The Legal Analyst

A Toolkit for Thinking about the Law

Ward Farnsworth

• Publisher: University of Chicago Press
• Year: 2008
• ISBN:
  9780226238364
• Topic: Droit
• Subtopic: Théorie et pratique du droit

There are two kinds of knowledge law school teaches: legal rules on the one hand, and tools for thinking about legal problems on the other. Although the tools are far more interesting and useful than the rules, they tend to be neglected in favor of other aspects of the curriculum. In The Legal Analyst, Ward Farnsworth brings together in one place all of the most powerful of those tools for thinking about law.From classic ideas in game theory such as the "Prisoner's Dilemma" and the "Stag Hunt" to psychological principles such as hindsight bias and framing effects, from ideas in jurisprudence such as the slippery slope to more than two dozen other such principles, Farnsworth's guide leads readers through the fascinating world of legal thought. Each chapter introduces a single tool and shows how it can be used to solve different types of problems. The explanations are written in clear, lively language and illustrated with a wide range of examples. The Legal Analyst is an indispensable user's manual for law students, experienced practitioners seeking a one-stop guide to legal principles, or anyone else with an interest in the law.

PART I

Incentives

1

Ex Ante and Ex Post

A thief walks into a bank, puts a gun to the head of one of the customers, and announces that he will shoot unless the teller hands over all the money in the drawer. The teller does nothing. The thief shoots the customer, runs off, and never is seen again. The customer dies of his injuries, and his estate brings a lawsuit against the bank, complaining that the teller should have given the money—it was only $5,000, let us imagine—to the thief. What should the court say?

There are two quite different ways to think about this question. The first looks at the case like this: it is a dispute between the bank and the estate of one of its customers. They’ve come to court because they haven’t been able to resolve their differences informally; the judge serves as an adjuster of last resort, producing an answer the parties agree to accept because it has the force of law. One of them will walk away a winner; the question is who. We find the answer by looking back at what happened and asking whether justice requires that the bank pay damages to the plaintiff. Did the bank do anything wrong? If the teller’s refusal saved the bank some money, would it be fair for the bank to pay nothing to the party who was injured as a result (or, more precisely, to his estate)? Can the bank bear the economic burden of the customer’s death more easily than his kin? We might consult our sense of fairness. And it might be possible to draw analogies to this case from others that seem similar—perhaps cases where a customer slipped on a puddle left behind by the bank’s janitor, or where a bank was held liable because one of its tellers tipped off a thief that a customer soon would leave the bank with a great deal of money. We could try to decide this case like others that sort of resemble it.

That much is, again, a first way of looking at the case. But now here is a second one: what happened that day at the bank is unfortunate but of secondary interest at this point; it’s over, and nothing a court says can change what happened. Money can be moved from one person to another now, or punishments inflicted, but these are just rearrangements, and they are too late to be of any serious consequence. What’s appalling about the day at the bank (or any crime or accident) is the waste of it: the lost life, the demolished car, the broken window, or whatever the harm might be. When any of those things happen, the world is made poorer—irrevocably so. This is intuitive and obvious to the family of the person who was killed. They are painfully conscious that nothing the law says can bring him back. But it’s also true, if less obvious, after an accident that merely wrecks a car. The car can be repaired, and whoever was to blame can be made to pay for it, and now the car’s owner might feel that the law did bring back the car: a miracle! It’s as if the accident never happened. But it isn’t a miracle. It’s a waste. No matter how satisfied the owner feels, the world would have been better off without the accident, because it ate up money that would have been better spent on something else—anything else. Just ask the person who ended up paying for it. The point is that the law doesn’t fix bad things that have happened. It can’t: they’ve happened. All the law can do is redistribute the suffering a bit. That isn’t nothing, of course; assigning blame and making the right person pay for it might make the victim feel better, or make the rest of us feel better. But the law’s dream—anyone’s dream—would be to turn the clock back and stop the bad thing from happening in the first place. That would be much better than quarreling afterwards about who should justly suffer for it. Too bad it’s impossible.

Yet in a sense perhaps it is possible, for the court may be able to do something almost as good. It might make a rule that would cause horrible events like the bank robbery to be less likely from now on. This wouldn’t undo the killing that already happened, but it could stop some killings in the future—and that’s equally good, isn’t it? Or better than equally good: for we would be stopping lots of bad things, not just undoing one of them. (If they aren’t prevented, we’ll be just as unhappy about those killings after they occur as we are now about the one that already did; we’ll wish we could turn the clock back—perhaps to today.) So here we arrive at a different vision of what the law can do about a case. Instead of looking back and deciding who should bear the suffering, it can look ahead and decide what ruling will make the suffering less likely to occur later.

What would this train of thought look like? In the case of the bank, it means asking what incentives people will have after the case is over. Here is an interesting possibility: if the court says that the plaintiff wins, then from now on banks will have an incentive to hand over the money when thieves take hostages (to avoid paying in court again next time). And that, in turn, means that thieves will have an incentive to take hostages. True, thieves might not know the legal rule, but then they might not need to know it. Maybe they would just observe that taking hostages has become wonderfully effective: the teller always hands over the money. The trouble becomes apparent. Letting the customer’s estate win might cause more hostages to be taken in the future. Perhaps fewer of them would be shot since the thieves would get the money they demand, but it’s hard to be sure what will happen once guns have been drawn and hostages taken.

On this view the bank has to win the case brought by the customer’s estate. It’s not a question of what’s fair as we look back on the fateful day. It’s a question of making the right rule for the future. The bank has to win for the same basic reason that governments won’t give in to hijackers of airplanes when they demand money or whatever else. It’s tempting to make the payment and save the passengers’ lives. You probably would make the payment if you knew it never was going to happen again. But that’s the point: not just that it might happen again, but that if you make the payment it becomes more likely that it will happen again.

So now we have seen two distinct ways of thinking about a problem. The first can be called the ex post perspective. It involves looking back at a disaster or other event after it has occurred and deciding what to do about it or how to clean it up. The second perspective is called the ex ante point of view. It involves looking forward and asking what effects the decision about this case will have in the future—on parties who are entering similar situations and haven’t yet decided what to do, and whose choices may be influenced by the consequences the law says will follow from them. (The first perspective also might be called static, since it accepts the parties’ positions as given and fixed; the second perspective is dynamic, since it assumes their behavior may change in response to what others do, including judges.) Part of what makes judicial decisions interesting is that courts think in both of these ways when they decide cases, and the two styles of thought can point to different conclusions. Does a court’s decision settle the dispute between these parties or does it make a rule for others in the future? It almost always does both.

This chapter (and much of the rest of this book) emphasizes the ex ante perspective because it creates lots of interesting possibilities for thought and argument, and because it usually comes more slowly than thinking ex post. It’s only natural to think in an ex post fashion when a lawsuit comes up: here are two parties in a vigorous dispute, perhaps blaming each other for a disaster. Their attention is fixed on the past and what should be done about it. They don’t care what effect the decision has on others down the line; they just care who wins right now. (That’s especially true in a case involving a random accident. The bank in our example may be a little different; it may be worried both about winning the case and about the rule in the long run, because it deals with these situations repeatedly.) But the court has to care about both perspectives, for it will be declaring a winner now and making a rule that affects others later.

In fact most courts to consider the problem of the bank teller have found in favor of the bank on the ex ante grounds sketched a moment ago. The Illinois Supreme Court put the point like so: “In this particular case the result may appear to be harsh and unjust, but, for the protection of future business invitees, we cannot afford to extend to the criminal another weapon in his arsenal.”¹ In other words, the argument from the ex ante perspective won out over the ex post. The practical challenge is to learn to see that sort of argument every time it’s available: to learn to think the ex ante way when a case is full of cues tempting everyone to only look backwards. The simplest way to find the ex ante angle is to picture either side winning the case and then imagine how the parties will think a week later—what either of them might do differently now that they know how the courts will respond. Thus you can imagine the bank learning of the decision and changing its rules to make sure that its tellers hand over the money; and then you can ask how this would, in turn, change anyone else’s incentives, starting with the thief. In real life, of course, the parties whose incentives we care about the most probably aren’t the ones before the court—this bank, that thief, and so on. We’re worried about banks generally and thieves generally. But sometimes it helps to think concretely about the parties you can see and then try generalizing to the ones you can’t.

Another way to come up with ex ante arguments is to imagine how a legislature would think about the problem facing the court. Legislatures make general rules for the future; they don’t resolve individual disputes that have already occurred. So if a legislative committee or a government agency were considering a rule about how banks should handle hostage takers, it less likely would be distracted by the equities of any one case. Its role would be to make decisions that have consequences ex ante—going forward. Granted, a court isn’t in quite the same role. It is supposed to decide cases according to the authorities on point: statutes, case law, and the like; and in the case of the bank teller, that is where a real court would start its search for an answer. But it often happens that there are no authorities entirely on point. In effect a judicial decision then serves as a little piece of legislation as well as the resolution of a conflict. The balance between these two functions varies among courts, and students of these matters often argue about how the balance should be struck.² But judges at most levels of the legal system accept the importance of giving at least some thought to the ex ante effects of their decisions, as do the lawyers who make arguments in their courtrooms. When you hear that a lawyer is making a “policy” argument, that usually means it’s an argument from the ex ante perspective—that is, a claim about the effects the decision might have on someone’s incentives.

Let’s look at some other examples. I build a house. Unfortunately I was mistaken in my measurements, and the structure extends onto your neighboring property by eighteen inches. There is no cheap way to correct the problem; either the house stays where it is or much of it will have to be torn down. What to do? The ex post style of thought accepts the situation as it is and asks how it might best be resolved. Nothing the court says can change the fact that the house encroaches; all we can do now is try to keep the damage caused by the mistake to a minimum. So the natural remedy might be an order that I pay you for the strip of land I built on, perhaps with a little premium since the sale is, in effect, being forced on you. Why waste a perfectly good house by tearing much of it down? But the ex ante perspective is entirely different. On this view it matters little how our particular mess gets resolved. The important question is how the resolution of it will affect our behavior in the future—and the behavior of others like us. From this standpoint an award of damages—a forced sale—might seem a terrible solution. It deals sensibly with the problem we already have, but it doesn’t give me or anyone else an incentive to be more careful next time. Indeed, it might create the opposite incentive: if I wanted to build on a bit of your property but were unsure whether you would be willing to sell, my best plan would be to go ahead and build and then let you sue for the value of the land. Even if I have to pay a premium, I still might be better off this way than by negotiating with you.

So the usual rule in encroachment cases is that the plaintiff—the complaining neighbor—gets an injunction entitling him to insist that the house be removed. There is more to say about this case, and we will return to it in other chapters. For now the important point is to see that it’s another win for the ex ante perspective. Ex post—after the thing is done—having one neighbor pay the other might seem attractive; ex ante—thinking about the incentives for next time, before anything has happened—payment of money seems decidedly unattractive precisely because it makes the bad thing more likely to happen again.³ Notice the analogy to making the bank pay damages to the hostage’s estate, which might cause the number of hostages to increase.

And notice also that the ex ante point of view is more than just a useful tool for courts to use in deciding cases. It’s also important to remember when deciding more broadly how well a rule works. A rule requiring buildings to be torn down when they encroach might look ugly if you just study the cases where the rule actually gets used. All you would see are buildings getting torn down or neighbors demanding extortionate prices from each other to avoid that result. But the consequences of the rule don’t appear only in those cases, or even primarily in those cases. The results of the rule also include all the cases where the building never encroaches in the first place because everyone was careful to get a proper survey done—for fear that otherwise the house would have to be torn down later. So new houses that don’t get torn down are evidence of the rule’s operation, too. A rule that looks brutal and wasteful when invoked might actually be working beautifully, if invisibly, by causing the occasions for its use to be rare.

These same trade-offs arise often when courts decide whether to allow evidence into a case. A police officer sees a fellow with a butcher knife chasing someone out the front door of a building. The officer shoots and kills him. Or at least that’s what she says happened. The trouble is that no knife was found on the dead man afterwards. The officer has some sessions with a therapist to discuss the incident. Meanwhile she is sued by the estate of the man she killed. The plaintiff demands to see the notes the therapist took during the sessions with the officer. Should this be allowed? Ex post—now that the notes exist and the two sides are arguing about them—the answer might seem clear: of course the plaintiff should be allowed to have them. They might show something important. What if the officer privately admitted that the story about the knife was an invention? What if she admitted that she shot the man because she had a grudge against him? Surely the jury ought to know those things. And if the officer didn’t say anything incriminating, what’s the harm in finding out? All this might seem a little unfair to the officer if she thought she was speaking in confidence, but that interest pales in comparison to getting at the truth behind the killing, does it not?

To repeat, however, this whole line of reasoning starts by taking for granted that the notes exist. The ex ante perspective doesn’t take it for granted. It asks how everyone’s incentives will be changed next time if the plaintiff wins this time. Next time the officer won’t go to the therapist, or won’t be as likely to talk; for a good therapist will start the session with a warning that anything said there can be used in court. Seizing the notes of the therapist, one might say, is an example of a trick that only works once. The trade-off here isn’t just between getting all that good information on the one hand and discouraging candid discussions with therapists on the other. That good information will be less likely to exist next time if it can be brought into court. This doesn’t necessarily mean, of course, that the conversation ought to be kept confidential, or “privileged.” That final decision depends on the details—in particular on how important it is for people to talk to therapists and how important confidentiality is in making those conversations happen. In the case of the officer and the butcher knife, the Supreme Court answered both those questions in favor of the officer and said that the notes of therapists are off-limits.⁴

This style of reasoning comes into play often when information of any sort is created.⁵ There frequently is tension between the rules that are best for distributing information once it exists (ex post) and for causing it to be created in the first place (ex ante). The information revealed in the therapist’s office is one example; information revealed to a lawyer is another: the case for the attorney-client privilege runs along the same lines as the case just discussed. When an attorney knows something important about a case, the court might like to hear about it; but forcing the attorney to talk this time makes it less likely that a client will talk to an attorney next time. The same goes for the rule that you can’t bring statements into court that were made during negotiations to settle the case. We’d like very much to have that information (we’d like to have all information), but ex ante the effect of allowing this would be to discourage frank settlement negotiations next time, and we want them to occur.

Or leave behind the question of evidence and think of intellectual property. Once I discover a useful drug, it might seem sensible to let you copy it. Your copying the drug doesn’t prevent me or anyone else from having it; and why should the first person to combine chemicals in a certain way be able to prevent others from doing the same? Yes, that is the ex post analysis—the analysis that makes sense once the drug exists, two parties are feuding over it, and we take those things as givens. But the ex ante point is that if you can copy the drug freely, other drugs may not come into existence in the first place. People won’t have as much incentive to create them. It’s a delicate trade-off, and the patent laws are the ...