Concept
Custom Is Not Controlling
Custom as evidence of the standard of care
Custom Is Not Controlling
A proved custom enters a negligence case as evidence, not as the rule of decision. Section 295A of the Restatement (Second) of Torts says that the customs of the community, or of others under like circumstances, are factors to be taken into account in deciding whether conduct is negligent, “but are not controlling where a reasonable man would not follow them.”1 The earlier parts of this chapter showed how a party proves a custom and why a proved custom is relevant. This part limits what the fact-finder may do with it once the proof is in. Even a custom that is well defined, followed in the defendant’s own calling, and aimed at the risk that caused the harm goes onto the scale as one weight among several.
The limit follows from what the negligence standard measures. Where no enactment, regulation, or decision fixes the standard of conduct, the judge or jury applies the reasonable-person standard to the facts of the case.2 Custom helps with that task because it reports how other actors have handled the same risk, and a report of practice can be wrong. If practice were itself the standard, the actors under review would write the test that judges them. Treating custom as evidence leaves the jury free to ask whether the practice measures up to reasonable prudence.
The principle cuts in two directions. In one, the defendant proves it did what others in its calling do and asks to be excused. In the other, the plaintiff proves that others took a precaution the defendant omitted and asks the jury to infer fault. Each direction has its own logic, and in both custom informs the jury without binding it.
Following Custom Is Not Enough
A defendant who proves it did what everyone in its calling does has offered evidence of due care, and no more. Conformity to an applicable customary standard of care, like violation of one, is some evidence of due care but is rarely conclusive, and that is now the universal rule in negligence law.3 The jury remains the final arbiter of whether a reasonably prudent person would have done more or less than was customary.4
The rule was not always so. Courts in the late nineteenth century trusted the free market to determine the socially optimal level of safety precaution. The Great Depression shook that faith, and Judge Hand and other judges were no longer willing to give the marketplace carte blanche to set the standard of care in negligence cases.5 The worry is easy to see. A practice can continue long after thoughtful analysis would compel its rejection, whether to save costs, out of inertia, or from tradition, and a new precaution can be ignored despite its obvious benefits.6 A standard fixed by such practice would carry its lag into the law.
Judge Learned Hand stated the modern rule in its classic form in T.J. Hooper (2d Cir. 1932).7 The defendant’s tugs were towing the plaintiff’s barges, loaded with coal, when a gale off the New Jersey coast caught them and the barges were lost. The tugs carried no radio receiver that would have brought a Weather Bureau warning in time to seek shelter.8 The owners proved that it was the custom of barge lines not to supply receiving sets to their crews, and the verdict against them stood anyway.9 Hand wrote that “in most cases reasonable prudence is in fact common prudence; but strictly it is never its measure; a whole calling may have unduly lagged in the adoption of new and available devices.” A calling, he added, “may never set its own tests, however persuasive be its usages,” and “[c]ourts must in the end say what is required.”10
The passage makes three moves. Common prudence usually is reasonable prudence, so proof of custom belongs in the case. Common prudence is never the measure, because a whole calling may have lagged. And the court, not the calling, decides what is required. The tug owners’ proof explained why their tugs had no receivers; it could not settle whether reasonable prudence required them.
Hand added that some precautions are so imperative that even their universal disregard will not excuse their omission.11 Mayhew v. Sullivan Mining Co. shows that point at its starkest. A mining company left the ladder-holes in its working platforms unguarded and unlit, and a workman fell through one.12 The court said that even if every mining establishment since the days of Tubal Cain had cut such holes without guarding or lighting them, the practice would have “no tendency to show that the act was consistent with ordinary prudence.”13 The two cases mark different strengths of one principle. In Hooper, the custom counted as evidence and lost the weighing, because a calling had lagged in adopting a new and available device. In Mayhew, the court gave the practice no weight at all, because leaving an open hole in a floor where people work, unguarded and unlit, is careless whoever else does it.
Products cases apply the same limit. In Hillrichs v. Avco Corp., cornpicker manufacturers customarily omitted a $50 emergency stop device, and the custom was no bar to a negligent design claim.14 The label “state of the art” has been defined in widely varying ways, but the proposition beneath it, that manufacturers should be held responsible only for what they can foresee and reasonably prevent, is now widely accepted.15 A showing that no competitor sold the safer design proves a custom. It does not show that the harm was unforeseeable or that the manufacturer could not reasonably have prevented it.
A suit against a landlord over a glass shower enclosure, Trimarco, shows what a jury does with a proved custom. Once the jury credits a custom’s existence, the court said, the custom is not necessarily a conclusive or even a compelling test of negligence, because the jury must be satisfied that it is reasonable, just as it must be satisfied of the reasonableness of conduct that followed it or the unreasonableness of conduct that did not.16 Customs and usages run the gamut of merit like everything else, and the court quoted Holmes: “[w]hat usually is done may be evidence of what ought to be done, but what ought to be done is fixed by a standard of reasonable prudence, whether it usually is complied with or not.”17 The court also left to the jury whether, by the time of the accident, the modest cost and ready availability of safety glass and the growing custom of using it for shower enclosures had made a door once thought reasonably safe one that no longer could be so regarded.18 A custom in motion can therefore work against a defendant who kept to the older practice, because the jury measures the defendant against conditions when the injury occurred.
Holmes wrote that line in Texas & Pacific Railway Co. v. Behymer, and the facts there show how conditions can make following a custom unreasonable.19 The railway made up trains by bumping cars together with brakemen standing on top of them, and the occasion in suit was one when the cars were covered with a sheet of ice.20 A practice that works on an ordinary day does not carry its own justification to a day when the footing is gone.
A custom’s weight depends on where it came from. When a roofer falls from a two-story house he was reshingling, it bears on his own negligence that roofers ordinarily do not wear safety harnesses on such jobs, since the trade accepts the risk in exchange for freedom of movement and time saved. If most roofers regard that as an acceptable risk, they may well be right.21 The custom comes from people who face the risk and bear its costs, so conformity to it can support a finding of reasonable care. The answer shifts if an enterprising roofer develops an easy-snap, no-hassle harness. The reasonable roofer might then use it, and even if it is hard to teach old roofers new tricks, following the older custom under changed circumstances may be unreasonable.22 A custom that records a real balance of cost and benefit deserves weight, and one that records cost-saving, inertia, or a failure to adopt an available device invites the jury to reject it. Neither kind is conclusive.
Departing From Custom Is Not Fault
The second clause of § 295A works from the plaintiff’s side as well: custom is a factor, and it does not control where a reasonable person would not follow it.24 A defendant’s violation of an applicable customary standard, like its conformity to one, is some evidence about whether it exercised due care, though it is rarely conclusive.25 If every other maker builds a certain safety switch into its presses, the absence of that switch from the defendant’s press tends to show the omission was unreasonable. The jury remains the final arbiter of whether a reasonably prudent person would have done more or less than was customary.26
Departure falls short of fault because the plaintiff’s proof must clear two checks, and the jury applies both. The first asks whether a general custom exists at all. In Trimarco, the trial judge told the jurors that “the mere fact that another person or landlord may have used a better or safer practice does not establish a standard,” and that it was for them “to determine whether or not the evidence in this case does establish a general custom or practice”; the court of last resort held the charge correct.27 Without that check, the most cautious member of a field would set the legal minimum for everyone else.
The Trimarco plaintiff did not rest on one safer landlord. Bulletins of nationally recognized safety and consumer organizations and official federal publications had warned of the dangers of plain glass in hazardous locations, including bathtub enclosures. The defendants’ managing agent, long familiar with the management of multiple dwellings in New York City, agreed that since at least 1965 landlords customarily used plastic or safety glass for shower enclosures.28 One better practice proves a better practice. A pattern of warnings and an admission from the defendants’ own agent prove a custom.
The second check asks whether the custom, once credited, is reasonable under the conditions the defendant actually faced. The jury’s duty to be satisfied with a custom’s reasonableness cuts both ways, because it applies to conduct that followed the custom and to conduct that did not.29 A departure made because following the custom would have been more dangerous in the circumstances at hand is an exercise of care, not a breach. An unexplained departure and an explained one therefore call for different treatment.
A driver who leaves the right-hand lane of a private road, where no traffic statute applies, shows both checks at work. The custom of keeping right is conceded in both versions below, so the only question is whether it fits the road the driver was on.
Only the reason for the departure changed the answer. On the quarry road, the premise of the custom, that the right lane is the safer one, is false, so a reasonable person would not follow it, and § 295A says custom does not control in that case.30 On the campus road, the premise holds and the driver offers nothing to displace it, so the jury may treat the departure as evidence of carelessness. Even there, the custom is one factor beside everything else the jury hears about the collision. The conditions that made the usual way unreasonable in Behymer can make the unusual way reasonable here.
Commercial defendants meet the same weighing. A plaintiff who proves that a manufacturer omitted a safeguard others customarily use will usually add specific evidence on the foreseeability and magnitude of the risk, its hidden nature, and the feasibility, utility, and cost-effectiveness of avoiding it.31 The manufacturer answers with the opposite showing: that the risk was remote, open and obvious, and impracticable and costly to remove.32 Custom supplements these basic proofs of how and why the defendant was or was not negligent.33 A defendant with no reason to give has nothing to set against the departure, though the jury still decides. A defendant with a real reason has handed the jury a possible justification.