It is useful as a preliminary matter to describe the approaches that scientists and lawyers take to situations involving various kinds of risk—approaches that often are in conflict but that are sometimes in harmony.
1.1A Line of Clash: Falsifiability and Propositions That Cannot Be Falsified
A frequent line of clash between lawyers and scientists has to do with a method of problem solving that is common to scientists. This method embraces a winnowing process in which investigators define a hypothesis and collect data in an effort to test whether that hypothesis can be “falsified.” This process is continuous. Testing proceeds in a way that narrows the inquiry in the direction of what might provisionally be called truth.
An important distinction concerns the criterion of falsifiability, the possibility of empirically testing a hypothesis. One might investigate, for example, the question of whether contact of a particular chemical with the skin of human beings causes a specific kind of irritation or illness. This might be termed a “scientific” question. We can contrast the question of whether a supermarket is “negligent” if it leaves a spilled, sticky liquid for twelve minutes in an aisle where a customer slips on it and breaks a bone. Answering that question requires a judgment which some might call a policy judgment. But the answer is not something one can deduce from data. It might be argued that one could verify the determination that a market was negligent when there was a collection of case results in which several courts unanimously concluded, on facts identical to those summarized above, that a market had or had not behaved below the appropriate standard of care in failing to clean up the spill. But in practice this set of identical circumstances will not occur frequently, if at all. And still the determination of negligence would be a judgment that is not a scientific conclusion. So scientists construct and prove propositions for their factual testability, but lawyers do not do that—at least not with propositions of law.
At the same time, we should note that both lawyers and scientists deal with problems of proof and that sometimes their professional processes overlap. Lawyers often use an adversary process to “prove” propositions, and sometimes those propositions are scientific in nature, in which case lawyers will employ scientists as “experts” for proof. In a roughly parallel way, it is not at all unusual for scientists to engage in what is basically an adversary process to argue about the appropriate “factual” or “scientific” conclusions from data. Those arguments may prove to be as contentious, and even as bitter, as arguments between lawyers in litigation.
1.2Some Harmonizing Elements
We further observe that there are some harmonizing elements among the professional standards and approaches of lawyers and scientists. Both professions place a premium on analysis. Their professional education develops an ability to break a problem into its component parts. It instills a demand for relevance: for focusing on the data that count with reference to the problem at hand.
Both professions involve ongoing processes of discovery. This is elementally a part of scientific education. The method of falsification assumes that usually there are more truths to be discovered, so knowledge is always tentative.1
Lawyers, by comparison, may not consciously think about this kind of process as often as scientists, but it is in their bones also. The process of judge-made law is an ongoing one. The common law tradition, discussed in further detail in Chapter 3, involves an ongoing process of development that includes synthesis of precedents, sometimes culminating in landmark decisions. One of the most famous examples is the 1916 case of MacPherson v. Buick Motor Co.,2 in which a judicial master, Benjamin Cardozo, strung together a series of nineteenth-century decisions, including English decisions, to arrive at the conclusion that an injured consumer could sue the manufacturer of a negligently made product. The prior law, using the concept of “privity of contract,” had limited consumers to suits against only the entity from which they bought the product; for example, a retailer. Cardozo’s conclusion was pithy and powerful: Against the idea that contract law defined the relation between the parties, he said, “We have put the source of the obligation where it ought to be. We have put its source in the law.”3 The process of common law reasoning does not proceed in straight lines and does not necessarily lead to expansions of the law. A pair of Texas decisions is illustrative. In one case, the state supreme court approved a claim for “negligent infliction of mental anguish.”4 Just six years later, a majority of the court rethought the prior decision and declared flatly that “there is no general duty in Texas not to negligently inflict emotional distress.”5
Legislation, a very different form of law, is malleable in a very different way. One could hardly say that it reflects a constant search for truth, unless one speaks of political “truths.” Each statute—the term for laws passed by legislatures—changes the landscape of rules by which we live. Just one example is the Occupational Safety and Health Act, discussed in Chapter 4, which firmly placed the federal government in a governing business it had not previously entered, at least not so directly—“to assure so far as possible every working man and woman in the Nation safe and healthful working conditions.”6 This statute created a new governmental agency—the Occupational Safety and Health Administration—and an entire new body of law, including many regulations and a new field of professional work for lawyers.
The professions of science and law share a problem of communication. This enigma is how to explain the technical aspects of their fields to a public that is uncomprehending and frequently skeptical. In science, a recurring problem in public perception arises from the use of statistics. There are at least four problems in this category. One is lies in arguments among experts about what the numbers are. Another is the broad ranges of some numbers. A good example of both is a much-cited study by the prestigious Institute of Medicine, which estimated that the range of “preventable” fatalities arising from the process of medical care was between 44,000 and 98,000.7 Even more confusing to a layperson would have been a later estimate where the range for that number was between 210,000 and 400,000.8 Dr. Ashish Jha of the Harvard School of Public Health told a subcommittee hearing chaired by Senator Bernie Sanders that “the IOM probably got it wrong.”9
A related set of problems lies in the difficulty of distinguishing among causes of injury. Dr. Jha responded to Senator Sanders’s question about why the new statistics had not become front page news by saying that it is “very easy to confuse the fact that somebody might have died because of a fatal consequence of their disease, versus they died from a complication from a medical error.” He said that it had “taken a lot to prove to all of us that many of these deaths are not a natural consequence of the underlying disease” but rather that “[t]hat they are purely failures of the system.”10
An area of vigorous dispute arises at the very intersection of science and law. A good example is the question of whether a court can find that exposure to asbestos is a cause of colon cancer. This question divided judges in the New Jersey state courts and federal courts in New York. In the federal case, the trial court had refused to credit epidemiological studies that showed a standardized mortality ratio (SMR) of 1.14 to 1.47, saying that the minimum ratio it would accept ...