skip navigation

Norfolk & Western Railway Co. v. Hiles (95-6), 516 U.S. 400 (1996).
[ Thomas ]
HTML version
WordPerfect version
HTML version
WordPerfect version

NOTICE: This opinion is subject to formal revision before publication in the preliminary print of the United States Reports. Readers are requested to notify the Reporter of Decisions, Supreme Court of the United States, Washington, D.C. 20543, of any typographical or other formal errors, in order that corrections may be made before the preliminary print goes to press.


No. 95-6


on writ of certiorari to the appellate court of illinois, fifth district

[February 27, 1996]

Justice Thomas delivered the opinion of the Court.

Railroad cars in a train are connected by couplers located at both ends of each car. A coupler consists of a knuckle joined to the end of a drawbar, which itself is fastened to a housing mechanism on the car. A knuckle is a clamp that interlocks with its mate, just as two cupped hands--placed palms together with the fingertips pointing in opposite directions--interlock when the fingers are curled. [n.1] When cars come together, the open knuckle on one car engages a closed knuckle on the other car, automatically coupling the cars. The drawbar extends the knuckle out from the end of the car and is designed to pivot in its housing, allowing the knuckled end some lateral play to prevent moving cars from derailing on a curved track. As a consequence of this lateral movement, drawbars may remain off center when cars are uncoupled. This misalignment, if not corrected, may prevent cars from coupling by allowing the knuckles to pass by each other. To ensure proper coupling, railroad employees must realign drawbars manually.

Respondent William J. Hiles was a member of a switching crew at petitioner Norfolk & Western Railway Company's Luther Yard in St. Louis, Missouri. His duties included coupling and uncoupling railroad cars and aligning drawbars. On July 18, 1990, Hiles injured his back while attempting to realign an off center drawbar. Hiles sued in Illinois state court, alleging that Norfolk & Western had violated the SAA, which requires that cars be equipped with "couplers coupling automatically by impact, and capable of being uncoupled, without the necessity of individuals going between the ends of the vehicles." 49 U. S. C. A. §20302(a)(1)(A) (Supp. 1995). Norfolk and Western argued that the misaligned drawbar did not result from defective equipment. The trial court granted Hiles' motion for directed verdict on liability.

The Illinois appellate court affirmed. 268 Ill. App. 3d 561, 644 N. E. 2d 508 (1994). The Illinois Appellate Court recognized a deep split of authority over the proper interpretation of the SAA, but determined that it would not reconsider its "longstanding authority permitting a plaintiff . . . to recover under the Safety Appliance Act for injuries sustained while attempting to align a misaligned drawbar." Id., at 565, 644 N. E. 2d, at 511. The Illinois Supreme Court denied review, and we granted certiorari, 515 U. S. __ (1995), to resolve the conflict among the lower courts. [n.2]

For most of the nineteenth century, the link and pin coupler was the standard coupler used to hook together freight cars. It consisted of a tubelike body that received an oblong link. During coupling, a railworker had to stand between the cars as they came together and guide the link into the coupler pocket. Once the cars were joined, the employee inserted a pin into a hole a few inches from the end of the tube to hold the link in place. See J. White, American Railroad Freight Car 490 (1993) (hereinafter White). The link and pin coupler, though widely used, ultimately proved unsatisfactory because (i) it made a loose connection between the cars with too much give and play; (ii) there was no standard design and train crews often spent hours trying to match pins and links while coupling cars; (iii) links and pins were frequently lost, resulting in substantial replacement costs; and (iv) crew members had to go between moving cars during coupling and were frequently injured and sometimes killed. Id., at 490-497.

In 1873, Eli H. Janney patented a knuckle style coupler that was to become the standard for the freightcar couplers used even today. [n.3] See Figure 1. The coupler had a bifurcated drawhead and a revolving hook, which, when brought in contact with another coupler, would automatically interlock with its mate.

The Janney coupler had several advantages over link and pin couplers. Not only did it alleviate the problem of loose parts that plagued the link and pin coupler, [n.4] it also allowed railworkers to couple and uncouple cars without having to go between the cars to guide the link and set the pin. [n.5] One commentator described the automatic coupling operation as follows:

"While the cars were apart, the brakeman had to make sure the knuckle of the coupler on the waiting car stood in an open position and that the pin had been lifted into its set position. When the opposite coupler was closed and locked in position, the brakeman was able to stand safely out of the way and signal the engineer to move the cars together. When the knuckle of the coupler of the moving car hit the lever arm of the revolving knuckle on the open coupler, it revolved around the locked one, while concurrently the locking pin dropped automatically from its set position into the coupler, locking the knuckle in place. Although the brakeman had to set up the entire situation by hand, the actual locking operation was automatic and did not require the brakeman to stand between the cars." Clark 191.

Though the market was flooded with literally thousands of patented couplers, [n.6] Janney's design was clearly among the best and slowly achieved recognition in the industry. See id., at 193-201. In 1888, the Master Car Builders Association Executive Committee obtained a limited waiver of patent rights--placing much of Janney's design in the public domain--and adopted the design as its standard. Conversion to the new standard proceeded slowly, [n.7] partly as a result of the sheer number of competing designs on the market. The lack of standardized couplers itself caused safety problems, [n.8] and reformers pushed Congress to pass legislation requiring the use of standardized automatic couplers.

In 1893, satisfied that an automatic coupler could meet the demands of commercial railroad operations and, at the same time, be manipulated safely, see Clark 206, Congress passed the Safety Appliance Act. Its success in promoting switchyard safety was stunning. Between 1877 and 1887, approximately 38% of all railworker accidents involved coupling. Id., at 179. That percentage fell as the railroads began to replace link and pin couplers with automatic couplers. The descent accelerated during the SAA's 7 year grace period and by 1902, only two years after the SAA's effective date, coupling accidents constituted only 4% of all employee accidents. In absolute numbers, coupler related accidents dropped from nearly 11,000 in 1892 to just over 2,000 in 1902, even though the number of railroad employees steadily increased during that decade.

Employee Coupling Accidents, 1892-1902[n.9]





Accidents Employee


Accidents Percentage



As originally passed, §2 of the SAA provided:

"[I]t shall be unlawful for any . . . common carrier to haul or permit to be hauled or used on its line any car used in moving interstate traffic not equipped with couplers coupling automatically by impact, and which can be uncoupled[,] without the necessity of men going between the ends of the cars." Act of Mar. 2, 1893, 27 Stat. 531, 45 U.S.C. § 2 (1988 ed.), recodified, as amended, 49 U. S. C. A.§20302(a) (Supp. 1995). [n.10]

The text of §2 requires that rail cars be equipped with automatic couplers and that all couplers be sufficiently compatible so that they will couple on impact. Johnson v. Southern Pacific Co., 196 U.S. 1, 16-17 (1904). The railroad is liable for an employee's injury or death caused by a violation of the SAA. See St. Louis, I. M. & S. R. Co. v. Taylor, 210 U.S. 281, 295 (1908) ("If the railroad . . . use[s] cars which do not comply with the standard, it violates the plain prohibitions of the law, and there arises from that violation the liability to make compensation to one who is injured by it"). [n.11]

Early SAA cases involved injuries that occurred when an employee was forced to go between the cars during coupling operations. See, e. g., Johnson, supra, at 2 (hand crushed between cars during coupling); San Antonio & Aransas Pass R. Co. v. Wagner, 241 U.S. 476, 478 (1916) (foot crushed between couplers); Atlantic City R. Co. v. Parker, 242 U.S. 56, 58 (1916) (arm caught in drawhead between cars during coupling). Our later cases extended the reach of SAA liability beyond injuries occurring between cars during coupling to other injuries caused by the failure of cars to automatically couple. Affolder v. New York, C. & St. L. R. Co., 339 U.S. 96,97 (1950) (railroad employee who ran after a runaway train caused by failure to couple lost a leg when he fell under a car); Carter v. Atlanta & St. Andrews Bay R. Co., 338 U.S. 430, 432-433 (1949) (plaintiff successfully boarded runaway cars that failed to couple, but was injured when the cars collided with another train); O'Donnell v. Elgin, J. & E. R. Co., 338 U.S. 384, 385-386 (1949) (railworker killed when 2 runaway cars--the result of a broken coupler--collided with cars whose couplers he was adjusting). Liability in each of these cases was predicated on the failure of coupling equipment to perform as required by the SAA, and we held that the SAA creates an absolute duty requiring not only that automatic couplers be present, but also that they actually perform. See, e. g., Affolder, supra, at 98; Carter, supra, at 433-434.

Hiles urges that railroads have an absolute duty to outfit their cars with safe equipment and that the SAA is violated if an employee is required to go between the ends of cars to manually adjust a misaligned drawbar. We cannot agree. Hiles correctly points out that failure to perform as required by the SAA is itself an actionable wrong dependent on neither negligence nor proof of a defect, [n.12] see Affolder, supra, at 98-99; O'Donnell, supra, at 390, 393, but the absolute duty to which we have referred on numerous occasions is not breached as a matter of law when a drawbar becomes misaligned during the ordinary course of railroad operations.

In Affolder, the plaintiff was working with a crew coupling cars. The 25th and 26th cars failed to couple and, after a few more cars were added, the first 25 cars began rolling down a slight incline. The plaintiff ran after the runaway cars in an attempt to board and stop them, but instead fell under a car and lost his leg. At trial, the railroad attempted to prove that the coupler at issue was not defective and that the knuckle on the coupler was closed when the coupling attempt was made. Following O'Donnell, we reaffirmed that the failure of equipment to perform as required is sufficient to create SAA liability, Affolder, supra, at 99 (quoting O'Donnell, supra, at 390), but we noted that failure to couple would not create liability if the coupler was not properly set:

"Of course [imposition of failure to perform liability] assumes that the coupler was placed in a position to operate on impact. Thus, if `the failure of these two cars to couple on impact was because the coupler on the Pennsylvania car had not been properly opened,' the railroad had a good defense." 339 U. S., at 99. [n.13]

In Carter, we similarly conditioned the duty on the coupler's being "properly set." 338 U. S., at 434; see O'Donnell, supra, at 394, n. 7 (declining to consider "a situation where an adequate coupler failed to hold because it was improperly set").

In Affolder, we predicated failure to perform liability on placing the coupler "in a position to operate on impact." 339 U. S., at 99. We implicitly recognized that certain preliminary steps, such as ensuring that the knuckle is open, are necessary to proper performance of the coupler and that a failure to couple will not constitute an SAA violation if the railroad can show that the coupler had not been placed in a position to automatically couple. Though Affolder involved a claimed closed knuckle, its language was not so limited and, as a matter of common sense, could not have been. Hiles could not reasonably complain that an otherwise working electrical appliance failed to perform if he had neglected to plug in the power cord. Similarly, a court cannot reasonably find as a matter of law that an otherwise nondefective coupler has failed to perform when the drawbar has not been placed "in a position to operate on impact." We think Affolder's restriction on failure to perform liability logically extends to every step necessary to prepare a nondefective coupler for coupling, see supra, at 5 (describing the ordinary process of preparing for an automatic coupling), including ensuring proper alignment of the drawbar. [n.14]

Hiles contends that the distinction between a closed knuckle and a misaligned drawbar makes a difference because opening a knuckle can be accomplished without going between cars but realigning a drawbar cannot. This is particularly true, Hiles argues, given that Congress' "central policy" in enacting the SAA was to protect the worker by obviating the necessity of going between cars. Brief for Respondent 12-13. We decline to adopt an expansive interpretation of §2 that would prohibit railroad employees from going between cars to realign slued drawbars. The language of §2, which requires couplers that both will couple and can be uncoupled without the necessity of persons going between the cars, does not easily lend itself to Hiles' interpretation. Instead, as even Hiles apparently concedes, see Brief for Respondent 19, the text of §2 only requires railroads to use a particular kind of coupler with certain attributes, and there is no question that Norfolk & Western's cars are equipped with couplers with the necessary functional characteristics. [n.15]

Adopting Hiles' reading of §2 would require us to hold that a misaligned drawbar, by itself, is a violation of the SAA, and we are quite unwilling to do that. It is true that our failure to perform cases made clear that the railroad will be liable for injuries caused by malfunctioning equipment, even when cars are equipped with automatic couplers. But we cannot agree that a misaligned drawbar is, as a matter of law, a malfunctioning drawbar. Historically, misaligned drawbars were an inevitable byproduct of the ability to traverse curved track and, like the closed knuckle in Affolder, are part of the normal course of railroad car operations.

We are understandably hesitant to adopt a reading of §2 that would suggest that almost every railroad car in service for nearly a century has been in violation of the SAA. See United Transportation Union v. Lewis, 711 F. 2d 233, 251, n. 39 (CADC 1983). Our hesitance is augmented by the enforcement scheme Congress enacted with the SAA. From its beginning, §6 of the SAA provided that railroads in violation of §2 were liable for "a penalty of one hundred dollars for each and every such violation." Act of Mar. 2, 1893, 27 Stat. 531, 45 U.S.C. § 6 (1988 ed.), recodified, as amended, 49 U. S. C. A. §21302(a) (Supp. 1995). The amount of the penalty for a §2 violation has varied over the years, [n.16] but the threat of a penalty has not. Yet Hiles points to not a single instance in which a railroad has been fined for misaligned drawbars. It is not the case that the Government has simply neglected to enforce the penalty provisions of the SAA for nearly 100 years. [n.17] We think there is a better explanation than that the Government has failed to enforce this particular aspect of the SAA since its inception: a misaligned drawbar simply is not a violation of §2. [n.18]

Finally, relying on the railroads' experimental attempts to develop automatic realigning devices, Hiles argues that Congress' clear intent to protect railroad employees in coupling operations required the railroads to "develop a mechanism for automatic realignment of a drawbar." Brief for Respondent 27. Or, in the words of his amicus, "[t]he Legislative wisdom of Section 2 is that it is as flexible as technology." Brief for United Transportation Union as Amicus Curiae 17. We reject this argument, for we find no such command in the text of §2. Congress plainly instructed the railroads to install compatible and automatic couplers on all cars, at a time when this basic technology had been in existence for two decades and had received widespread testing and recognition as a feasible technology superior to what was then in primary use. In contrast, Hiles concedes that automatic realignment technology did not even exist in 1893 when Congress passed the SAA, see Brief for Respondent 26-27, and, according to Norfolk & Western, automatic realignment has never been shown to be effective. But this matters not, because Congress legislated working automatic couplers for employee safety, not employee safety by whatever method a court might deem appropriate.

The judgment of the Illinois Appellate Court is



1 See R. Reinhardt, Workin' on the Railroad 274 (1970) (the automatic coupler style "use[s] the principle of the hooked fingers of the human hand").

2 Compare, e. g., Kavorkian v. CSX Transportation, Inc., 33 F. 3d 570 (CA6 1994), Lisek v. Norfolk & Western R. Co., 30 F. 3d 823 (CA7 1994), cert. denied, 513 U. S. ___ (1995), Goedel v. Norfolk & Western R. Co., 13 F. 3d 807 (CA4 1994), Reed v. Philadelphia, Bethlehem & New England R. Co., 939 F. 2d 128 (CA3 1991), and Maldonado v. Missouri Pacific R. Co., 798 F. 2d 764 (CA5 1986), with Coleman v. Burlington Northern, Inc., 681 F. 2d 542 (CA8 1982), and Metcalfe v. Atchison, Topeka & Santa Fe R. Co., 491 F. 2d 892 (CA10 1974).

3 Janney was a dry goods clerk and former Confederate Army officer from Alexandria, Virginia, who used his lunch hours to whittle from wood an alternative to the link and pin coupler. See F. Wilner, Safety: "A great investment," Railway Age, Mar. 1993, p. 53 (hereinafter Wilner).

4 Automatic couplers also made possible the use of power air brakes, which had not been successfully used with link and pin couplers because of excessive slack in the coupling. See Hearings on S. 811 et al. before the Senate Committee on Interstate Commerce, 52d Cong., 1st Sess. (1892), reprinted in, S. Rep. No. 1049, 52d Cong., 1st Sess., 9 (1892) (hereinafter Sen. Hearings).

5 Ezra Miller is generally credited with creating the first semiautomatic coupling device for passenger cars--known as the Miller Hook--but it was never widely used on freight cars. See C. Clark, Development of the Semiautomatic Freight Car Coupler, 1863-1893, 13 Technology and Culture 170, 180-182 (1972) (hereinafter Clark); White 505-506.

6 In 1875, there were more than 900 car coupler patents. White 498. By 1887, the number of coupler patents had topped 4,000, ibid., and by 1900 approximately 8,000 coupler patents had been issued. Clark 179.

7 The Pennsylvania Railroad Company, for instance, adopted a policy of putting automatic couplers on all new cars and on every car that went into the shop for repairs. Sen. Hearings, at 27. In 1890, approximately 10% of all freight cars in use in the United States were equipped with automatic couplers and power brakes. Wilner 54. By 1893, approximately 16% of freight cars were so equipped. Ibid. Witnesses testifying before the Senate Committee in 1892 placed the figure between 12% and 20%. Sen. Hearings, at 12 (12% of cars fitted with Janney style couplers); id., at 27, 42 (20% of cars fitted with mutually interchangeable couplers).

8 The new automatic couplers had design modifications that permitted them to couple with link and pin style couplers, but not easily. See Clark 192 ("[T]he knuckle of the Janney was notched in order to allow the opposing link to enter the drawhead to the point of coupling . . . but in practical service it was most difficult to effect"); S. Rep. 1049, 52d Cong., 1st Sess., 5 (1892) ("These representative men, speaking for thousands of their associates, say that what they desire is uniformity, and that the danger of their calling has increased rather than diminished by the introduction of different types of couplers").

9 Clark 207.

10 In Johnson v. Southern Pacific Co., 196 U.S. 1, 18-19 (1904), we clarified that the statute should be read as though there were a comma after the word "uncoupled," so that the words "without the necessity of men going between the ends of the cars" applies to both coupling and uncoupling. When Congress recodified the SAA in 1994, it placed a comma behind the word "uncoupled." See 49 U. S. C. A. §20302(a)(1)(A) (Supp. 1995).

11 We have held that the Federal Employers' Liability Act (FELA), 45 U.S.C. § 51 et seq., makes railroads liable for a violation of the SAA, see O'Donnell v. Elgin, J. & E. R. Co., 338 U.S. 384, 391 (1949), although early cases, like Johnson, supra, preceded FELA's enactment in 1908. Hiles did not assert a negligence claim under FELA.

12 Hiles neither pleaded nor attempted to prove at trial that Norfolk & Western acted negligently or that the drawbar was defective.

13 Justice Jackson, dissenting on other grounds, agreed: "Before a failure to couple establishes a defective coupler, it must be found that it was properly set so it could couple. If it was not adjusted as such automatic couplers must be, of course the failure is not that of the device." Affolder v. New York, C. & St. L. R. Co., 339 U.S. 96, 101 (1950).

14 Our holding that Affolder's restriction on liability extends to misaligned drawbars suggests that, at least in this case, the absence of a failed coupling attempt is of no consequence. On this record, Hiles would not have been entitled to judgment as a matter of law even if he had been injured during a failed coupling attempt.

15 Hiles reads into the legislative history a singular congressional intent to keep railroad workers from going between cars. Our construction of §2 rests on the text of the statute and our prior interpretations of that language. In any event, we think Hiles' reading of the legislative history is erroneous. For instance, Hiles selectively quotes statements made by W. E. Rodgers during Senate Committee hearings to suggest that Congress wanted to force the railroads to adopt a coupler that would keep railworkers out from between cars altogether. Brief for Respondent 14. A full reading of these statements makes clear, however, that Mr. Rodgers believed that adoption of the Janney design, as it then existed, would fully satisfy the requirements of §2. Sen. Hearings 14. Hiles' reliance on statements made by H. S. Haines, vice president of the American Railway Association, is similarly misplaced. See Brief for Respondent 15. Mr. Haines evidently thought that 500 existing couplers would satisfy the requirements of the proposed bill. Sen. Hearings, at 41; see Hearings on Automatic Couplers and Power Brakes before the House Committee on Interstate and Foreign Commerce, 52d Cong., 1st Sess., 6 (1892). If Congress had any singular purpose in enacting §2, it was to require the railroads to equip cars with uniformly compatible automatic couplers that employees could operate without having to go between the cars. See H. R. Rep. No. 1678, 52d Cong., 1st Sess., 3 (1892) ("It is the judgment of this committee that all cars and locomotives should be equipped with automatic couplers, obviating the necessity of the men going between the cars"); S. Rep. No. 1049, supra n. 8, at 6 (1892) ("What the railroad employés need to secure greater safety in the performance of their duties is uniformity. They want all couplers alike and perfectly interchangeable"). We think Congress fairly intended to prohibit the practice of placing railworkers between moving cars to guide a link into its matching coupler pocket or, worse, into an unfamiliar coupler cavity. Cf. Hearings on Automatic Couplers and Power Brakes before the House Committee on Interstate and Foreign Commerce, 52d Cong., 1st Sess., 11 (1892) ("He goes in to make a coupling. He does not know the conditions that exist there. He can not tell whether it is a Janney or a Hinson, a Dowling, a Drexel, or some other kind of a drawbar"). Contrary to Hiles' assertion, the legislative history contains no suggestion that Congress intended to prevent an employee from going between cars to ensure that the knuckle is open, that the locking pin is set, see supra, at 5, or that the drawbar is aligned.

16 The statute currently requires the Secretary of Transportation to impose a penalty of "at least $500 but not more than $10,000." 49 U. S. C. A. §21302(a)(2) (Supp. 1995).

17 See, e.g., Chicago, B. & Q. R. Co. v. United States, 220 U.S. 559 (1911) (§1 power brake violations); Alabama Great Southern R. Co. v. United States, 233 F. 2d 520 (CA5 1956) (§2 coupler violation); United States v. St. Louis San Francisco R. Co., 271 F. Supp. 212 (Mo. 1967) (§9 brake violation); United States v. Gulf, M. & O. R. Co., 76 F. Supp. 289 (La. 1948) (§2 coupler violation).

18 Hiles' view of §2 also conflicts with regulations promulgated by the Federal Railroad Administration (FRA) that provide for the safety of employees who go between cars to "prepare rail cars for coupling," 49 CFR § 218.22(c)(5) (1994), or to "adjust a coupling device," §218.39(a). In its proposed rulemaking for §218.39, the FRA explained that the proposed rule would protect employees who place themselves between cars to couple air hoses or adjust coupling devices, including "adjusting drawbars." 48 Fed. Reg. 45272, 45273 (1983).