injury might rest on a variety of subtle and intricate distinctions related to the nature of the injury and the character of the tortious activity.... In short, the question whether one can recover for emotional injury may not be susceptible to an all-inclusive "yes” or “no” answer. As in other areas of law, broad pronouncements in this area may have to bow to the precise application of developing legal principles to the particular facts at hand.
480 U.S. at 569-70, 107 S.Ct. at 1417-18. Thus the Court did not intend to make unconscionable conduct a prerequisite for bringing a claim for emotional distress.
We fail to see why unconscionable conduct is required to bring a claim for negligent infliction of emotional distress. Even common law does not require it. See Brown v. Cadillac Motor Car Div., 468 So.2d 903, 904 (Fla.1985); Rickey v. Chicago Transit Auth., 98 Ill.2d 546, 550, 75 Ill.Dec. 211, 457 N.E.2d 1 (1983). Indeed, requiring outrageous conduct is logically inconsistent with the statute. The language of FELA refers to negligence only. The statutory language, however, has been extended to include liability for some intentional torts. See Jamison v. Encamacion, 281 U.S. 635, 641, 50 S.Ct. 440, 442, 74 L.Ed. 1082 (1930). In this context, we find it odd that coverage is provided in an intentional tort, yet excluded when the injury arises out of negligence, specifically referred to in the statute, just because the plaintiff suffered an emotional injury.
. The Jones Act incorporates FELA. 46 U.S.C. § 688; Ferguson v. Moore-McCormack Lines, Inc., 352 U.S. 521, 523, 77 S.Ct. 457, 458, 1 L.Ed.2d 511 (1957).
. In view of these conflicts, Justices White and Thomas believe that "a uniform rule should be announced by this Court on this important and recurring issue.” Ray v. Consolidated Rail Corp., — U.S. -, -, 112 S.Ct. 914, 914, 116 L.Ed.2d 813 (1992) (dissenting from a denial of certiorari); Carroll v. Consolidated Rail Corp., — U.S. -, -, 112 S.Ct. 916, 916, 116 L.Ed.2d 816 (1992) (same).
The jurisprudence among the district courts is no more consistent. See for example, Kiffe v. Neches-Gulf Marine, Inc., 709 F.Supp. 743, 745 (E.D.Tex.1989) (worker who witnessed death of coworker and who suffered purely emotional injuries failed to state a claim); Welby v. Consolidated Rail Corp., 671 F.Supp. 1015, 1021 (M.D.Pa.1987) (heart attack suffered as a result of improper and unsafe working conditions was actionable); Gillman v. Burlington N.R.R. Co., 673 F.Supp. 913, 917 (N.D.Ill.1987) (emotional distress claim is actionable only if it satisfies the zone of physical danger test), affirmed, 878 F.2d 1020 (7th Cir.1989); Teague v. National R.R. Passenger Corp., 708 F.Supp. 1344, 1350 (D.Mass.1989) ("[Tjhis Court therefore rejects the Gillman approach to the extent that it permits recovery only if the 'zone of danger’ test is satisfied. In addition to permitting recovery under such circumstances, the Court also adopts the common law majority rule that actual or threatened impact is unnecessary to a negligent infliction of claim under FELA and that objective symptomatology of the mental distress will suffice.”) (emphasis in original); Halko v. New Jersey Transit Rail Operations, Inc., 677 F.Supp. 135, 144 (S.D.N.Y.1987) (suicide as a result of harassment is actionable); Amendola v. Kansas City S. Ry. Co., 699 F.Supp. 1401, 1411 (W.D.Mo.1988) (mental anguish resulting from fear of contracting asbestos-related diseases is not actionable absent manifestations of physical injury); Masiello v. Metro-North Commuter R.R., 748 F.Supp. 199, 205 (S.D.N.Y.1990) (emotional distress with an accompanying physical manifestation is actionable); Smolsky v. Consolidated Rail Corp., 780 F.Supp. 283, 291 (E.D.Pa. 1991) (harassment resulting in purely emotional injuries is actionable).
. See Urie, 337 U.S. at 174, 69 S.Ct. at 1027 ("What constitutes negligence for the statute’s purpose is a federal question, not varying in accordance with the differing conceptions of negligence applicable under state and local laws for other purposes.”); id. 337 U.S. at 196, 69 S.Ct. at 1038 (Frankfurter concurring) ("At the risk of wearisome reiteration it is relevant to say again that the common-law concept of negligence is an antiquated and uncivilized basis for working out rights and duties for disabilities and deaths inevitably due to the conduct of modern industry.”); Sinkler, 356 U.S. at 329, 78 S.Ct. at 762 ("However, in interpreting the FELA, we need not depend upon common-law principles of liability. This statute [is] an avowed departure from the rules of the common law_”); Kernan, 355 U.S. at 438, 78 S.Ct. at 401 ("[Courts have] plainly rejected many of the refined distinctions necessary in common-law tort doctrine_"); Dice v. Akron, Canton & Youngstown R.R. Co., 342 U.S. 359, 361, 72 S.Ct. 312, 314, 96 L.Ed. 398 (1952) ("State laws are not controlling in determining what the incidents of this federal right shall be.").
. "Psychiatric/psychological examining techniques have increased markedly in sophistication since the advent of the physical impact, zone of danger and ensuing physical injury rules. As a result, successfully feigning a psychic injury is not an easy matter.” Bell, The Bell Tolls: Towards Full Tort Recovery for Psychic Injury, 36 U.Fla.L.Rev. 333, 351 (1984) (cited by Buell, 488 U.S. at 569-71 n. 20, 107 S.Ct. at 1418 n. 20). “The existence and extent of mental distress is no longer generally viewed as so substantially different from, and easier to feign than, physical injury as to require qualitatively different rules of recovery. This change in perception reflects, in part, an increasing judicial notice and acceptance of the advances in the medical and behavioral sciences.” Supra n. 1, Comment, 33 Vill.L.Rev. at 789. “In judging the genuineness of a claim of mental distress, courts and juries may look to 'the quality and genuineness of proof and rely to an extent on the contemporary sophistication of the medical profession and the ability of the court and jury to weed out dishonest claims.’" Rodrigues v. State, 52 Haw. 156, 172, 472 P.2d 509, 519-20 (1970). Accord Champion, 420 So.2d at 350; Niederman v. Brodsky, 436 Pa. 401, 261 A.2d 84, 86 & n. 3 (1970); Sinn v. Burd, 486 Pa. 146, 404 A.2d 672, 678-79 (1979); Battalia v. State, 10 N.Y.2d 237, 219 N.Y.S.2d 34, 37-38, 176 N.E.2d 729, 731-32 (1961); Towns, 579 P.2d at 1164; Versland, 671 P.2d at 588.
. Indeed, as medical science continued to make strides and as the initial fear of inability to objectively and definitively verify these claims began to subside, the common law trend shows a greater receptivity towards these claims. Pennsylvania serves as an example.
Before 1970, Pennsylvania acknowledged only the physical impact rule of liability. See for example, Knaub v. Gotwalt, 422 Pa. 267, 270, 220 A.2d 646, 647 (1966). The cases applied this rule with “obstinate rigidity in that recovery was denied not only when the complaining party was a nearby witness, but also to the actual victim of the tortfeasor’s negligence or frightening conduct." Sinn, 404 A.2d at 675 (citing Bosley v. Andrews, 393 Pa. 161, 142 A.2d 263 (1958)).
But in 1970, the Pennsylvania Supreme Court abandoned the physical impact test and adopted the more flexible zone of danger test, believing it was compelled by the "inherent humanitarianism of our judicial process.” Niederman, 261 A.2d at 85. In so doing, it rejected the arguments that damages are incalculable, that the possibility of recovery in such claims will encourage fraud and frivolity, and that courts will be deluged with an avalanche of cases. Id. at 86-89.
With experience, the Pennsylvania courts realized that even the zone of danger test can be too restrictive and arbitrary. In 1979, the Pennsylvania Supreme Court remedied the situation by adopting the bystander rule of liability, believing that the deficiencies in the zone of danger test are “glaringly apparent.” Sinn, 404 A.2d at 686. In so doing, it again rejected the argument that such a rule will lead to incalculable damages, fraud and frivolity, and a flood of litigation. Id. at 678-85. The court concluded, "It is clear that appellant’s injuries were of a nature reasonably foreseeable under the circumstances alleged.” Id. at 686. Thus a review of Pennsylvania law shows that Pennsylvania has reversed course, from applying almost case-barring restrictions on these types of claims towards applying generally applicable principles of tort law.
Other jurisdictions have followed similar trends. See for example, Champion, 420 So.2d at 352 (“Thus, we believe traditional principles of negligence analysis may be applied in such cases and the imposition of undue liability may be avoided by using the broader test of reasonable foreseeability.”); Leong, 55 Haw. at 403, 520 P.2d 758 (“A survey of cases from other jurisdictions does not show the development of *382a logical and consistent rule, but reveals that the trend is a hesitant abandonment of such artificial restrictions and barriers to recovery in favor of a greater reliance on general tort law principles and the contemporary sophistication of the medical profession to test the veracity of claims for relief."); Hunsley, 553 P.2d at 1102 (“Rather than add to the already existing confusion with the formulation of a new rule, we conclude that the wisest approach is to return to the traditional principles, theories, and standards of tort law.”).