Courting Danger

The risky business of liability law

Some call it a safety tax, but its exact relationship to safety is mysterious. It is paid on many items that are risky to use, like ski lifts and hedge trimmers, but it weighs even more heavily on other items whose whole purpose is to make life safer. It adds only a few cents to a pack of cigarettes, but it adds more to the price of a football helmet than the cost of making it. The tax falls especially hard on prescription drugs, doctors, surgeons, and all things medical.

The tax has closed, or limited the use of, city parks and school diving and gymnastics programs. Because of the tax, you can no longer buy many American-made brands of sporting goods. It has curtailed Little League and fireworks displays, evening concerts, sailboard races, and the use of public beaches and ice-skating rinks. It temporarily shut down the famed Cyclone at the Astroland amusement park on Coney Island.

The tax goes by the name of personal injury, or "tort," liability. It is one of the most ubiquitous taxes we pay, now levied on virtually everything we buy, sell, and use. It directly costs American individuals, businesses, and government bodies at least $80 billion a year.

It is collected and disbursed through litigation. The courts alone decide just who will pay, how much, and on what timetable. Unlike better-known taxes, this one was never put to a legislature or a public referendum, debated at any length in the usual public arenas, or approved by the president or by any state governor. And although the tax ostensibly is collected for the public benefit, lawyers and other middlemen pocket more than half the take.

For all practical purposes, the liability tax is a comparatively recent invention. Under legal rules that prevailed in this country for centuries, and for centuries before that in the English common law from which we inherited ours, courts viewed the barkeeper who sold Baker a beer very differently from the drunk who took a swing at him. The civil law was in fact carefully divided into two distinct territories. Contract law governed consensual dealings, and tort law governed uninvited encounters.

Most accidents used to be covered by contract, not by tort. Aside from collisions on the road, mishaps between strangers are relatively rare. Workplace accidents are preceded by a contract between willing employer and employee; surgical accidents by a consensual doctor-patient relationship; airplane crashes by the voluntary purchase of a ticket. And the common sense of an earlier jurisprudence suggested that the contingency of an accident should be addressed ahead of time, when tempers are cool and minds clear. Buyer and seller were allowed to set the terms of their own deal, on questions of safety just as on questions of price. The courts viewed consumers as intelligent adults, whose freely made contractual commitments were to be respected and enforced as bargained for.

The law today could not be more different. Contracts are no longer sacrosanct; indeed, in matters of health and safety, contract terms count for nothing at all. The intervening revolution has been as profound and far-reaching as any in the history of the common law.

The shadowy beginnings of the revolution can be traced back to the early 1900s. But the real action occurred much later, in the 1950s and early 1960s. It was then that judges across the country set out to rewrite the civil law of accidents, chapter and verse. The movement was led by a handful of thoughtful, well-intentioned legal academics at some of the most prestigious law schools and by judges on the most respected state benches.

Standardized contracts between a large corporate seller and a small consumer, these new tort scholars reasoned, are not really agreements at all; they are mere contracts "of adhesion" and should be dismissed out of hand. Joe Consumer never bothers to read the contract accompanying his airline or bus ticket, his toaster, or his car; even if he did, General Motors, General Electric, and United Airlines would laugh if Joe proposed to renegotiate the terms. So the consumer came to be viewed much as a fly near flypaper, ever in danger of sudden death by glue.

The flypaper theory of contracts was infinitely powerful. Take-it-or-leave-it contracts typify virtually all mass-market sales of consumer products and services. Used skillfully, the theory allowed judges to get rid of almost any unwanted language they chose, anywhere they might find it. Adhesion theories could transform any deal, no matter how seemingly consensual, into the legal equivalent of a car accident.

The death of contract created a daunting new challenge to the courts. "It took but a few seconds to cut off Lavoisier's head," one of his countrymen remarked in 1794, when the great French chemist Antoine Lavoisier was led to the Paris guillotine. "It will take France a century to grow another like it." The judges who had sentenced and executed the law of contract had not one head to replace, but a million.

Every sale of a car, toaster, or cigarette lighter, every hookup of an electric or gas customer, every ticket for an ocean cruise or roller-coaster ride, every college admission, hotel room rental, or tonsillectomy, is based on contract. The old law depended on buyer and seller to agree privately beforehand who would pay for what if an accident befell. If they failed to set out the terms of their deal with sufficient precision, the law would of course supply presumptions of one sort or another in the case of subsequent dispute. But what was critically important was that contract could prevail if the parties wished.

All of this was to disappear. When the courts abandoned contract, they left a cavernous void that had to be refilled, bucket by bucket, with a new set of public standards, rights, and responsibilities. The legal theorists set about rewriting liability law, with California leading the way. Traditionally, a seller's liability for goods or services was modest and limited. Fault, in some sense of that word, was a precondition to any successful liability claim—malicious intent, or at the very least negligence. But in a 1962 landmark decision, the California state supreme court declared that the new rule for product manufacturers would be "strict liability"—liability for any and all accidents caused by a "defect in manufacture" of their product.

Within a few years, California and other courts across the country had extended this notion to "design defects." This empowered judges and juries to decide questions of engineering and pharmacology, medicine and science, far beyond their experience or expertise. Just how does one go about locating a defect in a complex product? Sometimes the job is easy. Defects in manufacture are immediately apparent when we compare the car without a critical bolt in the steering column to hundreds of others that came off the same assembly line with it. In effect, the mass manufacturer establishes his own standards, by which any one of his products can be gauged. Manufacturing defect cases are straightforward. They are also comparatively rare. Far more difficult are cases in which the product is said to be defective in design, where there is no such simple point of comparison. Fully 80 percent of product liability cases today are of this kind.

How is a jury to decide whether a whole class of products—say, the intrauterine device (IUD) contraceptive—is inherently defective? The new tort system has apparently reached the conclusion that it is, having driven from the market not only the notorious Dalkon Shield but also its far safer substitutes, the Copper-7 and the Lippes Loop IUDs. The Food and Drug Administration, Planned Parenthood, and the vast majority of doctors do not endorse that sweeping verdict, but the verdict stands, nonetheless.

It is not enough to identify a safety shortcoming; the jury must also weigh the cost of remedying it. In the early 1970s, the Ford Pinto was to car-crashworthiness cases what the Dalkon Shield later became to contraceptive designs. The Pinto weighed under 2,000 pounds and cost less than $2,000. Ford's own tests revealed that its gas tank was vulnerable to rear-end collisions, but the company decided not to spend an extra $10 per car to reinforce the structure—a calculation on which plaintiffs' lawyers subsequently grew very rich. But did they really deserve to?

Ten dollars is not much, but full-force rear-end collisions aren't common, and there are innumerable equally rare hazards that could also be averted for $10 or thereabouts. Protecting against every one of them would cost thousands. But people with thousands to spare don't buy a Pinto, they buy a Mercedes. What about other cars in Pinto's class? Some certainly had safer gas tanks, but that is not to say they were safer cars. A jury later fined Honda $5 million for its "reckless" act of using lighter-gauge materials than some other manufacturers in a 1971-model vehicle. (The driver of that car admitted he had bought it for its economy.) Toyota lost a $3-million judgment on similar grounds.

The subtle message here may be that all economy cars are inherently defective for tort purposes. But millions of consumers, all major car companies, and the government's safety regulators seem to view the matter quite differently.

With lawn mowers, kitchen appliances, airplanes, and safety valves, the conclusion is almost always the same: safety is no exception to the golden rule that buyers can pay more and get more. Design is an infinitely variable and subtle process. It is always possible to strengthen an airplane wing or a column in a building; it is always possible to reduce the dosage of a drug or change the method or timing of its administration. But the follow-up questions are the difficult ones. There are questions of function: Will the plane still fly? Will it fly as fast? There are questions of cost: at what point is an incremental benefit in the car's safety no longer worth the price increase it would entail? There are questions of safety itself: has the therapeutic drug really been improved, or has one risk just been traded off for another, possibly a more serious one?

All that is certain, in modern design defect lawsuits, is that any result is possible. Today, a product can be legally "defective" because a safety device has been omitted (a paydozer without rear-view mirrors) or because certain parts are not as strong as they might have been (a car roof is not strong enough to withstand the impact of a runaway horse that lands on the roof after a front-end collision). A jury can find that a single-control shower faucet is defective because if one turns it on all the way to one side, it will allow only hot water to spray, or that children's cotton sleepwear is defective because it has no flame-retardant chemicals added. Juries across the country somberly weigh the designs of lawn mowers, electrical switches, bottles, pesticides, contraceptive pills, vaccines, morning sickness drugs, and intrauterine devices.

The courts thus slipped gently from enforcing private agreements to redesigning products. But the courts soon discovered that condemning "defects" in the abstract is very much easier than identifying them in the particular case. The law journals published endless, impenetrable articles, by people with no training whatsoever in science, engineering, or pharmacology, on the nature of product "defects." The increasingly amorphous new legal standards attracted litigants, lawyers, and expert witnesses and inflated product liability trials beyond all prior recognition.

Strange verdicts became commonplace. The FDA recommends the Sabin polio vaccine over the Salk, but juries have issued verdicts—including punitive awards—on the strength of the opposite conclusion. In the face of overwhelming scientific evidence indicating no such link, a jury concluded that a spermicide, Nonoxynol-9, caused birth defects in a child born after her mother had stopped using the product. In a 1978 decision, the California supreme court decided that recognizing design defects was so difficult that the task should be simplified (for plaintiffs, at least) by reversing the burden of proof, leaving it to defendants to prove (if they can) that their products are not defective.

The disarray in the system was hardly surprising. The people who dreamed up this new occupation for the courts didn't know the first thing about designing a car, a contraceptive, or an airplane engine. And unlike the consumer who would weigh risks and benefits ahead of time, they lacked even the discipline that comes when one is spending one's own money and risking one's own life and limb.

The inventors of the new tort law should have been the least surprised by the mounting randomness and incoherence of jury outcomes. A central tenet of their faith was that the consumer is unqualified to make intelligent choices about safety for himself. Private choice, in short, was to be replaced by public prescription. Whose public prescription? Not that of the pharmacologists at the FDA, or toxicologists at the Environmental Protection Agency, or mechanical engineers at the Federal Aviation Administration, but that of the juror, pulled off the voter lists at random, solemnly sworn to his duty, and instantly educated in a contest of courtroom experts also solemnly sworn, of course, and paid by the hour for their particular form of swearing. The member of the public judged incompetent to make wise choices in the marketplace for himself was now being called upon to make wise choices in the jury box for others.

The authors of these bizarre rules sincerely believed, nevertheless, that they would advance safety by adopting them. They focused single-mindedly on the demand side of things: if the liability courts inflicted enough verdicts on people—especially well-insured people—found near the scene of an accident, accidents would decrease in frequency and severity.

It looked simple enough on paper. Accidents are socially costly, and the law should encourage accident prevention by the most economical means. To that end, liability should fall not on the person most "at fault" in some traditional sense, but on the one who could prevent any given accident at the lowest cost. And just who might that be? Most often, those who provide goods and services, not those who consume them. Corporations, hospitals, city governments, universities, and other institutions, unlike the individual consumer they serve, are accident experts. They could compile accident reports and discover recurring safety problems that individual users might never suspect. They could warn consumers of likely hazards, redesign products, or keep dangerous items out of the hands of irresponsible buyers. Even when there was no immediately feasible way to eliminate a hazard, billing producers for accidents would spur them to search constantly for improvements in their wares.

By a happy coincidence, ultra-stringent producer liability would further other desirable goals as well. It would ensure compensation for injured parties who can ill afford to shoulder the costs of accidents. And very strict rules would also be simpler to administer because they would sweep aside distracting questions of individual right and wrong, care or carelessness, innocence or fault. By creating a new demand for safety, in short, the courts would ensure that life did indeed get safer.

But there was a critical oversight in all this. The supply side of the safety picture was not seriously considered at all; there was just an easy presumption that suppliers would remain on the scene but conduct themselves more carefully.

What has happened in practice? The immediate impact of the new legal rules has been a marked increase in price and a decline in the availability of a wide range of goods and services. If businesses know they are going to be sued, the costs of the lawsuits, or the insurance to cover them, become routine costs of production. Producers either find ways to incorporate those costs in the final sale, or they go out of business.

That much was expected. What was unexpected was the propensity of the new tort system to fall where it is least needed and most difficult to bear. Hospital emergency-room services are perilous in liability terms because emergency-room patients are in trouble to begin with. Vaccines are now risky (again, from the legal perspective) because the children who receive them are susceptible to a host of diseases and reactions often indistinguishable from vaccine side-effects. Running a municipal police department, ambulance service, town dump, or waste cleanup service invites litigation, because these activities are aimed at situations that are inherently risky.

How could a tort system so committed to increasing safety have landed some of its first punches on the very persons who work on the front line of helping others? Why does this system so often fail to distinguish risks that are part of the problem from risks that are part of the solution? The answers are complex. For one thing, juries have often (and quite understandably) proved unskilled at distinguishing the various parties found at the scene of the crime. Juries are too prone to arrest the firefighter, so to speak, along with the arsonist, the ambulance driver along with the drunk who made the ambulance necessary in the first place.

Another part of the reason lies in a slip between theoretical cup and real-world lip. The founders of the new jurisprudence were committed to deterring hazardous practices. But judges and juries were, for the most part, committed to running a generous sort of charity. If the new tort system cannot find a careless defendant after an accident, it will often settle for a merely wealthy one. But the wealthy defendant is more often part of the safety solution than part of the safety problem.

The larger fallacy in the grand scheme of the new liability system was more profound. As our right to sue the butcher, brewer, and baker after the sale has grown, our freedom to make the purchase in the first place has declined. The purveyors of meat, beer, and such withdraw only partially, by demanding a higher price; the purveyors of rare drugs, Yellowstone hiking, and rural obstetrical services have often been driven from the market altogether. An unbounded and impossible-to-waive right to sue necessarily overtakes and destroys the right to make deals with people who place a high premium on staying in business and out of court. While the consumer has indeed acquired a new and sometimes valuable right to sue, he has done so only by surrendering an older right, the right to contract, which in the long run is worth far more.

Who is it that flees the modern liability system most quickly? Those quickest on their feet, of course—the person of action, the company of initiative, the mover, the shaker, and the doer. When it comes to liability, the bold innovators are the most fleet-footed of potential defendants. More often than not, they adjusted to the expanding threat of liability by doing less. In the very markets where the legal pursuit was the most intense—on the trail of exotic drugs, contraceptives, pesticides, small planes and cars, hazardous waste disposal, and medical procedures—the mood among suppliers became most sullen, hostile, defensive, and then coldly stagnant. Soon tired of running, the fox retreated to its burrow and refused to come out.

Research expenditures by U.S. companies working on contraceptives, for example, peaked in 1973 and plummeted 90 percent in the next decade. Steroidal oral contraceptives in this country underwent no significant changes after 1976, and no truly new contraceptive chemical entities have been introduced since 1968. A new and effective IUD, the Copper-T 380A, won FDA approval, but no major firm was willing to market it for several years. "Who in his right mind," the president of a major pharmaceutical company asked in 1986, "would work on a product today that would be used by pregnant women?"

The story has been much the same in other high-tech markets favored with attention from the liability system in recent years. Between 1965 and 1985, the number of U.S. vaccine manufacturers shrank by more than half; by 1986 the nation depended on a single supplier for vaccines against polio, rubella, measles, mumps, and rabies. In the 1960s there were eight U.S. manufacturers of whooping cough vaccine; by 1986 there were only two. And only two major companies, Merck and Lederle Labs, were still investing heavily in vaccine research. America, once the world leader in this technology so vital to the public health, was quickly losing ground here, too.

Consulting engineers report that they systematically favor old products over new ones in their design specifications, fearing—quite correctly—that newer design options carry a greater risk of liability, whatever real decrease in risk they might actually represent. Liability-conscious universities decline to license patents to small companies, despite the fertile environment they offer for innovation, fearing that anyone suing over a patent-related product would be sure to go for the university's deep pocket, as well. Monsanto decided in 1987 not to market a promising new filler and insulator made of calcium sodium metaphosphate. The material is almost certainly safer than asbestos, which it could help replace in brakes and gaskets. But safer is not good enough in today's climate of infectious litigation.

The demise of the prescription drug Bendectin marked progress of a similar kind. Bendectin had been used for 27 years, in 33 million pregnancies, to relieve morning sickness. The overwhelming scientific consensus, in the FDA and in all respectable scientific circles, has not moved an inch: Bendectin does not cause birth defects. But claims against the drug's manufacturer nevertheless burgeoned in the 1970s. In 1985, under the mounting pressure of litigation, the drug was pulled from the market. If you suffer from morning sickness, go see your lawyer. Your doctor has nothing left to prescribe.

Obstetricians and gynecologists have retreated, too. Some of their services have become wholly unavailable in rural and less-affluent communities. So women and children won their expanded right to sue but lost services essential to their health and safety. The complication that develops in the back seat of the car heading to the delivery room at a distant hospital is not likely to yield a munificent tort award, even though it is more likely to end in a tragedy.

Next in line on the slope toward less liability and less safety is the do-it-yourselfer. By 1985, the courts were levying a tax on stepladders amounting to over 30 percent of their cost. Economic theory tells us that some consumers will then curtail their use of these hazardous instruments. But there remains the vexing problem of the light bulb and the kitchen chair. Light bulbs still go out. People would still rather replace the bulb than curse the darkness. And if they cannot afford a stepladder they use a kitchen chair, which is more dangerous. Without ever quite saying so, the creators of modern liability law had assumed that everything would work out in the end, because kitchen-chair makers would be sued, too, presumably even more often. But even with all the excesses of the new law, light-bulb changers who fall off do not generally think to sue kitchen-chair makers, and if they do they are likely to lose.

In theory, expanded liability was supposed to deter only "defective" products and such. But practice turned out rather differently. When it encourages improvement at all, the new tort system promotes the trivial and marginal change. Today's oral contraceptive manufacturer does work hard to fine-tune the warning, or microscopically adjust the hormonal balance in the pill. The drill press designer adds an extra hair guard or hand shield. The doctor administers more tests, shoots more X-rays, and piles on a paper trail of his own. Large companies hire risk managers, industrial hygienists, consumer psychologists, and quality control experts in droves. The individuals and institutions all vigorously insist that they are working relentlessly to improve safety and reduce their liability.

What one finds, however, is that liability-driven safety management has become a mirror image of the legal process itself—fussy, cumulative, bureaucratic, and preoccupied with paper. The risk-reduction initiatives that are encouraged by liability undoubtedly do some good some of the time. But the threat of liability also postpones or prevents the sharp break from tradition, the profound change in method or material, design, or manufacture. And the sharp break, the occasional bold leap forward, is all-important in the quest for safety.

In net effect, the modern rules do not deter risk: they deter behavior that gets people sued, which is not at all the same thing. A victim's consent to be left alone is much more likely to be treated as binding than his consent to accept help, so doing nothing is safer legally than doing something, even if more dangerous in all other ways. Consent aside, juries are far more prone to find fault with something that was done and went wrong, than with something that was not done, though it might have gone right. And liability, for quite obvious reasons, gravitates toward wealthy targets. The good thing about suing the wealthy is that they can afford to pay; the bad thing is that people and corporations most often become wealthy not because they do things that are wicked and dangerous but because they do things that are valuable and necessary.

"We are here on earth to help others," W.H. Auden once remarked, "but what the others are here for I cannot say." Today's liability lawyers seem to be of much the same mind. They are not going to spend a great deal of time worrying about just what drug manufacturers might or might not contribute to the larger scheme of public health and welfare; the lawyers' business is to see to it that people who cause accidents pay and people who suffer accidents get paid, and that is that.

Modern tort law is unchangeably pessimistic about the consumer's competence and unduly optimistic about the prospects for universal institutional care. Courts and juries, the law assumes, must operate for the consumer in loco parentis, now and forevermore. This attitude robs us of our most important economic freedom, the freedom to plan in advance, to make commitments, and to arrange deals on terms mutually agreeable to the parties involved. Applied as it has been in recent years, open-ended tort law serves only as an engine of social destruction. The freedom of contract is undermined, private bilateral deals are curtailed. Modern tort law abrogates our freedom to cooperate.

Worst of all, the courts are now openly at war with human individuality, preference, and variety. The new legal theories adopt a one-size-fits-all theory of safety and product defects. A small car is either "defective" or it is not; it does not matter that a particular buyer may not be able to afford any other car and deliberately chooses the compact as best for her needs and circumstances. In its single-minded assumption of uniformity, not just among products but among consumers themselves, the new tort jurisprudence is a compulsive homogenizer.

And that, in the most subtle but far-reaching way, undercuts safety at every level. Safety ultimately lies in distinguishing good risks from bad ones and keeping them carefully apart in both thought and action, not across the board but in proper context. Lumping people, products, and processes together is dangerous; safety lies in tailoring choices to individual needs and circumstances, individual capabilities and responsibilities, on both the consumer's side and the supplier's.

Not all consumers are alike; the prescription that is needlessly unsafe for one may save the life of another. In all but the simplest cases, accidents originate not in defective designs but in the unwise conjunction of a particular design with a particular use and user. IUDs, for example, are an inappropriate form of contraception for younger women who have the self-discipline to use other forms of protection correctly, but they are often the safest option for older women who smoke, among others. Oral contraceptives are safer for the former group and riskier for the latter. Botulinum is a paralytic poison, except when it is a sight-saving therapy; penicillin is a miracle drug against infection, except when it is a fatal allergenic; thalidomide a fearsome teratogen, except when it alleviates the misery of leprosy. Pure water is necessary for life but can trigger seizures in epileptics when consumed in excess. Similar user-specific risks and benefits inhere in every power tool, food additive, transportation system, or medical therapy.

It is precisely because we are not all equally tough or delicate, cautious or careless, hardy or sickly, thin-skinned, or thick-skulled that safety lies in individuation and differentiation. Safety is not improved by treating individuals as cookie-cutter copies of each other or as interchangeable parts in a mass-production assembly line. Safety depends on treating individuals as such, not as undifferentiated members of a faceless public. But the new tort system insists otherwise. The law itself has become hazardous to the public health.

What's to be done? How do we encourage broader and more reliable accident insurance coverage? How can we rediscover a legal framework that permits us, individually or collectively, to make the affirmative choices that are needed if we are to end up with better drugs, contraceptives, and vaccines, new and improved medical procedures, small planes, football helmets, or amusement parks? The guiding principle must be to promote consent and agreement wherever possible—to prefer contract over tort whenever the choice may be presented—and to elevate direct insurance over liability-driven compensation.

Direct insurance, from whatever source, should always provide first-dollar coverage for accidents. Before Congress intervened with expensive and convoluted legislation two years ago, drug makers were prepared to offer a fixed-sum insurance policy, capped at $1 million per injury, with every dose of childhood vaccine they sold—provided that this advance commitment would be truly binding and exclusive, with questions of cause and effect referred to outside experts. But most providers remain unwilling to offer any such direct insurance, and for good reason. Under current rules, most courts will find a way to take the guaranteed money up front and then use the contemporary law of torts to demand more.

Insurance policies might not pay the astronomical amounts awarded by juries. But at the same time, direct insurance need not depend on costly inquiries into negligence, defect, or fault, or the flypaper theories of contract currently necessary to underwrite the rickety Mutual Tort Insurance Company.

Insurance questions aside, we must categorically reject the notion of contract as flypaper, the idea that to offer one's goods and services to others is to entrap them. Whatever lawyers may tell us, we each do have it in our power to say no to the cigarette, the cut-rate lawn mower, the contraceptive, and the ride in the propeller aircraft. The theory of adhesive contracts notwithstanding, consumers in a free society are not flies. Of course, we cannot bargain one-on-one with United Airlines or General Motors. But we can do what amounts to the same thing by choosing among different companies, different modes of transportation, and different insurance packages that may be willingly offered by competing companies. Time and again, deals that could protect health, cure disease, or improve safety—not to mention increase the stock of human happiness in other ways—go unconsummated because the law will not allow them to be settled on terms both sides would gladly accept.

The old law erred in the other extreme, elevating freedom of contract to the point of enforcing deals that were grounded on blind ignorance all around. Few would suggest returning to the day when a warning had to be particularly sought out and paid for, and the unwary were simply out of luck. The answer is to reanchor the law of warning to its contractual roots by applying the modern rules symmetrically, with a firm sense of balance. Full disclosure should be an ordinary and essential part of fair dealing. When it is—as it commonly is today—a deal should be treated, once again, as a deal.

The courts may have to sort out the wreckage that results from uninformed choice. But informed choice, deliberately made outside the courtroom and before the accident, should count for everything. Not because the choices will always be wise—they surely won't—but because respecting the right to choose, with the responsibility that necessarily goes with it, will advance overall safety more often than it retards it.

The most urgently needed correction here concerns warnings prescribed by regulatory agencies like the FDA and the EPA. We should have one inviolable safe harbor from liability. If the agency has weighed the question of warning, and spelled out the language to be used in detail, that should be a certain end to the matter for tort purposes.

A return to contract, built on open warning and informed consent, would permit once again infinite calibration to the varying needs of individual consumers. No liability system, and no system of broader regulation, can ever do better; rarely can they do anywhere near as well. And the individuality and flexibility that contract allows is the ultimate key to the consumer's safety.

The path to a safer existence lies as much in beatifying good risks as in exorcising bad ones. It lies in the replacement of risks that are old and grave with risks that are new and less grave. The answer lies in choice, in the affirmative needed to counterbalance every negative. Aaron Wildavsky has made the point with customary eloquence. Conceiving of safety without risk, or progress without change, or freedom without choice, is like seeking love without courting the danger of rejection.

Peter W. Huber, who has degrees in mechanical engineering and law, is a senior fellow at the Manhattan Institute in New York City. This article is adapted from Liability: The Legal Revolution and Its Consequences, copyright © 1988 by Peter W. Huber, by permission of Basic Books.