No. 81-1405.United States Court of Appeals, Fifth Circuit.
December 6, 1982.
Page 433
Arnold Porter, Dennis G. Lyons, Charles H. Cochran, Washington, D.C., for plaintiffs-appellants.
Richard B. Vance, Asst. U.S. Atty., Fort Worth, Tex., Mark C. Rutzick, Dept. of Justice, Washington, D.C., for defendants-appellees.
Lawrence M. Mann, Washington, D.C., for intervenor-appellee.
Appeal from the United States District Court for the Northern District of Texas.
Before GEE and JOHNSON, Circuit Judges, and VAN PELT,[*]
District Judge.
GEE, Circuit Judge:
[1] In the fall of 1970, Congress enacted the Railway Safety Act, a comprehensive regulatory scheme granting broad rule-making authority in the interest of safety to the Secretary of Transportation.[1] Among the Act’s provisions is that codified as 45 U.S.C. § 438(a), reading in pertinent part:[2] The essential issue on this appeal is whether the Secretary is authorized by the above language to do as he has recently done: promulgate regulations dispensing with all considerations of knowledge or care (scienter) on the part of the railroads and imposing civil penalties on the basis of strict liability. The parties are agreed that the terms “disregard” and “disobey,” used in the statute, incorporate elements of volition and hence of knowledge, so that if the statute contained only these terms, regulations imposing a strict liability standard would be unauthorized.[3] The battle rages over the meaning of “fail to adhere.” The court below, in a painstaking and carefully considered opinion, concluded, as stated in its own summary of its holdings, thatIt shall be unlawful for any railroad to disobey, disregard, or fail to adhere to any rule . . . prescribed by the Secretary under this subchapter.[2]
[3] Fort Worth Denver Ry. Co. v. Goldschmidt, 518 F. Supp. 121, 134 (N.D.Tex. 1981). The district court follows this analysis by noting that, in its view, the statute authorizes the adoption of regulations “that impose a standard of strict liability or a lesser standard of liability.” [4] The question presented is one of pure law that we address free of the trammels of the “clearly erroneous” rule. It is, moreover, an extremely doubtful one in light of the ambiguous statutory language: Congress, having jettisoned the original language of the bill which became Section 438, language that clearly authorized a strict liability standard,[4] instead saw fit to adopt other language that possesses no accepted legal meaning.[5] Instead of adopting tested language for its standard, Congress settled on the term “adhere,” a verb to which six discrete meanings are assigned by Webster’s Unabridged, thus endorsing the struggle between rail employers and unions to us, without recourse. One aspect of the Congressional motive in doing so seems all too plain; it is, however, irrelevant to our task. What standard of liability the Congress meant to authorize by use of the formulation “fail to adhere” is the relevant question, a study in obscurity. The issue is close, but on balance we conclude that the scale tips against a strict liability standard. We therefore reverse. [5] The legislative history being doubtful, we follow the counsel of Justice Frankfurter, and go to the statute.[6] We reiterate its language for the reader’s convenience:Although there is neither an expressed statement nor any absolutely clear indication from Congress that it intended to authorize a standard of strict liability, the Court does find the following:
1. The plain meaning of the phrase “fail to adhere,” although ambiguous as used in the context of § 209(a), would support an interpretation that a scienter element is not required for violations of regulations adopted by the FRA [Federal Railroad Administrator, the Secretary’s delegate].
2. Since the official committee reports are silent as to the standard of liability, it is likely that Congress intended to continuePage 434
the long-standing policy of a standard of strict liability for violations of railroad safety legislation.
3. The FRA did adopt, prior to the 1979 revised regulations, emergency orders and regulations that did not require a scienter element; therefore the FRA has interpreted § 209(a) as authorizing a standard of liability that does not require a scienter element. Moreover, the FRA, by adopting the revised regulations, interprets § 209(a) as authorizing strict liability regulations, and the Court gives great deference to that interpretation.
Therefore, based upon these factors and the foregoing review of the indicia of Congressional intent, the Court finds that § 209(a) of the FRSA does authorize the FRA to adopt regulations which impose a standard of strict liability.
[6] Several things are at once apparent from its structure. One, unremarked by the parties or the court below, is that the form adopted is such as to itself lay down the standard for determining whether a regulation prescribed by the Secretary has been breached, as opposed to one authorizing the Secretary to promulgate his own standard (or standards) of judgment. Another, likewise unremarked, is that in terms the standard stated — “disregard, disobey, or fail to adhere to” — applies to “anyIt shall be unlawful for any railroad to disobey, disregard or fail to adhere to any rule, regulation, order, or standard prescribed by the Secretary under this subchapter.
rule . . . prescribed by the Secretary under this subchapter.”[7] This language and structure seem to suggest, then, that in enacting Section
Page 435
438 the Congress believed that it was itself laying down a standard of liability that applied to “any” (and, therefore, in context to every) rule or regulation that the Secretary might promulgate under the Railroad Safety Act. We therefore disagree with the district court’s conclusion that the section delegates to the Secretary power to pick and choose among standards of liability, applying one to one regulation and a different one to another; whatever standard (or standards) of liability Congress meant to specify applies to every rule or regulation prescribed under the Act.[8]
[7] But if Congress intended a unitary standard, of universal application, why did it excise S. 1933’s original and clear specification of absolute liability and present us with the Cerberus that we now contemplate? Several speculations are possible. One, advanced by the proponents of strict liability, is that a succession of standards, each more severe than the previous, was stated. Nor is this suggestion implausible: “disobey” denotes a knowing and intentional refusal to comply; “disregard” denotes the same, but connotes a neglect or failure to attend to; and “fail to adhere to” can mean, among other things, simply a failure to comply with or conform to — though its primary meaning is volitional: to fail “to hold, follow, or maintain loyalty steadily and consistently (as to a person, group, principle or way).”[9] Thus, the argument runs, Congress decreed liability for one who violates a prescribed regulation in any of three ways: by deliberate disobedience, by negligence, or by simple failure to conform. Such a construction possesses a pleasing symmetry and brings order to the verbal shambles in which we struggle. Unfortunately, however, it has little else to recommend it, while several considerations militate against it. [8] In the first place, if our threshold observation is correct and these three standards apply to every regulation promulgated under the Act, then on this construction the first two are mere surplusage: if one is to be rendered liable by a mere failure to comply, then notions of disobedience or neglect are, in context, beside the point. All three applying to every promulgation, none signifies except the most severe: strict liability, which comprises the other two and more besides. Second, if Congress intended such a result, why did it reject the original word-of-art language of S. 1933, which clearly specified that result and none other? Next, such a Chinese-box construction, a somewhat curious one, finds no more warrant in the legislative history than does any other; had such an unusual meaning been intended, one would expect a hint of it to have been dropped at some point. And finally, if strict liability is to be the universal rule, why did not Congress repeal the existing statutes specifying it in particular instances, committing a clear field to the Secretary’s discretionary regulation?[10] [9] Of the other construction, that pressed upon us by the railroads, we can say no more than that it is the only reasonably probable alternative to that we have just treated and seems slightly less unlikely. This is that the terms of the congressionalPage 436
triad — disobey, disregard, fail to adhere to — are simply a partially redundant statement of the same concept: knew or should have known, that of willfulness or negligence. Supporting our adoption of this construction are most of those factors that we have suggested above as militating against the other, though not that of avoiding ascribing surplusage to the legislator or (so strongly) that of failure to utilize readily available terms of art.[11] And to them we are able to add a few others.
[10] First, even the proponents of strict liability concede, as they must, that the primary dictionary meaning of all three terms of the triad incorporates the element of volition. It is only the secondary meaning of “adhere”[12] that can be read as nonvolitional: “To be consistent or in accord.” The primary meaning is “to . . . follow or maintain loyalty steadily and consistently” — in other words, purposefully and consistently to obey and, conversely, never to disobey. The most accessible meaning of the term “adhere,” therefore, incorporates an element of volition consistent with the scienter standard, and inconsistent with strict liability. [11] Second, the hoary canon of ejusdem generis holds out some slight comfort. Of it we have said, as appellant reminds us,[12] As terms describing knowing and intentional acts, “disobey” and “disregard” are more specific than “fail to adhere to,” which may carry either sense, volitional or nonvolitional. By Webster’s, “ADHERE is a general term somewhat more bookish than STICK to indicate any holding to, esp. steadily and over a period of time . . . .” Thus, by its terms, the canon applies and suggests a construction of “adhere” that is consonant with the more specific terms with which it is brigaded. This, the volitional meaning, is also its primary dictionary meaning. Finally, our decision today is consistent with the original and contemporaneous construction of the Act, adhered to by the Secretary for seven years, which with certain exceptions imposed a scienter standard of liability. [13] All of the above is pretty thin stuff — jejune, conjectural and artificial. We regret deeply that we must rest a determination of nationwide moment on such a shaky foundation: a “least-unlikely” basis. It is, however, not open to us to remain in doubt; we must decide. Facing language of consummate ambiguity and deprived of proper tools of construction by a legislative history all but meaningless, we apply such tools as we possess. We hold that the standard of liability is “known or should have known.” If in this we are mistaken, it may be that, given the weight of the matters at stake, Congress or the Court will tell us so. [14] REVERSED.“In divining legislative intent, however, a venerable precept of statutory construction, the doctrine of ejusdem generis, warns against expansively interpreting broad language which immediately follows narrow and specific terms. To the contrary, this maxim of statutory analysis counsels courts to construe the broad in light of the narrow, in a commonsense recognition that general and specific words, when present together, are associated with and take color from each other.” United States v. Insco, 496 F.2d 204, 206 (5th Cir. 1974).
Page 437
a standard of strict liability on the railroads.[2] Faced with a long-standing policy of a strict liability standard in the field of railroad safety legislation, Congress chose the language “disobey, disregard, or fail to adhere”[3] for the 1970 Act. Given the history of over seventy years of a strict liability standard, had Congress intended to change the standard, can it be doubted that it would have given some clue, some intimation, some inkling of a possible change in the standard. Yet there is not one vestige from legislative history that Congress had even the slightest intent to change the three score and ten strict liability standard. Indeed, if change had been the intent of Congress, certain affected groups would have been exceedingly vocal. From the very beginning, labor itself had been involved in the drafting of the legislation by virtue of its participation on the task force on rail safety. The task force which drafted the “administration bill” first used the language “disobey, disregard, or fail to adhere.”[4] If the task force, by this language, had intended to change the strict liability standard, surely labor would have given some indication of concern and uttered some word of protest. Labor’s active participation in the enactment of this bill, yet silence on this score, is indicative of the fact that there was never any intent on the part of Congress to abolish the old standard.
[17] In light of labor’s lack of protest, other aspects of legislative history, although admittedly ambiguous indicators of congressionalPage 438
intent, portend and support an interpretation of the statutory language which continues a strict liability standard. First, in isolating the areas of disagreement between members of the task force, the Administrator of the Federal Railroad Administration (FRA) in a letter to the Secretary of Transportation stated that management opposed the draft bill because it did not limit penalties to “knowing and willful” violations. Although this statement could simply imply that management wanted “actual knowledge” as the standard of liability, it could very well indicate — and this writer is convinced does indicate — that management interpreted the “fail to adhere” language as allowing a standard of strict liability. Other statements, made at hearings before the Subcommittee on Surface Transportation and relating to penalty levels, included references to the standard of liability. Referring to the difference in penalties between S. 1933 and the administration bill, the then Secretary of Transportation John Volpe stated: “This is clearly an area of judgment. It is my judgment that, in a safety statute such as this, where a violation invokes the penalty without regard to the knowledge or willful action of the violator, a minimum penalty of $500 is simply too high.” (emphasis added). Federal Railroad Safety Act of 1969: Hearings on S. 1933, S. 1915, and S. 3061, Before the Subcomm. on Surface Transportation of the Senate Commerce Comm., 91st Cong., 1st Sess. 340 (1969). In proposing a technical amendment, later rejected by the Senate, Thomas Goodfellow, president of the Association of American Railroads, expressed his concern about the penalty provision whereby each day of violation constituted a separate offense: “In theory, at least, the provision could lead to immense fines in circumstances where (1) a railroad had neither knowledge nor notice of a violation; or (2) where the violation, though known, cannot be corrected immediately.” (emphasis added). Id. at 363. Again, these witnesses could have been referring to circumstances where a railroad had neither actual knowledge nor actual notice of a violation but should have known about the violation, i.e., to circumstances where a railroad had constructive knowledge. Viewed in the context of congressional silence on the standard of liability, it seems far more likely — indeed to a level of certainty — that the witnesses objected to the penalty level because the proposed law imposed liability, as had its predecessor laws, without regard to negligence.
[18] Interpreting the language “disobey, disregard, or fail to adhere” as permitting a strict liability standard has more to recommend it than a historical or continuity factor. The language of the statute itself suggests this interpretation. Since the majority has dissected the syntax of the twenty-seven word sentence,[5] this writer will also make some syntactical observations. The statutory language at issue sets forth only one mandate: the railroads must comply with whatever regulations the Secretary prescribes, i.e., the railroads cannot fail to comply with any regulation whatsoever. The negative usage “fail to adhere” easily suggests a nonvolitional meaning, a simple failure to comply. If “adhere” were used in a positive sense, that is, if the statute read “the railroads must obey, observe and adhere to any regulation,” then a volitional meaning would be appropriate. Simply stated, there is a difference in connotation between “adhere” and “fail to adhere.” [19] Furthermore, a “succession of standards” construction[6] is not only allowed, but even suggested by the syntax. Under this construction, the sequence “disobey,” “disregard,” “fail to adhere” represents an escalating standard of liability: liability for deliberate disobedience, liability for negligence,Page 439
and liability for a mere failure to comply with or conform. The Secretary is thereby authorized to promulgate regulations with any of these three standards of liability. Far from viewing a “succession of standards”[7] interpretation as unusual, this writer is convinced that such an interpretation is altogether logical. Prior railroad safety statutes, which imposed a strict liability standard, dealt with specific safety problems e.g., brake systems or coupling devices. In contrast, the Railroad Safety Act dealt generally with all areas of railroad safety, providing that the Secretary of Transportation was to prescribe appropriate rules, regulations, and standards as necessary. Given the fact that the RSA was general rather than specific, it seems logical, not unusual, that Congress would have used language which would allow the Secretary to select the standard of liability appropriate for the particular regulations he was issuing. Furthermore, surely the rule of ejusdem generis
does not militate against a “succession of standards” interpretation. This rule of statutory construction is inapplicable unless broad language immediately follows narrow and specific terms. As the district court correctly noted, “By no stretch of the meaning of the words `disobey’ and `disregard’ could they be said to be `specific’ nor could the meaning of `fail to adhere’ be said to be `general.'”[8] Reliance o ejusdem generis as an aid to construing the statutory language at issue is misplaced.
Page 440
do look at administrative interpretation of the Act between 1973 and December 1979, though, it is apparent that the FRA did not consistently interpret the language at issue as authorizing only regulations with a scienter standard. Indeed, the FRA had issued various emergency orders and some regulations without a scienter element.[10] Surely, if “fail to adhere” authorizes strict liability emergency orders, it also allows strict liability regulations.[11]
[21] In view of the statutory language and the long history of prior railroad safety legislation, this writer remains convinced that the Act permits the Secretary to promulgate either scienter or nonscienter regulations. Prior laws gave the Secretary broad discretion. The legislative history of the 1970 Act gives no indication — no possible hint — of any intent on the part of Congress to change that discretion. Surely, if that had been congressional intent, there would have been an outcry from certain groups. No such outcry was made; not one voice of protest was heard. This writer would uphold the judgment of the district court that the language in section 209(a) of the Act (“fail to adhere”) authorizes the FRA to adopt regulations which impose a standard of strict liability.But this Court early swept all issues of negligence out of cases under the Safety Appliance Act. For reasons set forth at length in our books, the Court held that a failure of equipment to perform as required by the Safety Appliance Act is in itself an actionable wrong, in no way dependent upon negligence and for the proximate results of which there is liability — a liability that cannot be escaped by proof of care or diligence. These rigorous holdings were more recently epitomized by Chief Justice Hughes, speaking for the Court: “The statutory liability is not based upon the carrier’s negligence. The duty imposed is an absolute one, and the carrier is not excused by any showing of care, however assiduous.”
(citations omitted).
Earlier, in Brady v. Terminal R.R. Ass’n, 303 U.S. 10, 58 S.Ct. 426, 429, 82 L.Ed. 614 (1938), this Court had defined the carrier’s obligation as an absolute one which was triggered by a mere “failure to comply”:
This final question must be determined in the light of the nature of the obligation resting upon the carrier in relation to the use of a defective car. The statutory liability is not based upon the carrier’s negligence. The duty imposed is an absolute one, and the carrier is not excused by any showing of care, however assiduous. The breadth of the statutory requirements is shown by the fact that it embraces all locomotives, cars, and similar vehicles used on any railway that is a highway of interstate commerce and is not confined exclusively to vehicles engaged in such commerce. Laying down this comprehensive rule as a matter of public policy, Congress has made no exception of those employed in inspecting cars. The statute has been liberally construed “so as to give a right of recovery for every injury the proximate cause of which was a failure to comply with a requirement of the Act.” In Davis v. Wolfe,
reviewing the earlier cases, the Court held that one can recover “if the failure to comply with the requirements of the Act is a proximate cause of the accident resulting in injury to him while in the discharge of his duty, although not engaged in an operation in which the safety appliances are specifically designed to furnish him protection.”
(citations omitted).
§ 438. Penalties
(a) Unlawful acts. It shall be unlawful for any railroad to disobey, disregard, or fail to adhere to any rule, regulation, order, or standard prescribed by the Secretary under this title.
45 U.S.C. § 438(a).
(1980), intimates that such a construction might raise serious constitutional questions of overbroad delegation. This writer does not view Industrial Union as barring the delegation at issue in the instant case — a delegation to the agency of the authority to promulgate regulations with any one of three standards of liability.
For seven years, between the time of its promulgation of the original regulations under the FRSA in 1973 and the time of its promulgation of the amended regulations here in issue in December 1979, the FRA adhered to and enforced a liability policy based on a “knew or should have known” standard. Plaintiffs argue that this first interpretation of the standard of liability authorized by the FRSA is entitled to deference on judicial review. Moreover, plaintiffs argue that this initial interpretation is entitled to great weight because it was a contemporaneous construction made soon after the time of enactment and such interpretation has been consistently applied for a seven year period. Plaintiffs conclude by implying that based upon these [sic] set of circumstances, the Court should reject the 1979 interpretation and accept the 1973 interpretation.
Plaintiffs’ argument is a misleading interpretation of the law. The proper starting point is the 1979 regulations. It is these regulations which are entitled to “great deference.” Whether the 1979 regulations are given “peculiar” weight depends on whether they are the result of a contemporaneous construction or whether they have been consistently applied. Therefore, a correct statement of the law is that agency regulations which interpret an enabling statute deserve great deference by a reviewing court, and such interpretation is given a peculiar weight when the regulations are adopted contemporaneously with enactment of the enabling statute or where the agency has given a consistent interpretation of the enabling statute.
Surely the law is not what plaintiffs imply. Agencies are not forbidden to revamp, revise or even reject prior regulations, when it has been determined that the prior regulations are no longer responsive to the problems that the agency is statutorially [sic] directed to remedy. The Court has to assume that the FRA changed its regulatory scheme, not based on some whim, but because the FRA, relying on its expertise in the field of railroad safety, determined that a new regulatory scheme would be more efficient in meeting the goal of the FRSA. It is an anomalous argument to say that a regulatory scheme’s shortcomings should be used against an agency when it tries to rectify those shortcomings. This Court has to assume that the FRA relied on its expertise to determine that strict liability regulations would be more beneficial in promoting railway safety than “knew or should have known” regulations.
Therefore, in adopting regulations which impose a strict liability standard, the FRA interpreted the FRSA as authorizing such regulations. It is this interpretation which the Court is bound to give great deference. There is no case holding that a lesser weight or no weight should be afforded such an interpretation, merely because it is not consistent with prior regulations which simply were not adequate.
Moreover, the 1973 and 1980 regulations are not inconsistent. The FRA did not say in 1973 that it was adopting regulations to the fullest extent as authorized by the FRSA. At most it could be implied that the FRA determined that a “knew or should have known” standard was more appropriate in that early stage of development of the FRSA. If plaintiffs’ argument is carried to the extreme, every time an agency decides not to use its full statutory power, such a decision, based on conditions then present, would forever forbid the agency from exercising its full power in the event changed conditions so warrant.
Obviously the FRA cannot adopt strict liability regulations, if doing so exceeds its statutory authority. The Court only holds that merely because the FRA decided to change its regulatory scheme, cannot be used as an indication that the FRA is now exceeding its statutory authority.