delivered the opinion of the Court.

These cases require us to determine the proper standard for evaluating claims for negligent infliction of emotional distress that are brought under the Federal Employers’ Liability Act. Because the standard adopted by the Court of Appeals is inconsistent with the principles embodied in the statute and with relevant common-law doctrine, we reverse the judgments below.

I

Respondents James Gottshall and Alan Carlisle each brought suit under the Federal Employers’ Liability Act (FELA), 35 Stat. 65, as amended, 45 U. S. C. §§ 51-60, against their former employer, petitioner Consolidated Rail Corporation (Conrail). We set forth the facts of each case in turn.

A

Gottshall was a member of a Conrail work crew assigned to replace a stretch of defective track on an extremely hot and humid day. The crew was under time pressure, and so the men were discouraged from taking scheduled breaks. *536They were, however, allowed to obtain water as needed. Two and one-half hours into the job, a worker named Richard Johns, a longtime friend of Gottshall, collapsed. Gottshall and several others rushed to help Johns, who was pale and sweating profusely. They were able to revive him by administering a cold compress. Michael Norvick, the crew supervisor, then ordered the men to stop assisting Johns and to return to work. Five minutes later, Gottshall again went to Johns’ aid after seeing his friend stand up and collapse. Realizing that Johns was having a heart attack, Gottshall began cardiopulmonary resuscitation. He continued the process for 40 minutes.

Meanwhile, Norvick attempted to summon assistance, but found that his radio was inoperative; unbeknownst to him, Conrail had temporarily taken the nearest base station off the air for repairs. Norvick drove off to get help, but by the time he returned with paramedics, Johns had died. The paramedics covered the body with a sheet, ordered that it remain undisturbed until the coroner could examine it, and directed the crew not to leave until the coroner had arrived. Norvick ordered the men back to work, within sight of Johns’ covered body. The coroner, who arrived several hours later, reported that Johns had died from a heart attack brought on by the combined factors of heat, humidity, and heavy exertion.

The entire experience left Gottshall extremely agitated and distraught. Over the next several days, during which he continued to work in hot and humid weather conditions, Gottshall began to feel ill. He became preoccupied with the events surrounding Johns’ death, and worried that he would die under similar circumstances. Shortly after Johns’ funeral, Gottshall was admitted to a psychiatric institution, where he was diagnosed as suffering from major depression and posttraumatic stress disorder. During the three weeks he spent at the institution, Gottshall experienced nausea, insomnia, cold sweats, and repetitive nightmares concerning *537Johns’ death. He lost a great deal of weight and suffered from suicidal preoccupations and anxiety. Gottshall has continued to receive psychological treatment since his discharge from the hospital.

Gottshall sued Conrail under FELA for negligent infliction of emotional distress. He alleged that Conrail’s negligence had created the circumstances under which he had been forced to observe and participate in the events surrounding Johns’ death. The District Court granted Conrail’s motion for summary judgment, holding that FELA did not provide a remedy for Gottshall’s emotional injuries.

A divided panel of the United States Court of Appeals for the Third Circuit reversed and remanded for trial. Gottshall v. Consolidated Rail Corp., 988 F. 2d 355 (1993). The court observed that most States recognize a common-law cause of action for negligent infliction of emotional distress, but limit recovery to certain classes of plaintiffs or categories of claims through the application of one or more tests. Id., at 361 (discussing “physical impact,” “zone of danger,” and “relative bystander” tests). The Third Circuit suggested that because “an emotional injury is easier to fake” than a physical injury, these tests have been “judicially developed to screen causes of action and send only the meritorious ones to juries.” Ibid.

The court below identified what it considered to be a fundamental tension between the restrictive attitude of the common law toward claims for negligent infliction of emotional distress on the one hand, and the general policy underlying FELA on the other. According to the Third Circuit, the common law places harsh and arbitrary limits on recovery for emotional injury, while FELA has consistently been interpreted to accord liberal relief to railroad workers injured through the negligence of their employers. Id., at 367-368 (discussing cases).

In the Third Circuit’s view, the only way to reconcile the apparent tension was to give preference to the liberal recov*538ery policy embodied in FELA over the common law: “[D]octrinal common law distinctions are to be discarded when they bar recovery on meritorious FELA claims.” Id., at 369. Determining that judges could weed out fraudulent emotional injury claims through careful scrutiny of the facts, the court held that the facts alleged in support of a claim under FELA for negligent infliction of emotional distress must “provide a threshold assurance that there is a likelihood of genuine and serious emotional injury.” Id., at 371. The Third Circuit suggested that a court’s factual inquiry might include consideration of the plaintiff’s claim in light of the present state of the common law.

After reviewing the facts of Gottshall’s case, the Third Circuit concluded that Gottshall had made a sufficient showing that his injuries were genuine and severe. Id., at 374. Because his claim had met the court’s threshold “genuineness” test, the court next considered whether the claim adequately alleged the usual FELA elements of breach of a duty of care (that is, conduct unreasonable in the face of a foreseeable risk of harm), injury, and causation. The panel majority concluded that there were genuine issues of material fact concerning whether Gottshall’s injuries were foreseeable by Conrail, whether Conrail had acted unreasonably, and whether Conrail’s conduct had caused cognizable injury to Gottshall. The court therefore remanded for trial. Id., at 383.

Judge Roth dissented in part because she believed that there was no triable issue regarding breach of duty. She reasoned that “outside of the interruption of the communications link, the allegedly negligent conditions created by Conrail at the time of Johns’ collapse consisted in fact of the members of the work gang performing the negotiated duties of their jobs under conditions which may indeed have been difficult but which had occurred in the past and will probably occur again in the future.” Id., at 385. In her view, these *539negotiated duties could not support a finding of negligence. Judge Roth concluded that “Conrail could not reasonably have foreseen that its negligence in interrupting the work gang’s communication^] link might cause James Gottshall’s severe emotional reaction to the death of Richard Johns.” Id., at 386.

B

Respondent Carlisle began working as a train dispatcher for Conrail in 1976. In this position, he was responsible for ensuring the safe and timely movement of passengers and cargo. Aging railstock and outdated equipment made Car-lisle’s job difficult. Reductions in Conrail’s work force required Carlisle to take on additional duties and to work long hours. Carlisle and his fellow dispatchers frequently complained about safety concerns, the high level of stress in their jobs, and poor working conditions. In 1988, Carlisle became trainmaster in the South Philadelphia yards. With this promotion came added responsibilities that forced him to work erratic hours. Carlisle began to experience insomnia, headaches, depression, and weight loss. After an extended period during which he was required to work 12- to 15-hour shifts for weeks at a time, Carlisle suffered a nervous breakdown.

Carlisle sued Conrail under FELA for negligent infliction of emotional distress. He alleged that Conrail had breached its duty to provide him with a safe workplace by forcing him to work under unreasonably stressful conditions, and that this breach had resulted in foreseeable stress-related health problems. At trial, Carlisle called medical experts who testified that his breakdown and ensuing severe depression were caused at least in part by the strain of his job. The jury awarded Carlisle $386,500 in damages.

The Third Circuit affirmed, “uphold[ing] for the first time a claim under the FELA for negligent infliction of emotional distress arising from work-related stress.” Carlisle v. Con*540solidated Rail Corp., 990 F. 2d 90, 97-98 (1993). In rejecting Conrail’s argument that Carlisle had failed to make out a claim under FELA because he had not alleged any accident or physical injury or impact, the court noted that in Gottshall (decided the month before), it had “upheld recovery under the FELA for negligent infliction of emotional distress without proof of any physical impact.” 990 F. 2d, at 96. Restating its holding in Gottshall, the court advised that, when evaluating a claim under FELA for negligently inflicted emotional distress, district courts within the Third Circuit “should engage in an initial review of the factual indicia of the genuineness of a claim, taking into account broadly used common law standards, then should apply the traditional negligence elements of duty, foreseeability, breach, and causation in weighing the merits of that claim.” 990 F. 2d, at 98.

In the case before it, however, the court did not examine Carlisle’s suit in light of any of the various common-law tests for dealing with negligent infliction of emotional distress claims. Instead, it shifted its primary emphasis to the foreseeability of the alleged injury and held that “when it is reasonably foreseeable that extended exposure to dangerous and stressful working conditions will cause injury to the worker, the employer may be held to be liable under the FELA for the employee’s resulting injuries.” Id., at 97. The Third Circuit held that Carlisle had produced sufficient evidence that his injury had been foreseeable to Conrail. The court also found sufficient evidence that Conrail had breached its duty to provide Carlisle with a safe workplace by making his employment too demanding, and that this breach had caused Carlisle’s injury. Ibid.

Pursuant to this Court’s Rule 12.2, Conrail petitioned for review of the Third Circuit’s decisions in Gottshall and Car-lisle. We granted certiorari, 510 U. S. 912 (1993), to resolve a conflict among the Courts of Appeals concerning the threshold standard that must be met by plaintiffs bringing *541claims for negligent infliction of emotional distress under FELA.1

II

In these cases, we address questions left unanswered in Atchison, T & S. F. R. Co. v. Buell, 480 U. S. 557 (1987). That case involved a FELA complaint filed by a railroad car-man who alleged that the intentional and negligent actions of his employer had caused him to suffer emotional injuries. We rejected the railroad’s contention that the FELA action should be barred because the conduct complained of was subject to arbitration under the terms of the Railway Labor Act, 44 Stat. 577, as amended, 45 U. S. C. § 151 et seq. See 480 U. S., at 564-567. Because the record was not fully developed, however, we were unable to reach the railroad’s alternative argument that purely emotional injury was not compensable under FELA. Today, we must resolve one of the questions reserved in Buell: whether recovery for negligent infliction of emotional distress is available under FELA.2 If we conclude that it is, we must consider the proper scope of that availability. Our FELA jurisprudence outlines the analysis we must undertake when deciding whether, and to what extent, this new category of claims should be cognizable under the statute.

First, as in other cases involving the scope of the statute, we must look to FELA itself, its purposes and background, and the construction we have given it over the years. See, e. g., id., at 561-562. Second, because “FELA jurisprudence gleans guidance from common-law developments,” id., at 568, we must consider the common law’s treatment of the right *542of recovery asserted by respondents. See, e. g., Monessen Southwestern R. Co. v. Morgan, 486 U. S. 330, 336-339 (1988) (disallowing prejudgment interest under FELA in large part because such interest was unavailable at common law when FELA was enacted); Buell, supra, at 568-570. Cf. Urie v. Thompson, 337 U. S. 163, 174 (1949); Kernan v. American Dredging Co., 355 U. S. 426, 432 (1958).

A

We turn first to the statute. Section 1 of FELA provides that “[e]very common carrier by railroad . . . 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.” 45 U. S. C. § 51. Our task today is determining under what circumstances emotional distress may constitute “injury” resulting from “negligence” for purposes of the statute. As we previously have recognized when considering §51, when Congress enacted FELA in 1908, its “attention was focused primarily upon injuries and death resulting from accidents on interstate railroads.” Urie, supra, at 181. Cognizant of the physical dangers of railroading that resulted in the death or maiming of thousands of workers every year, Congress crafted a federal remedy that shifted part of the “ ‘human overhead’ ” of doing business from employees to their employers. Tiller v. Atlantic Coast Line R. Co., 318 U. S. 54, 58 (1943). See also Wilkerson v. McCarthy, 336 U. S. 53, 68 (1949) (Douglas, J., concurring) (FELA “was designed to put on the railroad industry some of the cost for the legs, eyes, arms, and lives which it consumed in its operations”). In order to further FELA’s humanitarian purposes, Congress did away with several common-law tort defenses that had effectively barred recovery by injured workers. Specifically, the statute abolished the fellow servant rule, rejected the doctrine of contributory negligence in favor of that of comparative negli*543gence, and prohibited employers from exempting themselves from FELA through contract; a 1939 amendment abolished the assumption of risk defense. See 45 U. S. C. §§ 51, 53-55.

We have liberally construed FELA to further Congress’ remedial goal. For example, we held in Rogers v. Missouri Pacific R. Co., 352 U. S. 500 (1957), that a relaxed standard of causation applies under FELA. We stated that “[u]nder this statute the test of a jury case is 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., at 506. In Kernan, supra, we extended the reach of the principle of negligence per se to cover injuries suffered by employees as a result of their employers’ statutory violations, even if the injuries sustained were not of a type that the relevant statute sought to prevent. See id., at 432-436. And in Urie, supra, we held that occupational diseases such as silicosis constitute compensable physical injuries under FELA, thereby rejecting the argument that the statute covered only injuries and deaths caused by accidents. See id., at 181.

That FELA is to be liberally construed, however, does not mean that it is a workers’ compensation statute. We have insisted that FELA “does not make the employer the insurer of the safety of his employees while they are on duty. The basis of his liability is his negligence, not the fact that injuries occur.” Ellis v. Union Pacific R. Co., 329 U. S. 649, 653 (1947). Accord, Inman v. Baltimore & Ohio R. Co., 361 U. S. 138, 140 (1959); Wilkerson, supra, at 61. And while “[wjhat constitutes negligence for the statute’s purposes is a federal question,” Urie, 337 U. S., at 174, we have made clear that this federal question generally turns on principles of common law: “[T]he Federal Employers’ Liability Act is founded on common-law concepts of negligence and injury, subject to such qualifications as Congress has imported into those terms,” id., at 182. Those qualifications, discussed above, *544are the modification or abrogation of several common-law defenses to liability, including contributory negligence and assumption of risk. See 45 U. S. C. §§51, 53-55. Only to the extent of these explicit statutory alterations is FELA “an avowed departure from the rules of the common law.” Sinkler v. Missouri Pacific R. Co., 356 U. S. 326, 329 (1958). Thus, although common-law principles are not necessarily dispositive of questions arising under FELA, unless they are expressly rejected in the text of the statute, they are entitled to great weight in our analysis. Cf. Buell, 480 U. S., at 568. Because FELA is silent on the issue of negligent infliction of emotional distress, common-law principles must play a significant role in our decision.

B

We turn, therefore, to consider the right of recovery pursued by respondents in light of the common law. Cf. Monessen, supra, at 336-339; Buell, 480 U. S., at 568-570. The term “negligent infliction of emotional distress” is largely self-explanatory, but a definitional point should be clarified at the outset. The injury we contemplate when considering negligent infliction of emotional distress is mental or emotional injury, cf. id., at 568, apart from the tort law concepts of pain and suffering. Although pain and suffering technically are mental harms, these terms traditionally “have been used to describe sensations stemming directly from a physical injury or condition.” Pearson, Liability to Bystanders for Negligently Inflicted Emotional Harm — A Comment on the Nature of Arbitrary Rules, 34 U. Fla. L. Rev. 477, 485, n. 45 (1982). The injury we deal with here is mental or emotional harm (such as fright or anxiety) that is caused by the negligence of another and that is not directly brought about by a physical injury, but that may manifest itself in physical symptoms.

Nearly all of the States have recognized a right to recover for negligent infliction of emotional distress, as we have de*545fined it.8 No jurisdiction, however, allows recovery for all emotional harms, no matter how intangible or trivial, that might be causally linked to the negligence of another. Indeed, significant limitations, taking the form of “tests” or “rules,” are placed by the common law on the right to recover for negligently inflicted emotional distress, and have been since the right was first recognized late in the last century.

Behind these limitations lie a variety of policy considerations, many of them based on the fundamental differences between emotional and physical injuries. “Because the etiology of emotional disturbance is usually not as readily apparent as that of a broken bone following an automobile accident, courts have been concerned . . . that recognition of a cause of action for [emotional] injury when not related to any physical trauma may inundate judicial resources with a flood of relatively trivial claims, many of which may be imagined or falsified, and that liability may be imposed for highly remote consequences of a negligent act.” Maloney v. Conroy, 208 Conn. 392, 397-398, 545 A. 2d 1059, 1061 (1988). The last concern has been particularly significant. Emotional injuries may occur far removed in time and space from the negligent conduct that triggered them. Moreover, in contrast to the situation with physical injury, there are no necessary finite limits on the number of persons who might suffer emotional injury as a result of a given negligent act.4 The *546incidence and severity of emotional injuries are also more difficult to predict than those of typical physical injuries because they depend on psychological factors that ordinarily are not apparent to potential tortfeasors.

. For all of these reasons, courts have realized that recognition of a cause of action for negligent infliction of emotional distress holds out the very real possibility of nearly infinite and unpredictable liability for defendants. Courts therefore have placed substantial limitations on the class of plaintiffs that may recover for emotional injuries and on the injuries that may be compensable. See, e. g., Thing v. La Chusa, 48 Cal. 3d 644, 654, 771 P. 2d 814, 819 (1989) (“[Pjolicy considerations mandat[e] that infinite liability be avoided by restrictions that... narrow the class of potential plaintiffs”); Tobin v. Grossman, 24 N. Y. 2d 609, 616, 249 N. E. 2d 419, 423 (1969).5 Some courts phrase the limitations in terms of proximate causation; that is, only certain plaintiffs or injuries are reasonably foreseeable. Other courts speak of the limitations in terms of duty; the defendant owes only a certain class of plaintiffs a duty to avoid inflicting emotional harm. See, e. g., Pearson, supra, at 489, n. 72 (discussing Palsgraf v. Long Island R. Co., 248 N. Y. 339, 162 N. E. 99 (1928)). These formulations are functionally equivalent. We shall refer to the common-law limitations as outlining the duty of defendants with regard to negligent infliction of emotional distress.

Three major limiting tests for evaluating claims alleging negligent infliction of emotional distress have developed in the common law. The first of these has come to be known *547as the “physical impact” test. It originated a century ago in some of the first cases recognizing recovery for negligently inflicted emotional distress. At the time Congress enacted FELA in 1908, most of the major industrial States had embraced this test. See Throckmorton, Damages for Fright, 34 Harv. L. Rev. 260, 263-264, and n. 25 (1921).6 Under the physical impact test, a plaintiff seeking damages for emotional injury stemming from a negligent act must have contemporaneously sustained a physical impact (no matter how slight) or injury due to the defendant’s conduct. Most jurisdictions have abandoned this test, but at least five States continue to adhere to it.7

The second test has come to be referred to as the “zone of danger” test. It came into use at roughly the same time as the physical impact test, and had been adopted by several jurisdictions at the time FELA was enacted. See Throckmorton, supra, at 264-265, and n. 28.8 See also Bohlen, Right to Recover for Injury Resulting from Negligence Without Impact, 50 Am. L. Reg. 141, and nn. 3-5 (1902). Perhaps based on the realization that “a near miss may be as frightening as a direct hit,” Pearson, U. Fla. L. Rev., at 488, the zone of danger test limits recovery for emotional injury to those plaintiffs who sustain a physical impact as a result *548of a defendant’s negligent conduct, or who are placed in immediate risk of physical harm by that conduct. That is, “those within the zone of danger of physical impact can recover for fright, and those outside of it cannot.” Id., at 489. The zone of danger test currently is followed in 14 jurisdictions.9

The third prominent limiting test is the “relative bystander” test, which was first enunciated in Dillon v. Legg, 68 Cal. 2d 728, 441 P. 2d 912 (1968). In Dillon, the California Supreme Court rejected the zone of danger test and suggested that the availability of recovery should turn, for the most part, on whether the defendant could reasonably have foreseen the emotional injury to the plaintiff. The court offered three factors to be considered as bearing on the question of reasonable foreseeability:

“(1) Whether plaintiff was located near the scene of the accident as contrasted with one who was a distance away from it. (2) Whether the shock resulted from a direct emotional impact upon plaintiff from the sensory and contemporaneous observance of the accident, as contrasted with learning of the accident from others after its occurrence. (3) Whether plaintiff and the victim were closely related, as contrasted with an absence of any relationship or the presence of only a distant relationship.” Id., at 740-741, 441 P. 2d, at 920.

*549The courts of nearly half the States now allow bystanders outside of the zone of danger to obtain recovery in certain circumstances for emotional distress brought on by witnessing the injury or death of a third party (who typically must be a close relative of the bystander) that is caused by the defendant’s negligence.10 Most of these jurisdictions have adopted the Dillon factors either verbatim or with variations and additions, and have held some or all of these factors to be substantive limitations on recovery.11

HI

A

Having laid out the relevant legal framework, we turn to the questions presented. As an initial matter, we agree *550with the Third Circuit that claims for damages for negligent infliction of emotional distress are cognizable under FELA. A combination of many of the factors discussed above makes this conclusion an easy one. A right to recover for negligently inflicted emotional distress was recognized in some form by many American jurisdictions at the time FELA was enacted, see nn. 6 and 8, supra, and this right is nearly universally recognized among the States today. See supra, at 546-549. Moreover, we have accorded broad scope to the statutory term “injury” in the past in light of FELA’s remedial purposes. Cf. Urie, 337 U. S., at 181. We see no reason why emotional injury should not be held to be encompassed within that term, especially given that “severe emotional injuries can be just as debilitating as physical injuries.” Gottshall, 988 F. 2d, at 361. We therefore hold that, as part of its “duty to use reasonable care in furnishing its employees with a safe place to work,” Buell, 480 U. S., at 558, a railroad has a duty under FELA to avoid subjecting its workers to negligently inflicted emotional injury. This latter duty, however, is not self-defining. Respondents defend the Third Circuit’s definition of the duty we recognize today; Conrail offers its own proposed delineation. We consider the proposals in turn.

B

When setting out its view of the proper scope of recovery for negligently inflicted emotional distress under FELA, the Third Circuit explicitly refused to adopt any of the common-law tests described above; indeed, the court in Gottskall went so far as to state that “doctrinal common law distinctions are to be discarded when they bar recovery on meritorious FELA claims.” 988 F. 2d, at 369. Instead, the court developed its own test, under which “[t]he issue is whether the factual circumstances . .. provide a threshold assurance that there is a likelihood of genuine and serious emotional injury.” Id., at 371. If this threshold test is satisfied, the claim should be evaluated in light of traditional tort concepts *551such as breach of duty, injury, and causation, with the focus resting on the foreseeability of the plaintiff’s injury. Id., at 374-375. In Gottshall, the Third Circuit did at least consider the plaintiff’s claim in light of the common law of negligent infliction of emotional distress as part of its factual “genuineness” inquiry. By the time the court next applied the Gottshall genuineness test, however, the common-law aspect of its analysis had completely disappeared; Carlisle’s stress-related claim was not evaluated under any of the common-law tests. In Carlisle, the Third Circuit refined its test to two questions — whether there was convincing evidence of the genuineness of the emotional injury claim (with “genuine” meaning authentic and serious), and if there was, whether the injury was foreseeable. If these questions could be answered affirmatively by the court, there was “no bar to recovery under the FELA.” 990 F. 2d, at 98.

The Third Circuit’s standard is fatally flawed in a number of respects. First, as discussed above, because negligent infliction of emotional distress is not explicitly addressed in the statute, the common-law background of this right of recovery must play a vital role in giving content to the scope of an employer’s duty under FELA to avoid inflicting emotional injury. Cf. Monessen, 486 U. S., at 336-339; Buell, supra, at 568-570; Urie, supra, at 182. By treating the common-law tests as mere arbitrary restrictions to be disregarded if they stand in the way of recovery on “meritorious” FELA claims, the Third Circuit put the cart before the horse: The common law must inform the availability of a right to recover under FELA for negligently inflicted emotional distress, so the “merit” of a FELA claim of this type cannot be ascertained without reference to the common law.

Perhaps the court below believed that its focus on the perceived genuineness of the claimed emotional injury adequately addressed the concerns of the common-law courts in dealing with emotional injury claims. But the potential for fraudulent and trivial claims — the concern identified by the *552Third Circuit — is only one of the difficulties created by allowing actions for negligently inflicted emotional distress. A more significant problem is the prospect that allowing such suits can lead to unpredictable and nearly infinite liability for defendants. The common law consistently has sought to place limits on this potential liability by restricting the class of plaintiffs who may recover and the types of harm for which plaintiffs may recover. This concern underlying the common-law tests has nothing to do with the potential for fraudulent claims; on the contrary, it is based upon the recognized possibility of genuine claims from the essentially infinite number of persons, in an infinite variety of situations, who might suffer real emotional harm as a result of a single instance of negligent conduct.

Second, we question the viability of the genuineness test on its own terms. The Third Circuit recognized that “there must be some finite limit to the railway’s potential liability” for emotional injury claims under FELA, and suggested that liability could be restricted through application of the genuineness test. Gottshall, supra, at 379. But as just explained, testing for the “genuineness” of an injury alone cannot appreciably diminish the possibility of infinite liability. Such a fact-specific test, moreover, would be bound to lead to haphazard results. Judges would be forced to make highly subjective determinations concerning the authenticity of claims for emotional injury, which are far less susceptible to objective medical proof than are their physical counterparts. To the extent the genuineness test could limit potential liability, it could do so only inconsistently. Employers such as Conrail would be given no standard against which to regulate their conduct under such an ad hoc approach. In the context of claims for intangible harms brought under a negligence statute, we find such an arbitrary result unacceptable. Cf. Stadler v. Cross, 295 N. W. 2d 552, 554 (Minn. 1980).

Third, to the extent the Third Circuit relied on the concept of foreseeability as a meaningful limitation on liability, we *553believe that reliance to be misplaced. If one takes a broad enough view, all consequences of a negligent act, no matter how far removed in time or space, may be foreseen. Conditioning liability on foreseeability, therefore, is hardly a condition at all. “Every injury has ramifying consequences, like the ripplings of the waters, without end. The problem for the law is to limit the legal consequences of wrongs to a controllable degree.” Tobin, 24 N. Y. 2d, at 619, 249 N. E. 2d, at 424. See also Thing, 48 Cal. 3d, at 668, 771 P. 2d, at 830 (“[T]here. are clear judicial days on which a court can foresee forever and thus determine liability but none on which that foresight alone provides a socially and judicially acceptable limit on recovery”).

This is true as a practical matter in the FELA context as well, even though the statute limits recovery to railroad workers. If emotional injury to Gottshall was foreseeable to Conrail, such injury to the other seven members of his work crew was also foreseeable. Because one need not witness an accident to suffer emotional injury therefrom, however, the potential liability would not necessarily have to end there; any Conrail employees who heard or read about the events surrounding Johns’ death could also foreseeably have suffered emotional injury as a result. Of course, not all of these workers would have been as traumatized by the tragedy as was Gottshall, but many could have been. Under the Third Circuit’s standard, Conrail thus could face the potential of unpredictable liability to a large number of employees far removed from the scene of the allegedly negligent conduct that led to Johns’ death.12

*554Finally, the Third Circuit in Carlisle erred in upholding “a claim under the FELA for negligent infliction of emotional distress arising from work-related stress.” 990 F. 2d, at 97-98. We find no support in the common law for this unprecedented holding, which would impose a duty to avoid creating a stressful work environment, and thereby dramatically expand employers’ FELA liability to cover the stresses and strains of everyday employment. Indeed, the Third Circuit’s ruling would tend to make railroads the insurers of the emotional well-being and mental health of their employees. We have made clear, however, that FELA is not an insurance statute. See, e. g., Ellis, 329 U. S., at 653. For the foregoing reasons, we reject the Third Circuit’s approach.

C

Conrail suggests that we adopt the common-law zone of danger test as delimiting the proper scope of an employer’s duty under FELA to avoid subjecting its employees to negligently inflicted emotional injury. We agree that the zone of danger test best reconciles the concerns of the common law with the principles underlying our FELA jurisprudence.

As we did in Monessen, we begin with the state of the common law in 1908, when FELA was enacted. In determining in Monessen whether prejudgment interest was available under FELA, we recognized that the common law in 1908 did not allow such interest in personal injury and wrongful-death suits. Because in enacting FELA, “Congress expressly dispensed with other common-law doctrines of that era, such as the defense of contributory negligence,” but “did not deal at all with the equally well established doctrine barring the recovery of prejudgment interest,” we concluded that Congress intended to leave the common-law rule intact. 486 U. S., at 337-338. In contrast, the right to recover for negligently inflicted emotional distress was well established in many jurisdictions in 1908. Although at that time, “the weight of American authority” favored the physi*555cal impact test, Throckmorton, 34 Harv. L. Rev., at 264, the zone of danger test had been adopted by a significant number of jurisdictions. See n. 8, supra. Moreover, because it was recognized as being a progressive rule of liability that was less restrictive than the physical impact test, the zone of danger test would have been more consistent than the physical impact test with FELA’s broad remedial goals. See Waube v. Warrington, 216 Wis. 603, 608, 258 N. W. 497, 499 (1935) (discussing early emotional injury cases and referring to zone of danger test as “the liberal rule”). Considering the question “in the appropriate historical context,” Monessen, supra, at 337, then, it is reasonable to conclude that Congress intended the scope of the duty to avoid inflicting emotional distress under FELA to be coextensive with that established under the zone of danger test. That is, an emotional injury constitutes “injury” resulting from the employer’s “negligence” for purposes of FELA only if it would be compensable under the terms of the zone of danger test. See 45 U. S. C. § 51. Cf. Urie, 337 U. S., at 182.

Current usage only confirms this historical pedigree. The zone of danger test presently is followed by 14 jurisdictions. It therefore remains to this day a well-established “common-law concep[t] of negligence,” ibid., that is suitable to inform our determination of the federal question of what constitutes negligence for purposes of FELA. Cf. Buell, 480 U. S., at 568-570; Kerman, 355 U. S., at 432.

The zone of danger test also is consistent with FELA’s central focus on physical perils. We have recognized that FELA was intended to provide compensation for the injuries and deaths caused by the physical dangers of railroad work by allowing employees or their estates to assert damages claims. Cf. Urie, supra, at 181. By imposing liability, FELA presumably also was meant to encourage employers to improve safety measures in order to avoid those claims. Cf. Wilkerson, 336 U. S., at 68 (Douglas, J., concurring). As the Seventh Circuit has observed, FELA was (and is) aimed *556at ensuring “the security of the person from physical invasions or menaces.” Lancaster v. Norfolk & Western R. Co., 773 F. 2d 807, 813 (1985), cert. denied, 480 U. S. 945 (1987). But while the statute may have been primarily focused on physical injury, it refers simply to “injury,” which may encompass both physical and emotional injury. We believe that allowing recovery for negligently inflicted emotional injury as provided for under the zone of danger test best harmonizes these considerations. Under this test, a worker within the zone of danger of physical impact will be able to recover for emotional injury caused by fear of physical injury to himself, whereas a worker outside the zone will not. Railroad employees thus will be able to recover for injuries— physical and emotional — caused by the negligent conduct of their employers that threatens them imminently with physical impact. This rule will further Congress’ goal in enacting the statute of alleviating the physical dangers of railroading.

The physical impact test, of course, would achieve many of the same ends as the zone of danger test. We see no reason, however, to allow an employer to escape liability for emotional injury caused by the apprehension of physical impact simply because of the fortuity that the impact did not occur. And the physical impact test has considerably less support in the current state of the common law than the zone of danger test. See supra, at 546-549.

As for the relative bystander test, we conclude that it is an inappropriate rule in the FELA context. As an initial matter, it was not developed until 60 years after FELA’s enactment, and therefore lacks historical support. Cf. Monessen, supra. Moreover, in most jurisdictions that adhere to it, this test limits recovery to persons who witness the severe injury or death of a close family member. Only railroad employees (and their estates) may bring FELA claims, however, and presumably it would be a rare occurrence for a worker to witness during the course of his employment the injury or death of a close family member. In *557any event, we discern from FELA and its emphasis on protecting employees from physical harms no basis to extend recovery to bystanders outside the zone of danger. Cf. Gaston v. Flowers Transp., 866 F. 2d 816, 820-821 (CA5 1989).

Respondents decry the zone of danger test as arbitrarily excluding valid claims for emotional injury. But “[characterizing a rule limiting liability as ‘unprincipled’ or ‘arbitrary’ is often the result of overemphasizing the policy considerations favoring imposition of liability, while at the same time failing to acknowledge any countervailing policies and the necessary compromise between competing and inconsistent policies informing the rule.” Cameron v. Pepin, 610 A. 2d 279, 283 (Me. 1992). Our FELA cases require that we look to the common law when considering the right to recover asserted by respondents, and the common law restricts recovery for negligent infliction of emotional distress on several policy grounds: the potential for a flood of trivial suits, the possibility of fraudulent claims that are difficult for judges and juries to detect, and the specter of unlimited and unpredictable liability. Although some of these grounds have been criticized by commentators, they all continue to give caution to courts. We believe the concerns that underlie the common-law tests, and particularly the fear of unlimited liability, to be well founded.

Perhaps the zone of danger test is “arbitrary” in the sense that it does not allow recovery for all emotional distress. But it is fully consistent with our understanding of the statute. And for the reasons discussed above, we conclude that the policy considerations of the common law as they are embodied in the zone of danger test best accord with the concerns that have motivated our FELA jurisprudence.

IV

Because the Third Circuit applied an erroneous standard for evaluating claims for negligent infliction of emotional *558distress brought under FELA, we reverse the judgments below. In Gottskall, we remand for reconsideration under the zone of danger test announced today. Gottshall asserts before this Court that he would in fact meet the requirements of the zone of danger test, while Conrail disagrees. The question was not adequately briefed or argued before us, however, and we believe it best to allow the Third Circuit to consider the question in the first instance in light of relevant common-law precedent.

In Carlisle, however, we remand with instructions to enter judgment for Conrail. Carlisle’s work-stress-related claim plainly does not fall within the common law’s conception of the zone of danger, and Carlisle makes no argument that it does. Without any support in the common law for such a claim, we will not take the radical step of reading FELA as compensating for stress arising in the ordinary course of employment. In short, the core of Carlisle’s complaint was that he “had been given too much — not too dangerous — work to do. That is not our idea of an FELA claim.” Lancaster, supra, at 813.

The judgments of the Court of Appeals are reversed, and the cases are remanded for further proceedings consistent with this opinion.

So ordered.

concurring.

I join the Court’s opinion holding that claims for negligent infliction of emotional distress are cognizable under the Federal Employers’ Liability Act (FELA), and that the zone of danger test is the appropriate rule for determining liability for such claims. I write separately to make explicit what I believe the Court’s duty to be in interpreting FELA. That duty is to develop a federal common law of negligence under FELA, informed by reference to the evolving common law. See Atchison, T. & S. F. R. Co. v. Buell, 480 U. S. 557, 568-570 (1987). As we have explained:

*559“[I]nstead of a detailed statute codifying common-law principles, Congress saw fit to enact a statute of the most general terms, thus leaving in large measure to the courts the duty of fashioning remedies for injured employees in a manner analogous to the development of tort remedies at common law. But it is clear that the general congressional intent was to provide liberal recovery for injured workers . . . and it is also clear that Congress intended the creation of no static remedy, but one which would be developed and enlarged to meet changing conditions and changing concepts of industry’s duty toward its workers.” Kernan v. American Dredging Co., 355 U. S. 426, 432 (1958).

Because I believe the Court’s decision today to be a faithful exercise of that duty, and because there can be no question that adoption of the zone of danger test is well within the discretion left to the federal courts under FELA, I join in its opinion.

The Federal Employers’ Liability Act (FELA or Act), 45 U. S. C. § 51 et seq., instructs interstate railroads “ ‘to use reasonable care in furnishing [their] employees with a safe place to work.’ ” Ante, at 550, quoting Atchison, T. & S. F. R. Co. v. Buell, 480 U. S. 557, 558 (1987). As the Court today recognizes, the FE LA-imposed obligation encompasses “a duty... to avoid subjecting [railroad] workers to negligently inflicted emotional injury.” Ante, at 550.

The Court limits the scope of the railroad’s liability, however, by selecting one of the various “tests” state courts have applied to restrict recovery by members of the public for negligently inflicted emotional distress. The Court derives its limitation largely from a concern, often expressed in state court opinions, about the prospect of “infinite liability” to an “infinite number of persons.” See ante, at 552. This *560concern should not control in the context of the FELA, as I see it, for the class of potential plaintiffs under the FELA is not the public at large; the Act covers only railroad workers who sustain injuries on the job. In view of the broad language of the Act,1 and this Court’s repeated reminders that the FELA is to be liberally construed, I cannot regard as faithful to the legislation and our case law under it the restrictive test announced in the Court’s opinion.

I

The FELA was designed to provide a federal “statutory negligence action . . . significantly different from the ordinary common-law negligence action.” Rogers v. Missouri Pacific R. Co., 352 U. S. 500, 509-510 (1957). An “avowed departure” from prevailing common-law rules, Sinkler v. Missouri Pacific R. Co., 356 U. S. 326, 329 (1958), the Act advanced twin purposes: “to eliminate a number of traditional defenses to tort liability and to facilitate recovery in meritorious cases.” Buell, supra, at 561.2 “Congress intended the creation of no static remedy, but one which would be developed and enlarged to meet changing conditions and changing concepts of industry’s duty toward its workers.” Kernan v. American Dredging Co., 355 U. S. 426, 432 (1958). Relying upon “the breadth of the statutory language, [and] the Act’s humanitarian purposes,” this Court has accorded the FELA a notably “liberal construction in order to accom*561plish [Congress’] objects.” Urie v. Thompson, 337 U. S. 163, 180 (1949); see Buell, supra, at 562.

In particular, the Court has given full scope to the key statutory term "injury." The Act prescribes that "[e]very common carrier by railroad.. . shall be liable in damages to any person suffering injury while he is employed by such carrier." 45 U. S. C. § 51. That prescription, this Court observed, is "not restrictive as to . . . the particular kind of injury~" Uric, 337 U. S., at 181. "[W]hen the statute was enacted," it is true, "Congress' attention was focused primarily upon . . . accidents on interstate railroads," for "these were the major causes of injury and death resulting from railroad operations." Ibid. But "accidental injuries were not the only ones likely to occur," and Congress chose "all-inclusive wording." Ibid. "To read into [language as broad as could be framed] a restriction [tied] to . . . the particular sorts of harms inflicted," the Court recognized, "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." Id., at 181-182.

II

Seven years ago, in Atchison, T. & S. F. R. Co. v. Buell, 480 U. S. 557 (1987), the Court left unresolved the question whether emotional injury is compensable under the FELA, because the record in that case did not adequately present the issue. Id., at 560-561,570-571. In his unanimous opinion for the Court, Justice Stevens explained why the question could not be resolved on a fact-thin record:

“[W]hether ‘emotional injury’ is cognizable under the FELA is not necessarily an abstract point of law or a pure question of statutory construction that might be answerable without exacting scrutiny of the facts of the case. Assuming, as we have, that FELA jurisprudence gleans guidance from common-law developments, see *562Urie v. Thompson, 337 U. S., at 174, whether 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.” Id., at 568.

“[T]he 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.” Id., at 570.

In deciding the cases now under review, the Court of Appeals endeavored to “ ‘field the Buell pitch.’ ” 988 F. 2d 355, 365 (CA3 1993), quoting Plaisance v. Texaco, Inc., 937 F. 2d 1004, 1009 (CA5 1991).

A

In respondent Gottshall’s case, the Court of Appeals first described the various rules state courts have applied to common-law actions for negligent infliction of emotional distress. 988 F. 2d, at 361-362. That court emphasized, however, that “[determining FELA liability is distinctly a federal question.” Id., at 362. State common-law decisions, the Court of Appeals observed, “do not necessarily etch the contours of the federal right,” ibid., for the common law that courts develop to fill the FELA’s interstices is “federal” in character. See id., at 367.

In addition to the FELA’s express abolition of traditional employer defenses, the Court of Appeals next noted, this Court’s decisions interpreting the FELA served as path-markers. The Court of Appeals referred to decisions that had relaxed “the strict requirements of causation in common law,” id., at 368, citing Rogers, 352 U. S., at 506, broadened the conception of negligence per se, see 988 F. 2d, at 368, citing Kernan, 355 U. S., at 437-439, and generously con*563strued the FELA’s injury requirement, 988 F. 2d, at 368, citing Urie, 337 U. S., at 181-182. The FELA, the Court of Appeals concluded:

“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.” 988 F. 2d, at 369.

Accordingly, the Court of Appeals “refused to designate a particular common law test as the test” applicable in FELA cases. Id., at 365. Instead, the court looked to the purposes of those tests: to distinguish “the meritorious [claim] from the feigned and frivolous,” id., at 369, and to assure that liability for negligently inflicted emotional distress does not expand “into the ‘fantastic realm of infinite liability.’” Id., at 372, quoting Amaya v. Home Ice, Fuel & Supply Co., 59 Cal. 2d 295, 315, 379 P. 2d 513, 525 (1963); see also 988 F. 2d, at 381-382.

FELA jurisprudence, the Court of Appeals reasoned, has evolved not through a “rules first” approach, but in the traditional, fact-bound, case-by-case common-law way. See id., at 371. The court therefore undertook to determine “whether the factual circumstances [in Gottshall’s case] provide a threshold assurance that there is a likelihood of genuine and serious emotional injury.” Ibid. “[O]ne consideration” in that inquiry, the court said, “is whether plaintiff has a ‘solid basis in the present state of common law to permit him to recover.’” Ibid., quoting Outten v. National Railroad Passenger Corp., 928 F. 2d 74, 79 (CA3 1991).

Gottshall’s claim, the Court of Appeals held, presented the requisite “threshold assurance.” His emotional distress, diagnosed by three doctors as major depression and posttraumatic stress disorder, 988 F. 2d, at 374, was unquestionably genuine and severe: He was institutionalized for three weeks, followed by continuing outpatient care; he lost 40 pounds; and he suffered from “suicidal preoccupations, anxi*564ety, sleep onset insomnia, cold sweats, . . . nausea, physical weakness, repetitive nightmares and a fear of leaving home.” Ibid.; see also id., at 373 (noting that Conrail “wisely declined” to attack Gottshall’s claim as fraudulent). Gottshall’s afflictions, the Court of Appeals observed, satisfied the “physical manifestation” limitation that some States, and the Second Restatement of Torts, place on emotional distress recovery. See id., at 373-374 (citing cases); Restatement (Second) of Torts §436A (1965) (no liability for emotional distress without “bodily harm or other compensable damage”); ibid., Comment c (“[L]ong continued nausea or headaches may amount to physical illness, which is bodily harm; . . . long continued mental disturbance . . . may be classified by the courts as illness” and thus be compensable). Cf. Buell, 480 U. S., at 570, n. 22 (suggesting a distinction between claims for “pure emotional injury” and those involving “physical symptoms in addition to . . . severe psychological illness”).

The Court of Appeals also inspected the facts under the “bystande[r]” test, versions of which have been adopted by nearly half the States. See ante, at 549. While acknowledging that Gottshall did not satisfy the more restrictive versions of the “bystander” test, the court observed that several States have allowed recovery even where, as here, the plaintiff and the victim of physical injury were unrelated by blood or marriage. See 988 F. 2d, at 371 (citing cases). Further, the court noted, given “the reality of the railway industry,” rarely will one “se[e] another family member injured while working in the railroad yard.” Id., at 372. A strict version of the bystander rule, therefore, would operate not to limit recovery to the most meritorious cases, but almost to preclude bystander recovery altogether.

To adapt the bystander rule to the FELA context, the court looked to the reasons for limiting bystander recovery: to avoid compensating plaintiffs with fraudulent or trivial claims, and to prevent liability from becoming “an intolerable *565burden upon society.” Id., at 369, 372. The court held that neither concern barred recovery in Gottshall’s case. The genuineness of Gottshall’s claim appeared not just in the manifestations of his distress, the court said, but also in the extraordinarily close, 15-year friendship between Gottshall and Johns, the decedent. Id., at 371. Liability to bystanders, the court concluded, would be far less burdensome in the FELA context, where only close co-workers are potential plaintiffs, than in the context of a common-law rule applicable to society as a whole. Id., at 372. In this regard, the Court of Appeals again recalled, this Court has constantly admonished lower courts that “recovery [under the FELA] should be liberally granted,” ibid., “so that the remedial and humanitarian goals of the statute can be fully implemented,” id., at 373.

Satisfied that Gottshall had crossed the “genuine and severe” injury threshold, the Court of Appeals inquired whether he had a triable case on breach of duty and causation. Id., at 374. Here, the court emphasized that Gottshall’s distress was attributable not to “the ordinary stress of the job,” id., at 375, but instead, to Conrail’s decision to send a crew of men, most of them 50 to 60 years old and many of them overweight, out into 97-degree heat at high noon, in a remote, sun-baked location, requiring them to replace heavy steel rails at an extraordinarily fast pace without breaks, and without maintaining radio contact or taking any other precautions to protect the men’s safety, id., at 376-377.

The Court of Appeals stated, further, that even if Conrail could be said to have acted reasonably up to the time of Johns’ death, “its conduct after the death raises an issue of whether it breached a legal duty.” Id., at 378. The Conrail supervisor required the crew to return to work immediately after Johns’ corpse was laid by the side of the road, covered but still in view. Ibid. The next day, Gottshall alleged, the supervisor “reprimanded him for administering CPR to Johns,” id., at 359, then pushed the crew even harder under *566the same conditions, requiring a full day, plus three or four hours of overtime, id., at 378. These circumstances, the Court of Appeals concluded, “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.” Id., at 378.

B

Upholding a jury verdict for plaintiff in Carlisle, the Court of Appeals “reaffirmed]” its Gottshall holding that “no single common law standard” governs in “weighing the genuineness of emotional injury claims.” Instead, the court said:

“[C]ourts . . . should engage in an initial review of the factual indicia of the genuineness of a claim, taking into account broadly used common law standards, then should apply the traditional negligence elements of duty, foreseeability, breach, and causation in weighing the merits of that claim.” 990 F. 2d 90, 98 (CA3 1993).

The Court of Appeals held that the evidence submitted to the jury amply established the claim’s genuineness. Carlisle testified that, after Conrail’s 1984 reduction in force, the pressure on train dispatchers in Philadelphia, already substantial, increased dramatically. As the person chiefly responsible for ensuring the safety of “trains carrying passengers, freight and hazardous materials,” Carlisle became “increasingly anxious” over the sharp reduction in staff, together with the outdated equipment and “Conrail’s repeated instructions to ignore safety concerns, such as malfunctioning equipment or poor maintenance.” Id., at 92. When Carlisle was compelled to work 12- to 15-hour shifts for 15 consecutive days, the resulting additional pressures, and the difficulty of working for “an abusive, alcoholic supervisor,” led, according to Carlisle’s expert witness, to the nervous breakdown he suffered. Ibid.

*567Other evidence confirmed Carlisle’s testimony. Depositions taken from “Carlisle’s co-workers and subordinates” averred that “their jobs as dispatchers and supervisors in the Philadelphia Conrail offices had caused them to suffer cardiac arrests, nervous breakdowns, and a variety of emotional problems such as depression, paranoia and insomnia.” Ibid. An official report prepared by the Federal Railway Administration “criticized the outdated equipment and hazardous working conditions at Conrail’s Philadelphia dispatching office.” Id., at 93. Furthermore, the Court of Appeals pointed out, Carlisle’s emotional injury was “accompanied by obvious physical manifestations”: “insomnia, fatigue, headaches, . . . sleepwalking and substantial weight-loss.” Id., at 97, n. 11, 92. The court specifically noted: “We do not face and do not decide the issue of whether purely emotional injury, caused by extended exposure to stressful, dangerous working conditions, would be compensable under the FELA.” Id., at 97, n. 11.

Satisfied that the jury could indeed find Carlisle’s injury genuine, and continuing to follow the path it had marked in Gottshall, the court next examined the negligence elements of Carlisle’s claim. Emphasizing that “Conrail had ample notice of the stressful and dangerous conditions under which Carlisle was forced to work,” including actual notice of physical and emotional injuries sustained by Carlisle’s co-workers, 990 F. 2d, at 97, the Court of Appeals affirmed the District Court’s denial of Conrail’s motions for judgment n.o.v. or in the alternative for a new trial. Carlisle’s “extended exposure to dangerous and stressful working conditions,” the court concluded, constituted a breach of Conrail’s duty to provide a safe workplace, and the breach caused Carlisle’s injuries. Id., at 97-98.

Ill

The Court initially “agree[s] with the Third Circuit that claims for damages for negligent infliction of emotional distress are cognizable under FELA.” Ante, at 549-550. *568This conclusion, “an easy one” for the Court, ante, at 550, is informed by prior decisions giving full scope to the FELA’s term “injury.” The Court had explained in Urie that an occupational disease incurred in the course of employment— silicosis in that particular case — is as much “injury ... as scalding from a boiler’s explosion.” 337 U. S., at 187. Rejecting a reading of the statute that would confine coverage to “accidental injury” of the kind that particularly prompted the 1908 Congress to enact the FELA, the Court said of the occupational disease at issue:

“[W]hen the employer’s negligence impairs or destroys an employee’s health by requiring him to work under conditions likely to bring about such harmful consequences, the injury to the employee is just as great when it follows, often inevitably, from a carrier’s negligent course pursued over an extended period of time as when it comes with the suddenness of lightning.” Id., at 186-187.

Similarly, as the Court recognizes today, “ ‘severe emotional injuries can be just as debilitating as physical injuries,’” hence there is “no reason why emotional injury should not be held to be encompassed within th[e] term [‘injury’].” Ante, at 550, quoting Gottshall, 988 F. 2d, at 361.

In my view, the Court of Appeals correctly determined that Gottshall’s submissions should survive Conrail’s motion for summary judgment, and that the jury’s verdict in favor of Carlisle should stand. Both workers suffered severe injury on the job, and plausibly tied their afflictions to Conrail’s negligence. Both experienced not just emotional, but also physical, distress: Gottshall lost 40 pounds and suffered from insomnia, physical weakness, and cold sweats, while Carlisle experienced “insomnia, fatigue, headaches,. . . sleepwalking and substantial weight-loss.” Id., at 374; 990 F. 2d, at 92, 97, n. 11. The Court emphasizes the “significant role” that “common-law principles must play.” Ante, at 544. Notably *569in that regard, both Gottshall and Carlisle satisfy the “physical manifestation” test endorsed by the Restatement of Torts. See supra, at 564, 567; see also W. Keeton, D. Dobbs, R. Keeton, & D. Owen, Prosser and Keeton on Law of Torts 364 (5th ed. 1984) (“the great majority of courts have now repudiated the requirement of ‘impact,’ regarding as sufficient the requirement that the mental distress be certified by some physical injury, illness or other objective physical manifestation”); id., at 364, n. 55 (citing cases). Thus, without gainsaying that “FELA jurisprudence gleans guidance from common-law developments,” Buell, 480 U. S., at 568, one can readily conclude that both Gottshall and Carlisle have made sufficient showings of “injuries” compensable under the FELA.3

Notwithstanding its recognition that the word “injury,” as used in the FELA, “may encompass both physical and emotional injury,” the Court elects to render compensable only emotional distress stemming from a worker’s placement in the “zone of danger.” Ante, at 556. In other words, to recover for emotional distress, the railroad employee must show that negligence attributable to his employer threatened him “imminently with physical impact.” Ibid. Based on the “zone” test, the Court reverses the judgment for Carlisle outright and remands Gottshall’s case for reconsideration under that standard. Ante, at 557-558.

The Court offers three justifications for its adoption of the “zone of danger” test. First, the Court suggests that the “zone” test is most firmly rooted in “the common law.” The Court mentions that several jurisdictions had adopted the zone of danger test by 1908, ante, at 546,547, n. 8 (citing cases from eight States), and that the test “currently is followed in 14 jurisdictions.” Ante, at 548. But that very exposition *570tells us that the “zone” test never held sway in a majority of States.

Moreover, the Court never decides firmly on the point of reference, present or historical, from which to evaluate the relative support the different common-law rules have enjoyed. If the Court regarded as decisive the degree of support a rule currently enjoys among state courts, the Court would allow bystander recovery, permitted in some form in “nearly half the States.” Ante, at 549. But cf. ante, at 556 (bystander rule “was not developed until 60 years after FELA’s enactment, and therefore lacks historical support”). If, on the other hand, the Court decided that historical support carried the day, then the impact rule, preferred by most jurisdictions in 1908, would be the Court’s choice. But cf. ibid, (preferring the zone of danger test to the impact rule, because, inter alia, the latter “has considerably less support in the current state of the common law” than the former).

The Court further maintains that the zone of danger test is preferable because it is “consistent with FELA’s central focus on physical perils.” Ante, at 555. But, as already underscored, see supra, at 561, the FELA’s language “is as broad as could be framed____ On its face, every injury suffered [on the job] by any employee ... by reason of the carrier’s negligence was made compensable.” Urie, 337 U. S., at 181. And the FELA’s strikingly broad language, characteristically, “has been construed even more broadly,” in line with Congress’ dominant remedial objective. Buell, 480 U. S., at 562; Urie, supra, at 181 (“[N]othing in either the language or the legislative history discloses expressly any intent to exclude from the Act’s coverage any injury resulting ‘in whole or in part from the negligence’ of the carrier”).

The Court’s principal reason for restricting the FELA’s coverage of emotional distress claims is its fear of “infinite liability” to an “infinite number of persons.” See ante, at 552; see also ante, at 557 (referring to “the specter of unlimited and unpredictable liability,” and stating that “the fear of unlimited liability . . . [is] well founded”). The universe *571of potential FELA plaintiffs, however, is hardly “infinite.” The statute does not govern the public at large. Only persons “suffering injury ... while employed” by a railroad may recover under the FELA, ¿nd to do so, the complainant must show that the injury resulted from the railroad’s negligence. 45 U. S. C. § 51. The Court expresses concern that the approach Gottshall and Carlisle advocate would require “[jjudges ... to make highly subjective determinations concerning the authenticity of claims for emotional injury, which are far less susceptible to objective medical proof than are their physical counterparts.” Ante, at 552. One solution to this problem — a solution the Court does not explore — would be to require such “objective medical proof” and to exclude, as too insubstantial to count as “injury,” claims lacking this proof.

IV

While recognizing today that emotional distress may qualify as an “injury” compensable under the FELA, the Court rejects the Court of Appeals’ thoughtfully developed and comprehensively explained approach as “inconsistent with the principles embodied in the statute and with relevant common-law doctrine.” Ante, at 535. The Court’s formulation, requiring consistency with both the FELA and “common-law doctrine,” is odd, for there is no unitary common law governing claims for negligent infliction of emotional distress.4 The “common law” of emotional distress *572exists not in the singular, but emphatically in the plural; and while the rule the Court has selected is consistent with one common-law rule that some States have adopted, it is inevitably inconsistent with others.

Most critically, the Court selects a common-law rule perhaps appropriate were the task to choose a law governing the generality of federal tort claims. The “zone” rule the Court selects, however, seems to me inappropriate for a federal statute designed to govern the discrete category of on-the-job injuries sustained by railroad workers. In that domain our charge from Congress is to fashion remedies constantly “liberal,” and appropriately “enlarged to meet changing conditions and changing concepts of industry’s duty toward its workers.” Kernan v. American Dredging Co., 355 U. S., at 432. The Court’s choice does not fit that bill. Instead of the restrictive “zone” test that leaves severely harmed workers remediless, however negligent their employers, the appropriate FELA claim threshold should be keyed to the genuineness and gravity of the worker’s injury.

In my view, the Court of Appeals developed the appropriate FELA common-law approach and correctly applied that approach in these cases. I would therefore affirm the Court of Appeals’ judgments.