The problem of statistical injury
It is a battery to intentionally touch someone in a harmful or offensive way. If I handed you an exploding soda bottle, that would be a battery. If I poured sulfuric acid into your lungs, that would be a battery. If I ran my car into you, that would be a battery.
If I do something that is "substantially certain" to touch you in a harmful or injurious way, that's also a battery: we assume that people intend the inevitable results of their actions. So if a modern day Abraham pulls the trigger on his gun, which is pointing at his son, in order to prove the intensity of his faith in the goodness of God and his willingness to obey His commands, and moreover with absolute confidence and faith that an angel will catch the bullet before it reaches its target -- if the angel fails to show up, that is battery. It is battery even if the fact finder is convinced that Abraham had no desire, conscious or unconscious, to hurt Isaac. So long as Abraham intended to pull the trigger, and so long as pulling the trigger is "substantially certain" to result in the bullet hitting him, his faith is irrelevant.
On the other hand, it is not a battery, but mere negligence, to unreasonably increase the risk that someone will be injured. If I drive too fast on a windy road and run into you coming around a blind turn, that is negligent, but it is not a battery: I had no intent to harm you.
However, in the modern world and with modern understandings of risk, the distinction between these two scenarios often cannot be maintained. The problem is that even highly unlikely events become near certainties if the underlying act is repeated often enough. Winning the Megabucks lottery is highly unlikely. However, it would be a little strange to say that the designer of the Megabucks lottery created only a risk of winning: however unlikely it may be that you will win, it is certain that someone will. The odds of any one person being a winner may be one in several millions. The odds of anyone being a winner are just one.
What is true of lotteries that you'd like to win is equally true of lotteries that you'd prefer to lose. If I drive my car fast, the odds are pretty good that I'll be lucky enough not to hurt anyone; all I'm doing is increasing various people's risks. But if I run a fleet of 10,000 taxicabs 24 hours a day for years on end, all of them driving at risky speeds, it is a certainty that I will repeatedly injure people. If I make soda bottles that do not explode 99.99% of the time, the odds are extremely good that I won't injure any particular person with any particular bottle. But if I manufacture ten million bottles a year, those odds suggest that 1000 or so will explode each year -- it is substantially certain that I will injure a great many people over the life of my factory. If my factory emits sulfur dioxide as a by-product of burning coal, and sulfur dioxide predictably and regularly turns into sulfuric acid in the atmosphere, then it is entirely predictable that I am exposing many people to sulfuric acid in their lungs and on their skin. It is certain -- not merely probable -- that if I emit enough sulfur dioxide over a long enough period, someone will be injured.
Moreover, a modern enterprise, if it is properly run, knows the odds. You can't run the Megabucks lottery without knowing how often people will win on average -- otherwise, you wouldn't know how much to charge for tickets so as to have enough to pay for the prize and the costs of running the thing and have enough left over to fund education. Similarly, you shouldn't be running a modern factory without keeping careful records of how often your product fails and in which ways. If you don't know whether or how often you are injuring people, you can't make the product safer, fix the problems in production, know what your guarantee is going to cost you or even know whether in the end you are going to make a profit selling it. Exploding soda bottles are not good for future business; factories that kill their employees are likely to have to pay higher wages, or at least pay more for recruiting and training; blockbuster drugs that cause occasional heart attacks are going to attract controversy. Even if you don't keep these records, your insurance company is going to want them -- unless they know how often you are creating problems, they have no way of determining how much to charge you for insuring them.
So any modern enterprise is likely to know, in advance, that it is going to injure people. Care isn't the point: few economic activities can be made completely risk free. Even if the company is very careful, by the time it takes the risk millions of times, it is substantially certain to have bad luck sometimes. Moreover, competent enterprises will have a pretty good idea of how many they are going to injure and how badly: it is the careful, well-run, enterprises that will test their products, keep track of returns and problems and side-effects. In the course of improving the product and correcting existing problems they will inevitably learn exactly how many injuries they are causing.
In short, the decision to mass produce -- more, to engage in any modern, repetitive, institutionalized practice -- is a decision to injure. That is an intentional injury in any ordinary sense of the word intentional. The enterprise knows it will injure and makes a formal, conscious, decision to proceed nonetheless. What could be more intentional than that?
Now, I don't mean that the decision is wrong or that a company which (or executive who) decides to produce a useful product even knowing that it inevitably hurt someone, somewhere, sometime has necessarily done something bad. To say that no one should ever intentionally injure someone in this way would be to condemn all modern economic activity, which would mean approving of the starvation and misery that would result in its absence. It's hard to imagine anyone seriously taking that position. For many activities, the injuries will be balanced by the social benefits: I doubt that we, collectively, would want to give up soda in order to save the associated explosions, let alone cars to eliminate auto accidents (although we'd probably be happy to have safer bottles or cars).
Nor do I mean to condemn cost-benefit analysis. It is far better than the alternative: to intentionally engage in a risky activity that is certain to injure people without knowing what the odds are. To put people at risk without caring enough to find out what the risk is, or to think about whether the risk is worthwhile, shows an extraordinarily solipsistic disregard for fellow human beings. That is, in effect, a judgment that they don’t matter at all. Cost-benefit analysis often seems cruel and uncaring – especially when physical injuries are given low dollar values – but it at least reflects an acknowledgment that the injuries exist and must be given weight in decision-making. Cost-benefit analyses are a way of expressing concern for the risks, and associated injuries, that an enterprise inevitably causes.
Normally, one assumes, the injuries are an unwanted side-effect, not the point, of the act. So followers of the doctrine of double effect might conclude that a corporation that makes a useful product, knowing that inevitably it will injure someone, might not be morally responsible for the inevitable consequences of its action. But tort law has not traditionally followed the doctrine of double effect, and for good reason: a company that chooses not to use a safer production technique or to produce a safer product should not be forgiven simply because it intended to make a useful product, any more than it should be let off simply because, on balance, its product did more good than harm. The issue for a decent (and efficient) economy always must be whether it did the most good possible at the least cost, and injuries are always a cost.
In the end, virtually every modern enterprise engaged in repetitive activities is intentionally injuring people. Even if the risk of any particular act injuring a particular person is infintesimal, if you repeat a bet often enough, eventually you will lose, however overwhelming the odds.
The law of battery ought to confront this reality rather than evade it. These are intentional injuries and they ought to compensable as such. Under the standard definitions of battery, they are battery: intentional, harmful touchings. Alternatively, if the association of civil battery with criminal battery is so strong that it is impossible to call these intentional injuries battery without also condemning them, than tort needs a new category: intentional infliction of statistical injury. Regardless of the label, the victims are entitled to be made whole, and the tort-feasors must be required to pay for the damage they have caused.
Even, indeed especially, if this taking is for the social good, there is no justification for private individuals having to subsidize the enterprise or its consumers. From the perspective of the victims, their health or property has been taken from them without their consent. That is a form of theft, and they are morally entitled to compensation. From the perspective of the tort-feasor, the costs of the injuries are costs of production just as much, and in exactly the same sense, as the costs of wages paid to employees or the cost of raw materials used in creating the product. Micro-economic pricing decisions will be fallaciously low if producers are allowed to externalize them, leading to socially inefficient over-production of the injurious product. Just as it would be unfair and inefficient to allow producers to simply steal raw materials or force employees to work without compensation, it is would be both unjust and inefficient to allow an enterprise to knowingly, intentionally injure people or property without compensating them.
Punitive damages, however, are a separate issue. Ordinarily, we can assume that the injuries enterprises cause are unwanted, regrettable, side-effects of attempts to perform a socially useful service or create a socially useful product. The very fact that the enterprise makes itself aware of the injuries that it causes, willingly compensates its victims, includes the cost of that compensation in its prices, and is still able to sell its product for a profit is some evidence that, on balance, the injuries are a reasonable price to pay for the benefits society gains. We need to compensate, not to punish -- to replace a failed market (much as quasi-contract doctrines do), not to ban the behavior. Thus, compensation should be set at the closest approximation we can of the price at which the victim would have willingly contracted. If the enterprise can continue to make a profit while thus paying its full costs, it has proven its social utility.
On the other hand, sometimes an enterprise will demonstrate extreme moral callousness. Failing to test a products or failing to collect the information that would be necessary to determine whether more testing or other changes are needed, prima facie demonstrate a lack of concern for the injuries that the enterprise will cause or the people who will suffer them. Unless some other explanation appears, the failure to keep records or to perform a cost-benefit analysis seems to demonstrate a simple willingness to injure without regret. That's a form of moral depravity that might warrant punitive damages: organizations that injure without concern, much like individuals who hurt without guilt, are a special danger to society and ought to be stopped.
Finally, a note on current law. Ordinarily, courts treat these cases under the doctrine of negligence. In exceptional cases, they may apply the law of strict liability for product failure.
The former leads to an odd set of questions and very odd results. Under negligence law, the question is whether the actor has acted with the care that a reasonable person would under the circumstances. This leads to the truly bizarre result that if a company has performed its cost-benefit analysis and concluded that the product is socially beneficial – or even if it didn’t, but the court, ex post, reaches that conclusion – it is then allowed to expropriate the persons and property of private individuals freely. Under the famous Hand formula, for example, once a firm demonstrates that it could have paid for the injuries that it caused, it has demonstrated that its actions were reasonable – and, vindicated, it is not required to make the payments it could have made! If we applied this rule to ordinary conversion, I’d be allowed to take your possessions freely so long as I could prove, after the fact, that if asked I’d have been willing to pay for them. This result is simply indefensible.
Strict liability leads to more sensible results. But courts are reluctant to apply it, in part because its language is peculiar and unconvincing and, again, focuses the fact finder on the wrong questions. Why should anyone be “strictly liable” without fault? Why should an intentional injury be deemed “without fault”? If liability is “strict,” what happens to causation, often a difficult and potentially controversial issue in economic cases: it is rarely self-evident that one enterprise, rather than another or the individual tort-victim, is the sole cause of any given injury.
--
More subtly, the legal issue can be understood as a slightly off-kilter way of discussing a difficult political problem. Tort law sharply distinguishes between compensable behavior and non-compensable bad luck. If you are struck by lightening, you are unfortunate, but tort law offers you no remedy. Instead, you must look to your own private resources or to social welfare safety nets. If, in contrast, someone deliberately or carelessly injures you or your property, tort law insists that they make you whole – restore you to the status quo ex ante. Controversies arise when people differ as to whether an accident is just bad luck, or more properly seen as the result of someone’s action.
Often, the controversy can be understood as a result of differing base lines. Tort remedies small wrongs – where an identifiable person or institution has made someone worse off than they were prior to an identifiable act. It has nothing to say about people who are just badly off, or even people who become worse off as a result of a more general change in the world or specific events that are not the consequence of an identifiable act by an identifiable actor.
When people, or political views, differ on the background conditions, they will differ on whether particular injuries are compensable in tort. Thus, everyone agrees that an act of God is not compensable in tort. But when a building collapses in an earthquake, views will differ as to causation and fault. On one view, the injury may seem clearly the result of the earthquake – non-compensable bad luck. After all, if the earthquake hadn’t happened, no one would have been injured. Taking existing buildings and patterns of life as given and viewing the earthquake as the change, injuries due to the earthquake are just bad luck. While a society may wish to help out the injured, tort – which concerns itself with compensation, not charity – has no place in the effort.
But earthquakes only cause injuries under specific social circumstances. In particular, building methods may be more or less earthquake-resistant. When earthquakes hit grasslands inhabited by tent-dwelling nomads, it is rare that anyone is injured. Similarly, modern highrises, if built to earthquake resistant standards, are usually quite safe; injuries in cities built to modern earthquake building codes are usually quite rare and often involve pre-code buildings. Thus, in the last large San Francisco earthquake, the actual fatalities were almost entirely the result of a In contrast, cities built of brick, unreinforced masonry or concrete without consideration of earthquake danger have massively larger numbers of collapses, injuries and fatalities. Think of the death and destruction in Mexico City’s 19– earthquake as compared with the relatively minor damage to San Francisco a few years later – itself concentrated in a small number of collapses of structures that were not built to current standards: an apartment building built on fill, and several sections of highway and bridges.
On this view, the relevant variable is not earthquakes – they are background, a fact of life – but the building codes and the builders’ compliance with them. The injuries in Mexico City were the result of mid-rise buildings made of unreinforced concrete that collapsed in the tremor. From this perspective, the issue is first and foremost one of tort. Builders built in a fashion that used less labor, steel and design skill. One of the costs of this type of building method is potential collapse in earthquakes. Earthquake damage is, therefore, a cost of this method of building; injuries should be charged to the building. Within the traditional doctrinal categories, the injury was caused by the builder’s lack of reasonable care. If the tort system does not force builders to pay for the injuries they cause, the result will be that random victims of collapse will be forced to subsidize building, builders and their customers, without consent or compensation. That is precisely the type of injury to which tort should be responsive.
To generalize, many tort controversies stem from differing views of the status of existing social structures. If existing social structures are taken as given, either because they are beyond human control or because they are so important that they are beyond political, economic or legal discussion, then many accidents will be seen as just bad luck. If cars are necessary parts of modern life, and cars are inherently dangerous and inherently polluting, then victims of car accidents or pollution are simply victims of bad luck. To be sure, they’ve been injured, but no one was at fault, no one caused the injury, no one is morally to blame for the injury, no one had any obligation to change their behavior to prevent the injury or should be expected to do anything to prevent similar ones in the future, and the injury is not the sort of cost that should be included in price.
On a narrow view, many accidents will seem to be mere bad luck. Earthquakes, car accidents, factory injuries, sinking tug boats, spilt coffee, flooded cities can all be understood as the result of victims who were hit by lightening – suffered the inevitable slings and arrows of misfortune – or who did not exercise sufficient care themselves. At most, the alleged tort-feasor failed to prevent an injury, and tort never requires one to prevent an injury to a stranger.
But those who see human fate as at least partly under human control, or view the social structure as changeable by political, legal or economic forces, will reject this understanding. If we can imagine a different way of running the factory, or designing the tug boat, or serving the coffee or building or protecting the city, then the decision to do it in a way that does not prevent injuries is a decision to cause injuries. The inevitable victims are not strangers at all, nor is their injury the result of the malevolence of God.
Politically, then, people and industries may resist tort liability not merely because it means that they will have to pay for something they are used to taking for free, but precisely because it suggests that they are “taking” something from someone with whom they have a relationship and therefore to whom they have at least negative obligations. To hold a business liable for the accidents associated with its product or service should not suggest that they did something wrong; after all, we make businesses pay for the energy, raw materials and labor they use without condemning them for using it. But it does, necessarily, mean that our judgment is that they caused the accident – that without their actions, it would not have happened. It means that tort views the accident not as something that just happened, due to the whims of God or nature, but as something for which a human being or human institution should take responsibility. It is a claim that, here at least, the tort-feasor is, indeed, his brother’s keeper, not a stranger entitled to turn away from another’s misfortune.
For more on this topic, see Do Corporations Have a Fiduciary Duty to Commit Torts and Taking Corporate Torts Seriously